Run your whole outbound workflow from 1 prompt

Connect ChatGPT or Claude to more than 40 lemlist actions: source, enrich, write, launch. A gtm engineer who set it up said, “the whole outbound loop runs end to end against an account.”

See how lemlist's MCP works

300 SDRs. 30,000 email sequences. A 0.5% reply rate.

That was Snowflake’s SDR process before they rebuilt it.

They had the team, the budget, and the scale.

But all that scale produced a reply rate of half a percent.

Then they changed the system.

Today the same team runs prospecting at a 7.6% reply rate (a 15x lift). Their outbound system has now processed 55,000 prospects and booked 2,000+ meetings.

Every time I break down Snowflake, I hear the same objection:

Snowflake has 300 SDRs and unlimited resources. This is not applicable to me and my 3 reps.

Fair objection. But here is what it misses:

Snowflake already had 300 SDRs and a big budget when they were getting 0.5% reply rates. That is why a 3-person team can run the same logic, at a smaller scale.

So the fix was the system.

And a system does not care whether you have 1 rep, 3 reps, or 300.

A 3-person team can run the same logic at a smaller scale.

Today, we are going to reverse-engineer Snowflake’s AI outbound system and pull out what is actually transferable to your team.

Move 1: Build the system in the backend, not as a new tool

Snowflake did not buy a new app or hand reps a new tool to learn. The system runs in the backend of the tools the reps already use.

Here is what that looks like. A rep is on LinkedIn, finds 20 prospects, and drops them into an Outreach sequence. That is the whole job. The workflow fires in the background, and 2 to 3 minutes later, when they open Outreach, all 20 emails are already drafted, each with an account brief and a call script next to it.

Building it there buys back time. Before this, an SDR writing one good email was jumping across 6 tabs: Outreach, Salesforce, Tableau, SQL, ZoomInfo, LinkedIn. Every one of those clicks is a minute not spent with a buyer.

Put the system in the backend and the clicks disappear, so the rep’s hours go to revenue-generating activity (for an SDR, prospecting like cold calls) instead of tab-switching.

The principle: when you improve your outbound stack, ask one question first. New tool = frontend or backend for your rep?

• A frontend means training, adoption, and another tab for your reps.

• A backend means the improvement shows up inside the workflow they already live in.

Move 2: Orchestration with one prompt per job

Most reps give AI one prompt: write me a cold email for this company. That is why the output is generic. One prompt cannot do the work a good rep does before they write a word.

Writing a relevant email is a process, not one step. A good SDR researches the account (first-party and third-party data), finds something relevant, picks the angle for the persona, writes, then checks it.

Snowflake broke that process down, gave each step its own prompt, and chained them.

Here is the actual workflow, the SDR process broken into steps. Everything after the rep drops a prospect into Outreach runs on its own:

1. Pull the research. First-party data from Salesforce and the data warehouse, third-party data from ZoomInfo, and a live web search for recent, relevant news.

2. Validate it. A prompt checks the research is about the right company and drops anything off-target, so the email is not built on the wrong facts.

3. Classify the persona. A prompt tags the prospect’s seniority, from IC to C-suite. This decides the angle, the length, and the depth of the email.

4. Match a customer story. Search your case studies for the one or two most relevant to this prospect.

5. Write the email. One prompt drafts it from the validated research, for that persona.

6. Score it. Another prompt grades the draft against a rubric and sends it back to be rewritten if it falls short. That guardrail is the next move.

7. Deliver it. The output is wrapped in XML tags so the system parses each piece cleanly and loads it into Outreach.

8. Build the extras. The same research becomes an account brief and a call script for the rep.

Snowflake uses Workato, but you can use orchestration tools like Clay, Cargo, Zapier, or n8n. Even one rep can chain three steps by hand: run a research prompt, feed its output into a writing prompt, then into a review prompt.

The tooling to automate the chain is what scale buys you. The thinking behind it, one prompt per job, works at any size and costs nothing. That is the step most people skip.

Move 3: They built a scoring gate before anything sends

This is the answer to the thing everyone worries about: you cannot just trust AI to write and hit send. Snowflake learned that the hard way.

Their early misfires:

• “We saw your SEC investigation.”

• “Congrats on the CEO transition,” sent to a company whose CEO had just been fired.

• “Hope the layoffs are going smoothly.”

So they built The Refinery. The AI scores its own draft against a 10-point rubric. Only a draft that scores 8 or higher goes out. Anything below that gets rewritten and scored again, and the system cannot skip the step.

The transferable version is simple: add a scoring or review step before every send. It can be a second prompt grading the first, or it can be you or your reps reading the draft against three questions before it goes out. Either way, nothing ships unreviewed.

I got the same issue with AI building these systems for my own customers: confident, well-written, and wrong about the one fact that mattered, or even hallucinating. In the early stage of this system, you need someone to control the quality.

Move 4: What they tell the AI to write

These are the performance numbers Snowflake built into the prompts.

• Lead with the prospect’s business problem, not Snowflake features. Business-first emails get 40% higher engagement than feature-first.

• Keep it short. Emails under 100 words reply at 5.4%, against 2.1% for longer ones.

• Ask a question. It lifts reply rates by 50%.

• Frame it as loss, not gain. Loss-aversion converts 2 to 1 over benefit framing.

• Ask for 15 minutes, not 30. The shorter ask gets 68% higher acceptance.

So where do you start?

Not at full automation. Snowflake did not either.

Build it with the reps, never for them.

Stand up a small pilot team (like a tiger team): a handful of reps who run the new workflow next to their normal one and flag what is off. Snowflake started with 10 SDRs on a 50/50 A/B test, half the prospects on the AI sequence, half on the normal one, so the reps saw the two side by side. That feedback loop is how a team comes to trust the output instead of working around it. Build it on top of what they already do, not as a thing done to them.

Who owns this system?

If you are early, one rep or a fractional GTM engineer owns it, builds it, tunes it. If you are bigger, centralize it. Snowflake made exactly that shift, from decentralized to centralized: instead of every rep running their own research and picking their own sequence, one team owns the whole workflow. Jeff Long in Ops owns the AI for all 300 SDRs. One owner means one place to fix a prompt, and the fix reaches the whole team at once instead of living in one rep’s private setup.

This is not a quick win

Snowflake broke their version 147 times before it held. Mine took 50+ passes with one customer. This is not a prompt you paste once and walk away from. It works, and it takes real reps to get there.

Here is why I am confident this system transfers to a small team.

The teams I have helped build their own version of this booked 20-100 meetings each from it. All in under three months.

• One closed $1.5M+ in revenue via AI-written email

• Another closed $30K+ in revenue in less than 5 weeks via AI-written email

• Another influenced $6M+ in revenue in 12 months via a modern outbound system like Snowflake's.

None of them had 300 reps.

Hope this was helpful

See you in the next newsletter!

Elric

P.S.: If you need help setting this system for your team, I can help. Reply to this email.

That is also what I am building inside the new Outbound Kitchen membership:the system, playbooks, the AI workflows, and the templates to install these systems yourself.

If you want to get notified when this is ready, join the waitlist here.

5 additional resources to learn more about the snowflake outbound system:

• How we crush prospecting at scale using AI

• How Snowflake 15x’d replies with AI orchestration

• How Snowflake Scaled SDR Efficiency with AI and Automation

• My first breakdown on Snowflake: How Snowflake Scaled Outbound to $4B ARR

• We Broke AI Prospecting 147 Times (So You Don’t Have To)

Listen on YouTube, Spotify, Apple Podcasts, and Snipd.

Florin Tatulea, GTM Engineer in Residence at ZoomInfo, Loopio’s first sales hire, who built the outbound engine at Common Room.

We covered the headcount math that over-hires outbound teams, how to size your real serviceable market before you hire, what to look for in your first SDRs, and what AI actually changes about the team you build. One number sets it up: of the 30,000 accounts that looked addressable, only about 6,000 were real.

We discuss

• The top-down headcount math that quietly over-hires outbound teams

• TAM vs SAM, and why the real number was 6,000 accounts, not 30,000

• Why you don’t need an SDR before $1M ARR or product-market fit

• Hiring two SDRs at once to separate the person from the process

• The four things Florin looks for in a first SDR, and how he vets curiosity

• Building the team like a sports team, with complementary skill sets

• What AI and GTM engineering actually change about the team you build

• Why Anthropic, OpenAI, and Clay are still hiring SDRs

Referenced:

• Prospecting from the Trenches (Florin’s newsletter)

• Florin Tatulea on LinkedIn

• ZoomInfo

• Hyperbound (AI roleplay)

Here are my biggest takeaways from the conversation

Read more

How ElevenLabs took outbound from 5% to 30% of pipeline

ElevenLabs moved its EMEA team off too many disconnected tools onto lemlist. One campaign hit 71% interaction and 25% replies. Jon (Their SDR lead): 'it has been built by people who really understand outbound.'

See how ElevenLabs did it

A few months ago I read a survey saying that:

Most SDR teams using AI are hitting about the same quota as the teams that aren’t. Using AI barely shows up in the results.

So a lot of people have landed on a simple takeaway:

AI doesn’t work for outbound.

But look at what most teams actually do with it. They bought the tool and stopped there. They open ChatGPT, type “write me an email,” get something decent once, then it falls apart on the next account. Or they hand it to the whole team with no system, so every rep uses it their own way and nothing moves.

The teams getting results point AI at one of 2 jobs:

• Make the conversation with the prospect better: more relevant, better timed, better prepped.

• Give the rep more time to actually have those conversations, by clearing the admin work around them.

Better conversations, AND more of them. That is the test. If a use of AI doesn’t do one of those two, drop it.

So I unpacked: 12 use cases, from 16 companies, 26 different examples, including the plays I run for my own clients.

Use the ones that fit your business.

1. Signal interpretation / account-state engine

Job to be done: turn raw signals into “what should we do now?”

This is the timing layer: which accounts are worth a rep’s attention right now. The signal itself is usually not AI. Job changes, funding, hiring, website visits, PLG activity, compliance-page changes: those are just inputs. AI becomes useful when it decides whether the signal matters, what it means, which account or persona or play it maps to, and whether a human should act.

What’s inside

Raw signal → account fit check → why-now interpretation → persona mapping → play selection → route to AE / SDR / CS / ignore → store event history in CRM

ClickUp

AI monitors product usage on ClickUp’s PLG base. Usage spikes and paywall hits trigger a human touch, so reps stop watching dashboards.

• Pocus reads the signals

• A custom AI-SDR runs on Retool

• Human XDRs take the calls

• Result: human BDR meeting-booked rate 2.4%, inbound AI-SDR 28%

Tools: Pocus, Retool, Nooks, Outreach, Salesforce, ZoomInfo, Clari, Sales Nav, Gong.
Source: Kyle Coleman, LinkedIn

Pigment

AI turns unstructured inputs into clean GTM objects.

• A job description becomes the tech stack, the competitors in use, and the seniority level

• A LinkedIn profile becomes competitor usage and implementation history

• Web content becomes buying-intent signals

• Pigment weights each signal against its own value prop and stacks them into momentum models inside Palette, its own data warehouse built on BigQuery

• Without this layer, real-time activation would not scale

Source: Growth Twins playbook

Writesonic

By Collin Chlarson, Head of Sales Development.

A weekly system that lives in Claude and hands every SDR their hottest accounts. Two parts.

• First, it scores every account on nine criteria proven to lead to outbound meetings, and surfaces lookalikes of the accounts that hit

• Second, it prioritizes each rep’s owned accounts and assigns the extra hot ones round-robin

• Each rep gets a spreadsheet with their top accounts, suggested contacts, and a suggested pitch (10 to 20 new accounts max)

• Plus a tab showing how much action they took on last week’s accounts

• The score also writes back to HubSpot

Tools: Claude, HubSpot, spreadsheet.

Source: Collin Chlarson, shared this with me in the SDR Leaders of USA community.

2. Internal data synthesis / proprietary context

Job to be done: use private company data to create outbound angles competitors cannot copy.

This is the angle layer: the reason to reach out that no competitor can copy. Every vendor scrapes the same job-change and funding data. So do the 40 other vendors emailing your prospect. Your Salesforce history, product usage, closed-lost notes, and past champions are the part nobody else has. That is the moat, and it is what AI can turn into outreach.

What’s inside

CRM history + product usage + closed-lost data + previous champions + conversations + multi-threaded contacts → AI synthesizes context → AI suggests angle, proof point, gift, or next action

Sendoso

AI pulls from Snowflake product usage and Salesforce history to build email and gift angles only Sendoso can send.

• The history it draws on: closed-lost opps, previous champions now at new companies, multi-threaded contacts

• Result: those data-built angles took the BDR team from under 15% of pipeline to 30%+

Tools: UserGems/Gemmy, Snowflake, Salesforce, Claude, n8n, Orum, Qualified, Clay, Salesloft, Gong, Momentum.

Source: GTM AI Podcast

3. Account research / pre-call prep

Job to be done: remove manual research from the rep’s workflow.

This is the prep layer: everything the rep needs in hand before the call. Not “AI writes better emails.” AI prepares the rep before the call: account context, likely pain, why-now, proof points, talk track. The freed hours go to selling.

What’s inside

Company research + ICP fit + recent changes + likely pain + relevant proof + account-specific talk track → pre-call brief → rep calls with context

Owner.com

Built its own AI pre-call research because restaurant context is proprietary.

• A central data team owns list-building so BDRs make more calls

• The thesis from CRO Kyle Norton: buy infrastructure, own your intelligence

• Result: outbound revenue per BDR went from $72k to $120k a month (about $1.44M per rep a year)

Tools: internal AI PCR, Salesforce, Snowflake, Anthropic/OpenAI.

Source: How Owner.com’s AI SDR system hit $1.44M in outbound per rep

Glean

By Joey Lopez, Sales Development Manager, Strategic Accounts:

A Research and Contact Audit agent runs the full prospecting analysis on an account.

• The initiatives they have that Glean solves for, with sources and quotes

• The competitor’s version, a current customer to reference, and the key people leading those initiatives

• How Glean has interacted with the company before

• It audits which of those key people are missing from the CRM and writes up why they are worth a conversation

• It drafts example emails at 80% done for the SDR to finish, creates the missing contacts, and adds them to the right sequence

Tools: Glean, CRM.

Source: Joey Lopez (Glean) shared this with me in the SDR Leaders of USA community.

ClickUp (Andy Cabasso, Growth Ops)

ClickUp’s reps run a parallel dialer (Nooks) at up to 150 calls a day. At that volume, when someone picks up you have zero seconds to find context. So Growth Ops built a workflow that loads it for them.

• The moment a prospect enters a sequence, a Retool workflow pulls their CRM data, recent activity, and product usage

• It fills the gaps with enrichment, then fires three AI models in parallel: Perplexity for recent company news, GPT to summarize the context, Gemini to write the talk track

• All of it writes back to a CRM field that syncs to the dialer, so the rep sees the background and a talk track the second the call connects

• Result: reps went from about 35 to 150 calls a day, and a top rep booked 200 meetings in a quarter, without losing quality

Tools: Retool, Nooks, Salesforce, Perplexity, OpenAI GPT, Gemini.

Source: ClickUp built this before scaling dials

4. Content becomes the rep’s script

Job to be done: turn a marketing asset into a rep-ready opener at the moment of intent.

The asset already tells you what a lead cares about. AI turns it into the opener, the questions, and the CTA.

What’s inside

Prospect downloads gated asset → AI reads the asset → generates opener + questions + CTA → writes to a Salesforce field → BDR sees the relevant script

Rippling

A custom GPT: upload a gated asset and it writes an asset-specific opener, questions, and CTA straight into a Salesforce field.

Tools: custom GPT, Salesforce.

Sources: 95% Content podcast.

5. Personalized resources

Job to be done: hand a prospect something built for them specifically, so they see the value before they reply.

Instead of telling a prospect you can help, you show them. A page, a demo, or an analysis made for their account lands harder than any claim, and it is a prospecting touch in a medium they cannot ignore.

What’s inside

What you know about the account → AI builds a tailored resource (a landing page, a demo, or an analysis) → send it over email or LinkedIn, or use it on a call

Vercel (V0)

Reps built custom landing pages for prospects with V0, Vercel’s own AI app generator. The rep feeds in what they know about the account and V0 spins up the page.

Tools: V0.
Source: Ken Amar, Outbound Squad

Rippling (Ken Amar, Sr. Dir. Sales Development)

The same play, through a vendor.

• The SDR inputs everything they know about the account and the ideal prospect

• The tool builds a custom landing page with a tailored demo: here is what we know about your company, here are your areas of opportunity

• It goes out over email and LinkedIn, an in-depth prospecting touch in a different medium

Tools: AI landing-page / demo generator.
Source: Ken Amar, Outbound Squad

Outbound Kitchen (the Chef’s Bite)

I do a version of this for my own prospects. Instead of telling them I can help, I send a personalized sample of the work, what I call the Chef’s Bite.

• It might be a TAM analysis for one of their markets showing where the opportunity is

• A list of in-market accounts giving off pain signals online, with the reasons they fit

• A map of the business units inside a target account

• An agent built in Claude Code or Clay learns their ICP and top customers, then builds the sample

• They taste exactly what their reps would get, which beats any claim I could make

Tools: Claude Code, Clay.

6. Inbound qualification that frees outbound capacity

Job to be done: qualify inbound automatically so you can put more reps on outbound.

This one helps outbound indirectly. It is not an outbound play itself, but when AI handles the inbound, you free up people, and those people go do outbound. More humans on outbound is the win. (Reallocation, not cost-cutting.)

What’s inside

Demo request / website visitor / free signup → AI qualifies → AI asks questions → AI books or routes → human works higher-value outbound

Vercel

A form-to-enrich-to-qualify-to-Slack agent drafts the reply, a human reviews and sends.

• Result: 10 inbound SDRs collapsed to 1 reviewer

• 32x ROI ($60K cost against $2M+ saved)

• The 9 freed reps moved to outbound

Tools: internal AI lead agent, OpenAI Deep Research, Slack.
Source: drew.tech

ElevenLabs

An inbound AI SDR on its own Agents Platform.

• 78% of qualification decisions need no human

• It handles about 50 calls a week (two FTE SDRs)

• Across 38 countries, 24/7

Tools: ElevenLabs Agents Platform.
Source: ElevenLabs blog

Personio

Nia (by Qualified) books demos around the clock.

• Result: 140 meetings booked in the first seven days (Lacor)

• Roughly 80% of MQLs now route through it (Shantanu)

Tools: Qualified (Nia), Salesforce, Snowflake, Amazon Bedrock.
Source: SaaStr write-up

7. AI role-play

Job to be done: make reps practice against AI until they are ready, before a single live call.

The common advice is to rush new reps onto the phone, day two if you can. Top teams do the opposite: they make every rep complete around 100 simulations before their first live call, to build phone confidence through repetition and cut ramp. Do not skip it. Practice happens off live deals, not on them.

What’s inside

Script + common objections → AI role-play simulation → rep practices and is scored → pass to go live

JumpCloud

Structured AI role-play (Hyperbound) in onboarding: 100 role plays before the first call made.

• The four highest-practice reps averaged 143 role-play calls and reached first meeting in about 11 working days

• Slow-ramp reps previously took up to 70

Tools: Hyperbound.
Sources: Hyperbound case study, Outbound Kitchen: 5 AI cold-call training scenarios

Owner.com (Rob Yuen, Head of Enablement)

Every BDR runs 100 simulations against AI avatars (Avarra) before they pick up the phone for real. The point is confidence built through reps, which is what shortens ramp.

• The avatars include a crusty Brooklyn pizza-owner persona the team named Joe

• The sims run onboarding and certification: enablement only signs off that a rep is ready once they pass

• One cohort of 7 XDRs logged 168 hours practicing with avatars instead of live calls during ramp, and one rep alone did 37 hours in their first two weeks

• The payoff: new reps catch up to experienced restaurant-tech hires fast, and rollouts that include an avatar piece see 30 to 40% higher adoption

• How eagerly a rep engages the avatar even predicts their call volume and quota attainment

Tools: Avarra.
Source: Rob Yuen, Scaling Your GTM Playbooks Using Avatars.

8. Cold-call analysis

Job to be done: mine real call transcripts to find what works and what kills deals, then rebuild the script from it.

The best AI use case may not be writing the next email. It may be understanding why the last 212 calls failed, then fixing the script.

What’s inside

Calls → AI finds the patterns (objections, pattern interrupts, lost reasons) → rebuild the cold calling script → feed it back to the team

Mangomint

Analyzed 212 cold-call transcripts from Nooks in 3.5 minutes.

• Nooks (their BDR dialer and sequencer) does not support bulk transcript export, so Claude’s browser extension scraped them all

• Claude Code analyzed the patterns: top objections, the pattern interrupts that work, battlecard gaps, and regrettable vs non-regrettable losses

Tools: Nooks, Claude browser extension, Claude Code.
Source: GTM AI Academy interview

9. Build your own internal GTM app

Job to be done: one internal surface, so reps stop switching between dozens of systems to figure out who to reach and what to say.

The most committed teams are not buying an AI SDR. They are building the surface their reps live in.

What’s inside

Tools + data warehouse → one internal surface → build audience / enrich / sequence / read signals in plain English → push to CRM and sequencer

Ramp

“Ramp Revenue,” an internal CDP processing millions of records a day plus a unified action layer with agents embedded in the workflow.

• Self-reported: more than 80% of sales workflows now run on it

Tools: internal CDP, embedded agents, Salesforce.
Source: Parth Gujare, X

Pigment

Two internal tools on one warehouse.

• Palette is the GTM brain, a BigQuery warehouse that models momentum and buying signals in real time, kept reliable by AI that continuously normalizes company names, standardizes roles, and de-dupes contacts (clean data is what makes the automation safe)

• Newton sits on top: an AI account-planning engine that pulls CRM, Gong transcripts, past wins and losses, playbooks, and live market signals into one place

• Instead of hunting across tools, a rep asks Newton what changed in an account, why it matters now, which narrative to use, and what similar deals looked like

• 80+ weekly active users (about 70% of the target AEs and BDRs), 12,000+ rep queries since June 2025

Tools: BigQuery (Palette), Newton, Gong, CRM.
Source: Growth Twins playbook

Vibe

An internal GTM co-pilot built by two GTM engineers in two months.

• Build an audience in plain English, enrich in-app, push sequences to Smartlead, HeyReach, or Outreach

• A layer called Sonar connects the app to Claude Code

Tools: Claude Code (Sonar), Smartlead, HeyReach, Outreach, Salesforce.
Source: Reza Mokdad (Vibe), from the upcoming Outbound Kitchen podcast episode.

10. Yes, AI can write cold emails

Job to be done: let AI write the cold email or message, and have it land.

Half of LinkedIn will tell you AI cannot write a cold email. The teams getting results have the numbers to say otherwise. Here is what those takes miss: the email works because it is the last step of a workflow. The timing, the angle, and the prep from the layers above all feed it, so the AI is not staring at a blank “ChatGPT, write me an email.” And a human stays in the loop on quality. Wesley Baker at Nextiva calls the payoff “outbound that feels like inbound”: by the time a rep follows up, a personalized AI-written email has already landed.

What’s inside

Account data + signals + research → AI drafts the message → human reviews, then spot-checks at scale → send over email or LinkedIn

Snowflake (Jeff Long)

300-SDR org. When a rep adds a prospect to the Outreach cadence, a multi-step pipeline does the research, the summary, the search-term generation, and drafts the email on the rep’s behalf, then scores it and flags weak drafts to fix.

• It takes the research, the drafting, and the sequence pick off the rep’s plate, and the rep stays in the loop inside Outreach

• Result: reply rate 0.5% to 7.6% (over 15 times higher)

• 55,000 prospects, about 2,000 meetings booked

Tools: Outreach, internal scoring pipeline, Claude.
Source: Jeff Long, “How We Crush Prospecting at Scale Using AI”

Outbound Kitchen (what I built for different clients)

I built the same kind of chain for a client, end to end, across email and LinkedIn.

• Clay and the Claude API research every account and rank the best-fit ones, the Claude API writes the messages, and Lemlist sends them

• I stay on quality myself: I read the first ten or so by hand, and once I trust the output I spot-check about every hundredth one

• No rep dialing, no manual research

• Result: $1.6M in revenue generated in 2026 so far

Tools: Clay, Lemlist, Claude API.

Nextiva

By Wesley Baker, Sr. Director, Business Development and Automation:

The goal is end-to-end automation so SDRs spend time on calls, not prospecting admin.

• Custom intent signals built on-site with their Outbound database flag high intent around Nextiva’s newest product (AI receptionist and AI voice)

• Their outbound database builds those high-intent accounts, filters to the ICP buying group, and pulls 5 contacts per account

• The accounts and contacts auto-create in CRM and route to the SDR team round-robin

• Their Sales Engagement Platform research agent runs prompts tuned per industry (the restaurant agent looks for recent growth and reviews that complain about wait times or busy phones, anything the email can use)

• A Prospect AI agent is embedded on the first email, and Wesley is experimenting with full automation: the SDR just has call tasks due, while a highly personalized first email to a high-intent ICP contact has already gone out on their behalf. That first step is automatic

• The reps then power-dial, and because they can see the first email already sent for them, they lean on it in the conversation when they know they are calling cold intent outbound

Tools: ZoomInfo, Salesforce, Outreach, Nooks.

Source: Wesley Baker, Nextiva, shared this with me in the SDR Leaders of USA community.

11. AI reporting agents, a daily pulse on the team

Job to be done: automate team reporting so leaders get a daily pulse, without anyone building a deck.

AI does not only run the prospecting. It can run the reporting layer on top of it, so the people leading the team see how it is pacing every day.

What’s inside

Team activity and pipeline data → reporting agent in Claude → daily summary → posted to a Teams channel → revenue leaders see how the XDR team is pacing

Nextiva

By Wesley Baker, Sr. Director, Business Development and Automation):

Built reporting agents that automate the team’s reporting and post it to a team channel, so revenue leaders get a daily pulse on how the XDR team is performing, allowing for further level of SDR execution inspection and highly customizable reporting that you cannot find with traditional reporting methods..

Tools: Claude (Cowork and Code), Microsoft Teams.

Source: Wesley Baker, Nextiva, shared this with me in the SDR Leaders of USA community.

12. Build the data and lists you can’t buy

Job to be done: build the data or the list no provider sells, with AI, and hand the team something ready to work.

Sometimes the data or the list you need is not in Clay or any other provider. So you build it yourself, with Claude Code.

What’s inside

Find the source a vendor won’t sell you → point Claude Code at it → scrape and classify → check ICP fit and the CRM → fill the contact gaps → hand the team a ready-to-work list

Outbound Kitchen (custom buying signals)

When a client needs a buying signal that Clay and the other providers do not have, I build a custom scraper for the specific site that does, with Claude Code and Apify.

• The unglamorous part is the workflow around it: scrape the source, dedupe, check whether the account is already in the CRM, then write the signal onto the records that matter

• The client ends up with a signal nobody else can pull, sitting in their CRM, ready for the team to act on

Tools: Claude Code, Apify, CRM.

Outbound Kitchen (tradeshow attendee lists)

The same move, applied to events. I built this five times for clients in the past three months, and it replaced a seven-step manual slog across multiple Clay tables.

• I point Claude Code at the event and it scrapes the exhibitors, sponsors, speakers, and attendees, sorts each into person or company, checks ICP fit, matches against the CRM, and finds contact info for anyone missing (Exa looks up each LinkedIn by name and company)

There are three ways in, depending on what the event exposes:

• A public site, straight from the URL

• A web-app attendee list, scraped or fed to Claude Code as raw HTML

• A mobile-app-only list, where you screenshare your phone and let Claude Code read the screenshots

And when the official list is locked down, a fourth angle:

• Claude Code plus the Serper API (not LinkedIn directly) finds the posts where people announce they are going, which works well for the big events

• One event surfaced about 200 posts, and I reverse-engineered those into the right ICP contacts

• The output is an ICP score for the event itself, plus a built list so the client’s reps can book meetings or invite people to the dinners they run

Tools: Claude Code, Exa, Serper API, Apify, Clay, CRM.

What’s next

This is a snapshot of what the best teams are building right now.

The hard part is turning any of it into something that makes money.

One ask:

If your team put AI use cases into outbound in a way that works, reply and tell me how. I am collecting the best ones for the next issue.

Hope that was helpful,
Elric

P.S. That is what I am building inside the new Outbound Kitchen membership:

the playbooks, the AI workflows, and the templates to install these systems yourself.

If you want to get notified when this is ready, join the waitlist here.

Listen on YouTube, Spotify, Apple Podcasts, and Snipd.

Will Rohleder is the SDR director at TeachTown, special education software for school districts, where he took over a four-person SDR team sitting on $200K of pipeline and, in about a month, had the same four reps building roughly $600K a week. Today he runs 14 SDRs at a 92% show rate and 88% conversion to opportunity. He hires almost only former teachers, and he pays his SDRs to move deals forward, not just book the meeting.

We discuss

• How he rebuilt a four-person SDR team and 10x’d weekly pipeline in about a month

• The territory design and SDR-to-AE alignment behind the numbers

• His show-rate playbook: the calendar-accept trick, layered reminders, and letting the prospect talk 70% of the time

• Why half of SDR comp is tied to the opportunity reaching stage 3, not just booking the meeting

• Who owns moving a deal forward, and the six-point qualification gate

• Why he almost only hires former teachers, and the one interview question he swears by

• Where he uses AI, and the inbound AI SDR that emailed a lawyer

Referenced

• Will Rohleder on LinkedIn: linkedin.com/in/william-rohleder-2aa7a9143/

• TeachTown: teachtown.com

• SDR Leaders of USA community

My biggest takeaways from this conversation

Read more

There's a deal in your pipeline that died three weeks ago.

Your main champion took a new job a month ago and never told you. By the time you got the news, the deal was already gone. Expertise watches for the signals you'd never catch on your own and implements the next steps to save it. The only tool an AE needs. Rated 4.9 on G2.

See Expertise in action

A few weeks ago, I saw this post on LinkedIn from Kade (SDR at Common Room):

In the 164 comments, I’ve spotted two patterns:

• Pattern 1: Every prospect has a different channel preference.

• Pattern 2: What to do before you prospect them.

So let’s unpack them.

Pattern 1: Every prospect has a different channel preference

Some said call me. Some said short email. Some said video on LinkedIn. Some sales leaders said cold calls only. Sort the replies any way you want and no channel wins.

A few weeks ago I saw the same question in a Slack community:

What’s the best outbound channel in 2026?

That’s the wrong question.

Because “Best channel” assumes one universal answer.

There isn’t one.

Once you know what works for your market, the job is to optimize each channel in your context.

The better question:

What channel mix makes sense for your market, your segment, your account, and this person?

Channel preference works in layers.

Layer 1 - Market and segment set the default

• Selling to restaurants (SMB), 1 contact per account → phone. Most owners are on the floor and may not have a LinkedIn profile. And with 1 contact per account, no way to multi-thread.

• Selling to Fortune 500 (Strategic / Enterprise) → dinners, trade shows, gifts, sponsor intros, cold calling, email, LinkedIn, top-down approach, bottom-up approach.

• Selling in European markets where cold calling laws are strict → email and LinkedIn lead.

Your channels adapt to your market, segment (SMB, Mid-Market, Entreprise, Strategic) and Cost of Acquisition.

Layer 2 - The individual sets the preference inside the segment.

Until you talk to a specific prospect, you don’t know their preference. Two CROs in the same industry will answer differently. One picks up unknown numbers. The other only engages with what their team forwards them.

The market narrows the channel mix. The individual picks the lane.

It’s also not an either/or question. It should be a mix, if your Cost of Acquisition allows it and your market allows it. Each channel has a ceiling. You can’t get 100% positive reply via email or LinkedIn. Same for calls. Same for the number of emails or phone numbers you can actually find in your market. You can’t get contact info for every prospect.

And channel sits downstream of relevance.

Even if you knew every prospect’s preferred channel, it wouldn’t save you. The leaders in the comments didn’t say “I prefer email and I take any meeting that arrives by email.” They said the opposite. Channel works when relevance is there. Without relevance, no channel works.

So “what’s the best channel” assumes a single answer that doesn’t exist at either layer. The market narrows it. The individual decides it.

If your team is debating email vs. LinkedIn vs. phone before they’ve fixed who they’re going after and what they’re going to say, they’re optimizing the wrong stage of the system.

The channel is the way the work is done. The part that decides whether the delivery lands happens upstream.

Which raises the real question the best comments were actually answering: how do you get relevant? Pattern 2 answers it.

Pattern 2: What to do before you prospect them

Here are the 3 strongest comments. That’s where the interesting insights were.

Two more quotes from posts I captured a few months ago that line up with the same pattern:

The pattern across all 5:

Show me you understand my world before you ask for my time.

That’s the work upstream of the channel.

Move 1: Get inside through the team

Before you contact the decision maker, talk to their team. The individual contributors. The front line managers. Ask them:

• What’s the current process?

• What tools do they use?

• What are the current challenges?

• What have you done to fix that?

Get into the account through the team. Multi-thread so the exec hears about you before you ever contact them. Better: get someone they trust to hand you over. That’s what most of these leaders described in their own words: get introduced, multi-thread, get handed over by someone they trust.

That’s why an account-based approach makes outbound work.

Move 2: Make your POV the reason they vouch

You don’t earn the intro by asking for it. You earn it with a point of view sharp enough that someone inside will repeat it for you.

Show up with a hypothesis tied to a top-3 priority. A reframe. An insight from the front lines. Something that proves you understand their world before you ask for their time. That’s the email Orlob described, the one that makes him lean back in his chair. Your champion repeats it up the chain.

Outbound teams in 2026 have no excuse to show up without a hypothesis and research. AI is the best tool to do this at scale: using public, online information (strategic initiatives, the facts they share online) to build a real point of view about how your solution might help their business.

In 2026, outbound teams have no excuse to send generic outreach.

This is how outbound creates demand

Most outbound teams capture demand. That’s the main reason people say outbound is dead, that you can’t generate revenue with outbound, that outbound only works because of timing.

When your outbound team gets inside through the team, shows evidence you did the work, and educates the buyer, outbound creates demand at scale.

You don’t intercept that demand. You create it.

That's it for today.

Hope it was useful!

Elric

P.S. Also

• Building that system is what I’m doing with clients right now. If you want one for your team, reach out, happy to talk it through.

• And I’m building something new: a membership to upgrade the paid newsletter: a way to build these systems without me in the room. The frameworks I unpack in breakdowns like this one, plus the workflows to run them inside your own company. So if you're interested to join the beta let me know.

We’re back with 2 live sessions:

• This Thursday I’m going live with Will Rohleder, SDR Director at TeachTown, whose 14-person SDR team wins 42% of the deals they source, against 34% without them → Save your seat

• The week after, Florin Tatulea joins me live (Author of Prospecting from the Trenches & GTM Engineer in Residence at ZoomInfo). We’re going to unpack How to Scale an Outbound Team in 2026 → Save your seat

Other resources that might be helpful on the same topic:

Stop running outreach across four different tools

lemlist customers describe it as solving ‘the problem of having to use multiple tools for outreach.’ Email, LinkedIn, and WhatsApp in one workflow. 4.6 on G2 across 1,653 reviews.

Start your free trial on lemlist

Your outbound stack has one job.

Most teams build for a different one: how many tools do we need, should we consolidate, what's the best data provider, etc.

All the wrong questions.

The one job is this:

Get your reps in front of the prospects most likely to convert and stay, armed with the best message to educate them. The highest-fit, highest lifetime value (LTV) accounts, reached with the sharpest message. Every part of the stack serves that. Data and scoring find and rank the right accounts. Your playbook and AI sharpen the message. Automation clears the low-value work so the human hours land where they pay back. The goal is rep time on the right accounts, with the right message, at the right time. The number of tools is only ever an output of that.

That sets up the second decision, the one teams get wrong: how far you can build toward that goal is capped by what you can maintain. Your stack is a resources-to-maintain-complexity decision, and that gate decides every call. Build versus buy. Consolidate versus expand. Whether your next tools make reps faster or just busier.

The most expensive mistakes in outbound tooling is stack envy: copying what other teams run or copying a more advanced team runs and inheriting complexity you cannot keep alive.

So the first step is knowing what you can run. Tools come after. An 8-rep team and a 300-SDR org can sit at the same stage, and that gap is the whole point.

We're back with 2 new live sessions

This Thursday I’m going live with Will Rohleder, SDR Director at TeachTown, whose 14-person SDR team wins 42% of the deals they source, against 34% without them.

Save your seat

The week after, Florin Tatulea joins me live (Author of Prospecting from the Trenches & GTM Engineer in Residence at ZoomInfo). We're going to unpack How to Scale an Outbound Team in 2026.

Save your seat

Step 1: Find your stage

Two questions tell you where you are.

1. Do you have a RevOps or GTM engineer, in-house or fractional WITH bandwidth to support your team?

2. Do you have a dedicated data team building your data stack?

Your answers place you in one of three stages:

• Stage 1, the Single Kitchen: no ops, no engineer. You and your reps run everything.

• Stage 2, the Modular Kitchen: a RevOps or GTM engineer with bandwidth to build for outbound.

• Stage 3, the Signature Kitchen: a dedicated data team building your data stack, your own predictive scoring, the models that decide which accounts and people to prioritize.

These stages stack. Stage 2 is Stage 1 plus someone who can build. Stage 3 is Stage 2 plus a data team on top. You move up by adding the next layer, never by skipping one.

The gate is resources, and resources means bandwidth. A RevOps title with no hours behind it does not move you up. Most teams I know sit between Stage 1 and Stage 2: they have a RevOps person, but that person has no time for outbound. I worked with a company that had one ops person for 50 reps. That person owned the full funnel, from pipeline creation at the top to closing at the bottom to expansion and renewal after the sale. With all of that on their plate, no bandwidth was left for the SDR team’s top of funnel. A request that should take a day took two weeks, sometimes a month. The resource existed, but the time did not, so the team ran like Stage 1. That is the case for a fractional hire or an agency: someone, internal or not, whose only job is the part you actually need. The spread across the best teams I studied proves the gate is resources, not size.

I spent eight months breaking down how the best outbound teams build, from three-rep startups to three-hundred-SDR machines. Look at the two ends of that range. Snowflake runs roughly 300 SDRs. Behind those reps sits a dedicated SDR ops and enablement team plus internal data teams, and every tool earns its place in revenue. Now look at Gorgias. Eight SDRs total, running 5,000+ unique AI emails a day at $0.002 each, off a machine a dedicated team built before the first SDR was hired. By the test that matters, Gorgias sits at the same stage as Snowflake’s 300. If the gate were size, those two could not share a row. They share it because both have a team maintaining the machine.

One more finding, and it matters most if you are early. None of the eight elite teams I studied is Stage 1. Every one had already put someone on the system, a RevOps lead or a GTM engineer, before any of this worked. Stage 1 is a starting point you move through. You leave it by getting someone who can own the system, not by buying another tool.

That confirms the test works. Now build for your own stage.

Here is the full map: what you decide at each stage, and where each team lands. Read it as the center of gravity at each stage, not a checklist. You will not have every row, and that is fine. You will know your row the moment you read the quote in it.

Weigh that resource’s cost against what it unlocks before you add it. Headcount and revenue do not move you up; the person who can run the system does.

Treat your data the way a serious kitchen treats its ingredients, because the dish is only as good as what goes in it. Most teams source data like a diner running on supermarket and frozen food: one cheap provider, nobody checks the quality, and it shows up in every email, call, and forecast downstream. That is Stage 1, and it is the costliest habit on this page, because bad data poisons everything after it. Stage 2 goes farm-to-table: a specific provider for each ingredient, stacked for coverage and quality, sourced for your market. Stage 3 grows its own garden: it builds and scores its own data, so it controls the quality of the ingredient and improves it over time. Run your stack like a Michelin kitchen, where the sourcing is the first thing you obsess over.

Step 2: Forget vendor categories. Every tool does one of six functions

The categories vendors sort themselves into are the wrong mental model. G2 has dozens of them. Your stack has six functions:

1. Data. Company data, contact data, intent. This is the most market-dependent function in the stack, so build it around your industry. You stack several sources, each serving a different goal: Snowflake runs four intent providers at once, and tools like Clay chain providers to fill the gaps. In many verticals the data your buyers leave online sits outside the big databases, so the right data stack for the trades looks nothing like the one for enterprise IT.

2. Orchestration. What builds and runs your workflows. The glue.

3. Execution. The touches themselves. The real design decision lives inside this one: who runs each touch, a human or an automation, by tier and channel. Gorgias automates most of its email. Cold calls stay rep-owned. Same category, two different answers, decided on purpose.

4. System of record. Your CRM, or your data warehouse once you can maintain one.

5. Context. Your playbook. The only function with no logo and no line item, and the one that decides whether your AI tools work at all.

6. AI. The models themselves (OpenAI, Anthropic, and the rest). They run inside the other five, sharpening data, orchestration, execution, and context. AI is the layer you build on across the whole stack.

One axis runs under all of them: frontend is what reps touch, and it stays short. Backend is where mature spend goes. The reason to keep the frontend short is time: every minute a rep is not fighting a tool is a minute on the revenue activity, prospecting for SDRs, selling for AEs. You design the stack to protect that time and let the backend carry the load.

Budget by function, not by vendor category. The rest of this guide gets simpler: each stage is just a different answer to who owns each function.

Step 3: The play for your stage

What you do depends entirely on your stage. Here is the play for each. Run the one that matches where you are now. The plays above your stage will break you: they add complexity you cannot keep alive.

Stage 1: no ops or engineer

Buy one all-in-one platform that sits on top of your CRM and covers contact data plus sequencing, so you are not stitching vendors together. Lemlist, Amplemarket, or Apollo.io on top of your CRM is the honest default here. Pick speed to launch over flexibility you cannot use yet. You can stand this up in days, not weeks, and start learning what works before you spend on anything specialized.

What you skip on purpose: separate data providers, any standalone AI platform, anything that needs an admin to keep it running. One good data source is enough at this stage, and you have no validated data for an AI tool to act on anyway. Data first, AI second. If a line item needs someone to maintain it, it belongs to the stage above you, not to you.

Stage 2: a RevOps or GTM engineer with bandwidth

The trigger hire is what moves you here, and it changes what your stack can be. For the first time, someone can own and maintain complexity that reps never touch. So this is where you stop only buying and start building where it pays.

Go best-in-class, surgically integrated. Keep your CRM and your sales engagement platform (SEP) as the two surface tools reps touch daily, then bolt on the back-end tools that remove real work from a specific workflow. I run lemlist at this stage, and some teams keep it into Stage 3. Outreach or Salesloft are the heavier alternatives. Unit economics decide every add, not logo count. A $150-a-seat enrichment tool that removes 40 percent of non-fit dials stays, no matter how it looks on a tool-count slide.

This is also where AI workflows live, not Stage 3. Your one technical hire plus tools like n8n, Cursor, and the Claude or ChatGPT API can support a back end that used to need three people. When a vendor cannot deliver, you build. I am building an n8n account-research automation for a customer right now because the off-the-shelf output was not good enough for their context. That is the build trigger in practice: a real priority, a vendor limitation, and one technical resource who can own it.

What you skip on purpose: the data warehouse. You will be tempted. Do not, yet. A warehouse has to be maintained, and that takes a data team you do not have. Spend that build energy on AI workflows instead, where one hire can keep it alive.

Stage 3: a dedicated data team

The data team changes what you know, on top of what you can already build. Selective building already started at Stage 2. What is new at Stage 3 is that your decisions run on data: a predictive model scores every account and tells you where reps should spend their hours, so you know what is working, what is not, and what to do next instead of guessing. Three moves open up that were wrong everywhere below.

First, stand up the warehouse-on-CRM backbone. The warehouse stores, enriches, and scores on the back end. That split is only safe because you finally have the people to maintain it and run the models. This backbone is the engine that makes you data-driven.

Second, run a specialized tool for each job. A 24-tool stack is not bloat here, because every layer ties to revenue and you have the team to cut what does not convert. And reps do not touch all 24. They live in two or three tools, the surface, while the rest run on the back end where the team maintains them. Six of the eight elite teams I studied run Stage 3, and no two run it the same way. One leans on signal breadth, another automates first, Rippling scores and stack-ranks its accounts with an in-house data-science model. Same gate, six answers.

Third, build your own edge: build the differentiator and buy the commodity. Building carried over from Stage 2, and the data team lets you build the thing no vendor will, your own data and scoring. Owner’s CRO Kyle Norton weighs five questions before building: how critical is uptime, how much customization it needs, the engineering ROI, whether it is core proprietary intelligence, and whether it gives a real competitive advantage. His rule of thumb: buy infrastructure, build intelligence. That is why Owner buys its dialer and its CRM but builds its own scoring. The most advanced teams go further and build the rep surface itself. Pigment, Ramp, and Vibe each built their own app on top of the stack, so a rep works in one tab, the next action and every account in one place, instead of stitching six vendor screens together. Vibe walked me through theirs on the podcast, shaped around how their reps actually sell. It is the most mature build there is, far more than a workflow, and it is not for everyone. It only pays if you have the engineers to maintain it and you know the edge it buys. The CRM, the warehouse, and the data underneath stay bought. You build the part that is yours.

What you skip on purpose: the urge to consolidate for its own sake. The instinct that kept you lean at Stage 1 turns into a liability here. If a tool ties to revenue and you have the resources to maintain it, it stays. The question flips from “is this too many tools” to “can we still prove each one pays.”

Step 4: The one move you can make at any stage

There is one Stage 3 behavior a Stage 1 team can copy today, at no cost. Document your playbook, what Kevin Dorsey calls “what great looks like.” Start with one doc, somewhere your tools can pull from: your three best-performing sequences, the reason each works, and the objections that sink the others. This is the fifth function, context, the one with no logo and no line item.

Do it because it is the context your AI runs on. An AI workflow is only as good as what you feed it, and an undocumented playbook makes every AI tool you add later blind. The model draws on generic web text and you get generic output. You cannot stand up a warehouse at Stage 1, but you can write your playbook down today. Skip it and you automate a process you cannot even describe.

The decision sits above the tools, and it never changes. Know what each layer of your process needs. Then buy the best tool you can afford for that layer, at your stage. That discipline scales from 3 reps to 300. The tool count changes at every step, but the thinking stays the same. The right number of tools is the number you can maintain, pointed at the one job: your reps in front of the highest-fit accounts, with the best message, as much of their time as you can buy.

Hope you find this helpful!

See you in the next newsletter.

Elric

P.S. Two things.

• Building systems like this is what I do with clients right now. If you want help finding where your funnel leaks and fixing one stage with depth, reply and tell me what your motion looks like.

• And I’m building something new: a membership to upgrade the paid newsletter: a way to build these systems without me in the room. The frameworks I unpack in breakdowns like this one, plus the workflows to run them inside your own company.

Other resources that might be helpful on the same topic:

Your reps act on every signal the second it lands

Default unifies signals from every tool you use, enriches each account with intent, and delivers it to your reps wherever they work: Claude Cowork, Slack, or any custom agent with MCP access. Teams like Owner.com, Perplexity, and Cortex already use it.

See Default in action

In 2026, Owner.com generates $120,000 in new ARR per BDR, per month.

So what’s the one thing they added to get there? There isn’t one.

The easy answer is they bought an AI SDR and the number went up. Plenty of teams bought the same AI SDRs in the same window, and their per-rep numbers didn’t move like this. If the tool were the cause, everyone who bought it would see the jump. They don’t.

What moved Owner’s number was who owned the AI, and the call on what to build versus what to buy.

Last year I broke down part 1: how Owner scaled outbound to $30M+ ARR. Back then each BDR was sourcing about $72,000 in new closed-won ARR every month.

Today it’s $120,000 on average. That’s $1.44M per BDR, per year (source: GTM Atlas). Their top performer hit $174,000 in closed-won in a single month.

Owner is redefining what an outbound team can produce in 2026. That’s why they’re worth studying. The lesson lives in how they think.

Today let’s unpack what changed.

P.S. If you missed part 1, read it here.

Who Owner is, in numbers

Restaurant software, SMB sales motion, ACV around $8K to $10K, in-office XDR team.

The scale tells you the rebuild is real: $50M ARR in September 2025 to about $100M ARR in May 2026. Headcount 258 to 357 (+38% past 12 months). Sales team 64 to 93 reps (+45% past 12 months). Their average AE closes $2M+ in ARR per year on a $150K OTE, and 4x CW ARR / AE / Month vs competitors.

That growth traces back to two structural decisions:

1. One team owns AI

2. Buy the commodity, build the differentiator

Decision 1: one team owns AI

Most companies do AI the decentralized way. Every rep and every manager goes off to figure out their own prompts, their own tools, their own workflows. It feels empowering. It produces a hundred mediocre half-built things, and it eats rep time. A rep using ChatGPT to research is a rep not on the phone. That’s why you see surveys where companies adopt AI everywhere and report zero impact.

Owner went the other way. One central team owns AI for the whole GTM function. In September 2025, they hired a GTM AI lead: YG Hong. A trained expert in this category gets closer to 20x the output of an amateur, not the 50% or 100% you’d assume. So the owner bought the expertise instead of asking each of their reps to become an amateur prompt engineer.

The ops and AI lead sit with the reps to find the bottleneck, build, measure it, and feed it back. Observe, build, measure, repeat.

Decision 2: buy the commodity, build the differentiator

The second layer. Owner buys infrastructure: Salesforce, Snowflake, Anthropic + OpenAI, Momentum, Salesloft, Datalane, Avarra, 1Mind.

They build the intelligence layer itself:

• Pre-Call Research (PCR) agent,

• ML lead scoring

• Intelligent routing model,

• Cerebro (Salesforce config in plain language),

• The Gardener (master business-context file)

• Churn review agent.

Their build test, and they only build when most of these are yes: it’s not uptime-critical, it’s custom to their ICP, the engineering ROI is big, they own the resulting intelligence, and no vendor will ever build it for their niche.

These two decisions are the whole story. Everything below only compounds because one team owns the loop.

Now let’s unpack what that loop produced at each stage of the funnel.

Stage 1: rebuild the BDR’s day

At most companies, only 30% of an SDR’s time is spent on prospecting, and 50-60% of their hours go to list-building and research.

That is the non-prospecting activities and time Owner went after first.

The central team shadowed outbound BDRs work and built one tool to absorb all of it. They call it AI PCR, pre-call research. It drops a personalized brief into Salesloft before every call: a website audit the prospect can see (a score out of 100, the revenue the problems are costing them, a problem list), plus the nearest high-performing Owner customer for local social proof.

The rep dials more, and each dial lands warmer, because the opener is built on something real about that specific restaurant. More connects, and better ones. Kyle Norton (Owner's CRO) frames it as a 5x improvement in decision-maker connects from where they started, now 20 to 30 a day on 150 to 200 dials.

The build cost: one engineer, two weeks. One team built it well, instead of fifteen reps building it badly fifteen times. Kyle calls AI PCR a large part of how the BDR team gets to $120,000+ in closed-won per month.

That lift only exists because one team owned the loop and pointed it at the boring 60%. That’s the first lift: more meetings into the funnel.

Stage 2: rebuild the AE’s day

AEs were spending to 30 minutes before every demo manually researching each prospect. That did two things: it cut the time they could spend with customers, and it slowed qualification, especially when a restaurant owner wanted a demo within hours.

The same pre-call research that warms the BDR’s dial also preps the AE’s demo. The Grader report audits the restaurant’s online presence before the call and hands the rep a personalized business case. Instead of “you need a better website,” the rep can show the specific problems likely costing this owner money. The buyer walks in already problem-aware.

The result: aes stopped burning 20 to 30 minutes per demo, freed that time for more customer conversations, were better prepared, and saw a close-rate lift on top.

If the AEs close more of the meetings the BDRs source, every sourced opportunity is suddenly worth more. That’s the second lift, stacking on the first: more meetings in, and each one worth more. The win-rate work below is the same loop, one stage downstream.

Stage 3: the talent system

Owner ran AI scoring across 150 of their sales calls to find what actually correlated with winning. The surprise: discovery barely moved the needle. Ten of the twelve highest-correlated skills were end-of-call mechanics. Discovery scored zero of twelve.

Here are the 12, grouped by where they happen on the call:

End of call (8 of 12):

• Ask for the sale

• Book the next call, even a placeholder

• Help the prospect see the value of continuing

• Agree on a real purpose for the next call

• Gauge whether the prospect is leaning in

• Ask for the next meeting to move the buying journey forward

• Give next steps with clear reasons and value to the prospect

• Make an assumptive close (fill out the form together)

Pricing (1 of 12): concise, transparent delivery that reiterates value.
Prospect engagement (1 of 12): the prospect leaves with a clear idea of next steps.
General (2 of 12): rep stays in control of the call, reads the room and simplifies language.
Discovery (0 of 12).

One caveat before you copy that. Owner sells to SMBs on a 1 to 6 day sales cycle, 1 or 2 calls to close. In that motion the deal is won or lost in the last few minutes of the call, which is why end-of-call mechanics topped their list. So don’t copy their answer. Copy their system: score your own calls and coach whatever separates your won deals from your lost ones.

Here’s what they do, in four steps.

1. Hire for DNA, then audit it. They hire raw, coachable people, often straight out of school, for drive, coachability, organization, and curiosity. Then every quarter they run a hiring autopsy: line up interview scores against actual attainment and re-fit the scorecard.

2. Manufacture the craft instead of buying it. An oversized enablement team installs skill faster than you could hire it, using a four-step loop borrowed from MMA coaching: introduce, isolate, integrate, inspect. Most enablement skips isolate and integrate and jumps straight from “here’s the skill” to “now go do it.” That gap is why training doesn’t stick.

3. Let the data tell you which craft to coach. The 150-skill analysis pointed at end-of-call mechanics, not discovery. So they ran three sequential modules on closing only, ten weeks, with calls scored daily, changing one thing at a time. Win rate moved from 38% to 52%.

4. Wire it into one signal system. The Skills Intelligence Center, an Owner build, marries the human read of a call with the machine read, and feeds both coaching and hiring off the same signal.

Talent and enablement compound together. Owner hires coachable people, then runs them through a real coaching machine. Raw talent without the machine stays raw. The machine without coachable people has nothing to work with. Put both on the BDRs and they book more, ramp faster, and stay longer. Put both on the AEs and they close more of what the BDRs source. Two stages lifted at once. That’s the compounding.

That explains another part of the $120K closed-won per BDR per month story. Owner got more precise about who could actually operate inside the machine, then made each of those people better.

Why it compounds instead of adds

The jump from $72,000 to $120,000 came in small lifts at each stage of the funnel, and the lifts multiply.

Hire the right people. Coach them on the skills that actually correlate with winning. Hand the research to AI so reps (XDRs + AEs) get selling time back, which means more conversations. The conversations are better prepared, so better conversion. More of those meetings close, because the AEs got better at closing. Reps ramp faster and stay longer.

Move four stages of the same funnel 10 to 15% each, and the gains multiply. $72,000 to $120,000 in closed-won per BDR per month. $1.44M a year, per rep.

With a team of 15 BDRs. At $72,000 each, that team sourced $1.08M in new closed-won ARR a month. At $120,000 each, it’s $1.8M. Same 15 reps, $720,000 more every month, about $8.6M more a year. That’s the lift from compounding alone, before they hire a single new rep.

And they are hiring. Their sales team grew from 64 to 93 reps (past 12 months). So the two effects stack: each rep produces more, and there are more reps producing. Every new BDR now ramps into a $1.44M-a-year seat instead of an $864K one. The per-rep gain is the multiplier. Headcount is another multiplier on top.

This is the sub-two-hour marathon of outbound.

In April 2026, Sabastian Sawe won the London Marathon in 1:59:30, the first official sub-two-hour marathon in history, breaking a barrier the sport had called impossible for decades. Overnight it reset what every coach and runner believes a body can do.

Owner just did that for an SDR team. $1.44M sourced per BDR a year is the new proof of where the outbound ceiling sits in 2026.

What you can actually take from this

Even if you don’t have 100 reps. Or you don’t have an in-house GTM engineer building you AI pre-call research in two weeks.

Good news: the lesson doesn’t need any of that. The lesson is the order of operations.

First, give the structural decision its due. Pick one owner for AI and systems on your team, even if that owner is you for now. The compounding only starts when one person owns the loop instead of every rep building their own.

Then find the bottleneck. The manual work on your BDRs and AEs, the contacts added to a sequence by hand, the research before every call.

Then pick two compounding bets, not ten. Take the busywork off your reps so they sell more, and fix one stage of the funnel with depth, the way Owner fixed closing. Two stages moved on purpose beats ten tools bought in a panic.

This is the same equation I keep coming back to: strategy, leadership, talent, systems, enablement, and outbound process. Multiplied, not added.

So when someone asks what one thing Owner added, the honest answer is still: there isn’t one. They put AI under a central team, pointed it at the boring 60% of the job, coached the humans on the one skill that moved the number, and let it compound.

You can build a smaller version of that system this quarter.

Hope you enjoy this was helpful!

See you in the next newsletter.

Elric

P.S. Two things.

• Building systems like this is what I do with clients right now. If you want help finding where your funnel leaks and fixing one stage with depth, reply and tell me what your motion looks like.

• And I’m building something new (the Pantry) to upgrade the paid newsletter: a way to build these systems without me in the room. The frameworks I unpack in breakdowns like this one, plus the workflows to run them inside your own company.

Resources:

• Kyle Norton's newsletter & Podcast The Revenue Leadership Podcast,

  • Don’t Buy Your GTM Brain

  • Thoughts on a centralized vs decentralized AI transformation

• Part 1: How Owner.com Scaled Outbound to $40M+ ARR

• How Owner.com Scaled From Founder-Led Sales to $30M ARR

• GTM Atlas: Start with the data

• SaaStr: How Owner.com’s CRO Is Closing $2M+ in ARR Per Rep With AI

• SaaStr: Inside the AI Agent Stacks at SaaStr, Owner.com & Klaviyo | SaaStr AI Annual Final Day

• Scaling Your GTM Playbooks Using Avatars with Rob Yuen & Matt Huang

I ran the same 1,400 US contacts through all 10 B2B mobile data providers. Then I checked every number twice. Once for whether it is a real mobile. Once for whether it belongs to your prospect. I paid $2,521, because no vendor will hand you these numbers.

I ran it because of how teams actually pick a provider. They go on opinion. The one a rep used last job, the one a founder saw at a conference, the answer to “what’s your favorite data provider?” in a Slack thread. Almost no one tests. So I did.

No single tool is enough. The real question is not which provider is best, but which cover the right people for who you call.

What I measured, the same way for all 10: coverage, valid-mobile rate, right-person match, predicted reachability, and cost per right-person number.

The short version

No provider is both wide and accurate, and 1 in 4 of the mobiles you get back is the wrong person. So you do not pick one, you stack two: lead with Wiza for reach, add the one provider that fits who you call, and validate the line and the name before you dial (about $0.10 a number). The rest of this page shows you which second provider fits your persona.

The 4 numbers that matter

• The widest and the most accurate are never the same provider. The widest returns a valid mobile for 1,211 of 1,400, or 87%; the most accurate is the right person 69% of the time and reaches far fewer people. The top-right corner of the chart is empty.

• One in four mobiles is the wrong person. Of every valid mobile, 60% are the right person, 25% belong to someone else, and 15% cannot be confirmed without a call.

• High coverage, low right-person coverage. You can find a number for 1,390 of 1,400 (99%), but a right-person mobile for only 957 (68%).

• Bad data costs a 10-rep team about $53k a year. 50 dials a day, 1 in 4 on the wrong person, at an $85k wage, before dead numbers.

That empty corner is the whole story. Finding a number is the easy part. Reaching the right person on a working mobile is the hard part, and it is the only part that books meetings.

What this benchmark is. A head-to-head test of 10 B2B mobile data providers that sell phone numbers for cold calling and outbound:

• Apollo

• ZoomInfo

• Lusha

• LeadIQ

• LeadMagic

• Upcell

• Wiza

• Forager

• People Data Labs

• RocketReach

The same 1,400 US contacts, run through every provider at the same point in time. Together they handed back 3,446 distinct phone numbers, an average of 2.46 per contact. The sample is balanced across:

• 7 leader personas: security, engineering and product, finance, HR, ops, sales, and marketing.

• Seniority levels: C-level, VP, and Head-of.

• 2 company sizes: mid-market (50 to 1,000 employees) and enterprise (1,000-plus).

• 100 contacts in each persona-by-size cell.

I put every number through the same two checks. First the line type: using Trestle, a service that labels each number a real mobile, a landline, or a VOIP line. Then the name on it, using IPQS, a fraud-and-identity tool that returns the name actually tied to a phone number, which I matched against the prospect. That is what makes these results measured rather than claimed. People Data Labs is shortened to PDL throughout.

Why me, and why you can trust the numbers. I help GTM leaders build and scale outbound teams, I do not sell a data product, and no provider here paid me. I paid for the test myself, about $2,521, because the vendors will not hand you these facts. They do not publish coverage, and the one thing they do claim, that the number is a mobile, is wrong a meaningful share of the time.

Who this is for. Outbound to US B2B contacts you can find on LinkedIn. If you sell into a market where your buyers are not on LinkedIn, like local businesses or restaurants, these providers cover far less of your list and this benchmark will not match what you see. Vertical and SMB data is a different problem and out of scope here.

What bad data actually costs you

You already validate emails before you send them, because spraying un-verified addresses wastes sends and burns your domain. Phone data is the same. Dial an un-validated list and about 1 in 4 of your calls lands on the wrong person, on top of the dead and non-mobile lines. That is the quiet SDR tax: reps spending their day on numbers you could have dropped before they ever lifted the phone.

The math, scoped to what an SDR actually does, because they are not on the phones all day. Say two hours of cold calling, around 50 dials. One in four lands on the wrong person, so about 12 of those dials, roughly half an hour of selling time, are wasted every day, per rep, before you count the dead numbers. Across a 10-rep team that is five hours of selling gone daily. In salary, at an $85k on-target wage, that is about $5k a year per rep at two hours of calling: roughly $16k for 3 reps, $53k for 10, $106k for 20. Dial more and it climbs, dial less and it shrinks. None of it shows up as a line item, it shows up as reps who are not booking enough. More reps will not fix it. The fix is to validate the line type and the name before you dial, about $0.10 a number, so reps only ever call numbers that are real and likely the right person.

And the wasted salary is the smaller half. The bigger cost is the pipeline you never build: every dial spent on a stranger is a conversation that did not happen with a real prospect. A quarter of your team’s calling capacity going to wrong numbers is a quarter less pipeline, and that is lost revenue, not just lost payroll.

Why coverage numbers lie

Every provider advertises coverage: how often it returns a number. That is the widest, weakest layer, and it is the only one they talk about. A real number gets four questions, not one, and each one is harder than the last. The rest of this benchmark walks down them in order.

• Coverage: did a number come back at all. Almost always yes, 99% of your list.

• Valid mobile: is it actually a mobile line, not a landline or VOIP your dialer treats differently.

• Right person: does that mobile belong to your prospect, or to a stranger.

• Reachable: will the right person even pick up.

Coverage rate is how much of your list a provider matches, and it is the easy layer everyone passes. The other three are where providers separate. Measure all four, or you are measuring the easy one.

Layer 1 · Did a number come back?

Wiza returns a number for the most people. Forager the fewest.

Coverage is the easy layer, but it is not flat. Wiza returns a phone number for 94% of a list; Forager for 38%. This is before any check on whether it is a mobile or the right person, just “did a number come back.”

Layer 2 · Is it a valid mobile?

Not every number labeled “mobile” is one.

The one claim providers make is that the number is a mobile. It is the only checkable claim, and it fails often. A share of what gets labeled “mobile” is a landline or a VOIP line that no dialer will reach the way you expect. RocketReach is the worst offender: one in four of the numbers it returns is not a valid mobile.

Strip out the landlines and VOIP and you are left with the coverage that actually matters: the share of your list you get a real, dialable mobile for. RocketReach shows the fall clearest, a number for 81% of your list but a real mobile for only 845 (60%). Wiza, cleaner, holds 1,211 (87%). This is the coverage the rest of the page runs on, the hero chart at the top and the scorecard both use it.

Layer 3 · Is it the right person?

One in four mobiles belongs to a stranger.

Once you have a working mobile, the question that decides everything: does it belong to your prospect, or to a stranger? This is the hard layer, the one no vendor reports, and the one that separates the providers.

The leaderboard flips here. Wiza is the widest on coverage but sits mid-pack on accuracy at 59%, and RocketReach is the reverse, narrow coverage but the highest right-person rate. No provider tops both, which is the whole finding. Pool every provider together and the aggregate tells the same story from the other side.

I checked the name on every valid mobile in the test (via IPQS). 60% were the right person. 25% belonged to someone else, a wasted dial and a polluted CRM. 15% I could not confirm without a live call. The best provider money can buy still hands you a stranger one time in four. And the honest limit, mine included: no data check confirms the number is still your prospect’s. People change jobs and numbers. Only the call confirms it, which is exactly why the right-person rate is the number to buy on.

Layer 4 · Will they pick up?

Pickup is the same for everyone.

The last layer, once you have the right person’s mobile: will they even pick up? This is the one you cannot buy your way out of. You cannot raise the pickup rate, because it is capped by buyer behavior, not data. TitanX, across hundreds of millions of dials, finds only about 20% of any B2B market ever answers a cold call, no matter how many times you try. And whichever provider’s number you dial, the predicted reachability is the same.

You can see it in SureConnect’s prediction. It forecasts a reachability tier for each number without ever dialing it, unlike TitanX, which scores from a team making real calls. Only about 3% are a predicted live answer (P1), about a third voicemail (P2), and over half predicted unreachable (P3). Both reachable tiers, the live answer and the voicemail, sit at the same rate for every provider. There is no provider whose numbers answer more.

So the pickup rate is not yours to improve. The only thing you can change is how many right-person numbers you put in front of your reps, and that swings a lot by provider. If only about 1 in 5 of any market will ever answer, your one lever is to spend those scarce pickups on the right people instead of strangers. That is the goal of this whole benchmark. Better data does not raise the answer rate. It decides who is on the line when someone does answer. The provider that hands you more right-person numbers hands you more conversations. Full stop.

One more trap, and it is the sharpest one: reachable is not the same as right. Of the numbers SureConnect predicts will answer or hit voicemail, 1 in 6 belongs to a different person. And on the ones it is most confident will answer live, it is worse: more than 1 in 4 is a stranger, the same wrong-number rate as the list overall. A reachability score tells you a line will pick up; it cannot tell you whose line it is. That is why the right-person check is its own layer, and why a cheap “will it connect” score can never stand in for it.

No provider is both wide and accurate

That is the chart you saw at the top, and now you have walked the layers behind it. Coverage against quality (of the mobiles a provider returns, how many are the right person), every provider plotted, and the corner you would actually want, wide and accurate at once, is empty. The widest provider is only mid on quality. The ones whose mobiles are cleanest reach far fewer people.

That is the structural reason there is no single best provider, and why the stack starts at two. Everything that follows is how you pick your two for who you call.

The scorecard: all 10, ranked by how many right people they deliver

One base shows up on top every time: Wiza returns the most right-person numbers, for every persona and both company sizes. That is a count of right people delivered off the full list, not a per-number hit-rate, and it is the number that drives conversations. So the table is sorted by it.

How to read it: Right people reached is the bottom line and the sort. It is already net of every kind of waste, so it is the only column that decides. The rest show why it is high or low. A returned number fails two ways, and both are useless to a rep: Not a mobile, a landline, VOIP, or dead line, and wrong person, a real mobile that belongs to a stranger. Mobile coverage is how many real mobiles you get. Right people per valid mobile is, of those, the share that are actually your prospect; the rest are the wrong person. Multiply the two and you land on Right people reached. Forager is the clean example: only 2% not a mobile, but 36% of its real mobiles are still a stranger, so it ends at 24%. Right person = an exact name match (IPQS) on a valid mobile. Cost = per right-person number, current pricing. n = 100 per persona-by-size cell, single run, May 2026. ZoomInfo is a borrowed-account estimate ($0.60/record, excludes its $15k/yr floor). Capability for Apollo and Lusha is from provider docs; the rest are confirmed from the test returns.

Reach and quality are different vendors.

Wiza reaches the most people, but of the mobiles it returns, only 59% are the right person. RocketReach is the cleanest at 69%, though it makes you filter the most non-mobiles to get there: about 1 in 4 of the numbers it returns is not even a mobile. PDL and Forager are right behind, at 66% and 64%, and cleaner. Those three reach far fewer people than Wiza. So the buy is base plus complement: Wiza for reach, then the provider that adds the most new right people on top of it. That second provider is RocketReach overall, though Apollo and ZoomInfo are within a few contacts of it. No single tool is enough, which is exactly the point: identity is the hard layer, no provider owns it, so you cover the right people by stacking the two that fit your market.

Explore it for who you call (Interactive dashboard)

The same data, filtered to your buyers. Pick a persona and size; every view recomputes.

Open the interactive dashboard here: https://outbound.kitchen/research/mobile-data-benchmark-2026

Best for what you care about most

The full board is above. If you just need the quick pick for your one priority, here it is.

• Cheapest usable numbers: LeadMagic ($0.11 per right-person number), self-serve and direct, is the cheapest by a wide margin. Upcell ($0.27) is next, and the cheap way into it is through Clay.

• Cleanest mobiles: of the mobiles a provider returns, RocketReach’s are the most likely to be the right person (69%), with PDL and Forager just behind (66% and 64%). PDL and Forager get there with few non-mobiles; RocketReach makes you filter more first. When a wrong dial is expensive, these hit the prospect most often per mobile dialed.

• Most coverage: Wiza reaches the most of your list (1,211 valid mobiles, 87%, the most right people delivered), which is why it is the base of almost every stack.

• A full record, not just a number: PDL and RocketReach return verified email, company, and job history alongside the mobile, so the credit does double duty.

And whatever you stack, run it as a waterfall ordered by cost: lead with the cheap, accurate ones and layer the wide one to mop up. Same reach, about 10% less.

Build your stack: find your row

No single second provider wins everywhere, so here is the whole decision in one table. Find who you sell to, take the two names, start there. That is the stack.

Lead with Wiza, then add the one partner for your row. Two providers get you to between 50% (HR, the hardest persona to reach) and 68% (marketing) of the right people. The third barely moves the line, so only add it when your dialer says you are running dry.

One cut I ran and left out on purpose: company size. I split the same data by mid-market (50 to 1,000) versus enterprise (1,000+), and the provider picture barely moves between them. The same names sit in the same corners. What actually changes the answer is the persona, who you call, not how big their company is. That is why this table is by persona and not by company size.

What a third provider actually adds

Stack the providers in order and each one adds less new reach than the last: Wiza gets you to 51% of the right people, a second provider takes you to 60%, the third adds three points, the fourth two, the fifth barely registers.

That does not mean cap your stack at two. If you run credit-based tools in a waterfall, you only pay for the numbers you actually use, so adding a third or fourth as a cheap fallback costs you almost nothing. It just changes why you add one: never for more coverage, since a third mostly hands you numbers you already had, but because that one is cheap and catches the few your first two missed. The decision is cost, not count. The only thing to avoid is paying for a second expensive contract to chase those last few points.

The reason is overlap. Some providers are largely reselling the same underlying numbers, so a second or third often returns mobiles you already have. Most pairs overlap 20 to 30 percent, and the most redundant, like Upcell and Lusha or LeadMagic and Upcell, overlap about half. Stack two that overlap heavily and the second barely widens your reach, which is exactly what the curve above shows.

Run it as a waterfall, and order it by cost

A data waterfall sequences your providers in priority order: you call provider A, and only on a miss call provider B, so you pay for the second provider only on the gaps the first one left. That is how you get two providers’ reach without paying twice for the overlap. Clay, n8n, and Cargo all build one.

The question is what order to run them in. There are a few ways to sequence a waterfall: by cost (cheapest provider first), by coverage (the provider that returns the most right people first), or greedily by best cost per new contact. I modeled all three, and cost-first wins. The cost and coverage orders both reach the same 68% ceiling, but cost-first gets there for about 10% less, because the widest provider’s coverage overlaps the cheap ones, so leading with it bills you for numbers a cheaper provider would have caught anyway.

So: quality decides which providers, cost decides their order. Lead with the cheap, accurate ones (LeadMagic, Apollo), then layer Wiza to catch what they missed. This order is for this list.

What a usable number really costs, and what a credit buys

Sticker price per number is the wrong metric. You pay for every number, but only the right-person ones are usable, so the real cost is cost per right-person number.

But even that is not apples to apples, because a credit does not buy the same thing everywhere.

LeadMagic at $0.11 returns only a phone number. People Data Labs and RocketReach cost more, but they also return verified email, company, and job history alongside the number. Price the number against what else comes with it.

How you actually buy each one (and the catch)

The number is only half the cost. The other half is how you are allowed to buy it. Most of these are cheap and self-serve. Two force an expensive annual contract.

This is the hidden lever. ZoomInfo is the one you cannot just swipe a card for. Upcell is reachable for a fraction of its direct minimum if you pull it through Clay’s credits instead of signing direct. ZoomInfo has no self-serve at all, which is why I had to borrow an account to include it. Everything else you can start on this month for the price of a lunch.

What it cost me, and the Clay tax

I paid about $2,521 to run this. The same run today would cost roughly $4,990, nearly double, for one reason: in 2026 Clay raised its credit price 2.65x, from $0.016 to $0.0425 a credit. The five I ran through Clay got 11% to 166% more expensive overnight. The four I bought direct sit outside Clay, so that hike does not touch them. Even the scoring step, SureConnect, jumped from $880 to $2,337 from the same increase.

The takeaway: where a vendor sells direct, buying outside Clay stops your cost from depending on Clay’s rate, that is Apollo, Lusha, LeadIQ, and LeadMagic. The exception worth knowing is Upcell: it re-priced its credits to match Clay’s new rate, so the cheap way in is through Clay first (about $0.15 a number), and you only graduate to its roughly $2k direct deal once it has earned a place on your list. Among the Clay-routed five, only RocketReach also softened the hike; Wiza, PDL, and Forager pass Clay’s full increase straight through.

What I actually took away from this

The honest headline: no single tool is enough. There are winners by category. One provider leads on coverage, another on accuracy, another on cost. None leads on all three, and the gaps between them are smaller than the marketing suggests. Every provider has a strength and a soft spot, so the win is picking the two that fit who you call, not hunting for the one that does everything.

The bigger lever is what you do with the data after you buy it. No vendor can tell you whether a number is the right person on your specific list, that is the gap this whole benchmark is about, so your own dial results are the only ground truth there is. A team that validates numbers before dialing, never calls a dead one twice, and feeds those results back into which providers it keeps will beat a team sitting on “better” data with no loop, every time. The provider choice gets you in range. The feedback loop is the actual edge.

I built it with a team that books most of its meetings on the phone. We stacked a couple of mobile providers, validated each line was a real mobile before anyone dialed, scored reachability with the tools made for it (we tried TitanX and SureConnect), and fed every call result back into which provider earned the next list. That loop is the part that compounded.

The recommendation, in one place

If you read nothing else, this is what to do.

1. How many providers depends on how you pay. For reach, two get you most of the way and a third barely adds, so you do not need five for coverage. But if you buy credit-based tools and run them in a waterfall, you only pay for the numbers you actually use, so stacking several as cheap fallbacks costs almost nothing extra. The trap is contracts: do not sign multiple annual deals like ZoomInfo or Cognism. A clean setup is one contract tool as your base if you already have one, plus credit-based providers layered in a cost-ordered waterfall. For any provider that still locks you into an annual contract, run the experiment on your own list first with the month-to-month and credit providers, and sign the year only once the coverage proves out for your market.

2. Run them as a cost-ordered waterfall. Lead with the cheap, accurate ones and layer the wide one to mop up. Same reach, about 10% less spend.

3. Buy direct where you can. Apollo, Lusha, LeadIQ, and LeadMagic sell their own credits, insulated from Clay’s hikes. Through Clay you can pay up to 2.65x more for the same number.

4. Track the source. Stamp every number in your CRM or sequencer with the provider and date, or you will never know which one is earning its spend.

5. Validate before you dial, then score. Run each number through a line-type and name check (IPQS, Trestle) to confirm it is a live mobile that belongs to the prospect, about $0.10 each. That gets you most of the way on identity; the live call is the final confirmation. Then score for reachability (SureConnect, TitanX) so reps dial the best numbers first.

6. Close the loop. Pull dispositions back from the phone system, was it the right person, someone else, did they pick up, and feed that into which providers you keep.

The stack gets you in the game. The feedback loop is how you win it. The benchmark gives you the starting lineup; your dispositions tell you who to keep.

If you are already in one of these situations

• You already pay for ZoomInfo. Keep it for what it is good at, the full record: email, firmographics, org charts. Do not lead your dialer with its mobiles. They are mid on quality (53% of their mobiles are the right person) and the most expensive per usable number ($1.27). Use it as the enrichment base, layer cheaper, higher-match-rate mobile providers on top, and put ZoomInfo last in the waterfall.

• You run on Clay. Most of the credit-based providers are one toggle away: Wiza, Upcell, RocketReach, PDL, Forager. Order your waterfall cheap-and-accurate first, then Wiza for coverage. LeadMagic and Apollo are cheaper bought direct, so pull those in with your own keys. Upcell re-priced its credits to match Clay’s new rate, so start it through Clay before its direct deal.

• You are starting from zero. Begin with Wiza plus the one partner for who you sell to (the lookup above), bought whichever way is cheaper for that vendor. Add a third only when your dialer says you are running dry.

What I did not test, and why

• Waterfall aggregators (FullEnrich, BetterContact). Left out on purpose. They stack the same underlying providers I tested, so of course their coverage looks higher. It is not like-for-like, it is someone else’s stack with a markup. The waterfall section above is how you build the same thing and keep the margin.

• One-year-contract providers (Cognism, SalesIntel, Seamless.AI). I could not get access without signing an annual contract, so I left them out rather than guess. ZoomInfo is in only because a friend ran my list through his account.

• Live calling. I did not run a calling census, and even one has a ceiling: a dial only tells you the truth about the people who actually answer, and most will not. The pickup numbers here are predicted, not confirmed by an actual call. A future round with a human caller could measure real pickup, but it would still only confirm the share who pick up, not the rest.

Behind the numbers

The methodology

Elric Legloire, Outbound Kitchen. I help GTM leaders build and scale outbound teams and write a newsletter on it. I do not sell a data product, no provider here pays me, and none sponsored or reviewed this. Self-funded, about $2,521 out of pocket.

The sample. 1,400 US B2B contacts scraped fresh from LinkedIn in May 2026. 7 leader personas (security, engineering/product, finance, HR, ops, sales, marketing) at C-level, VP, and Head-of. 2 company sizes (mid-market 50 to 1,000, enterprise 1,000+), 100 contacts in each persona-by-size cell. Input to each provider: first name, last name, LinkedIn URL, and company name for some.

The depth. Those 1,400 people came back as 3,446 distinct phone numbers, 2.46 per contact across the 10 providers. I did not score one number per person and move on. Every number got checked on its own for line type and for the name on the line, which is what makes the right-person rates here measured rather than claimed.

How I sourced each provider. Phone numbers through my own Clay credits for five (Upcell, Wiza, RocketReach, Forager, People Data Labs). Direct via their own API for three I pay for (Lusha, Apollo, LeadIQ). My own account API for LeadMagic. ZoomInfo through Jed Mahrle’s account, since it is annual-only and I could not buy it on its own. Thanks Jed. If you want the sharpest outbound newsletter going, read his: Practical Prospecting.

The checks.

• Coverage: a number came back.

• Valid mobile: Trestle classified the line type, mobile versus VOIP versus landline. That is the only thing I used Trestle for.

• Right person: IPQS is a fraud-and-identity tool that returns the name tied to a phone line. I took that name and matched it to the prospect, counting only exact matches as right-person (nicknames and truncated names that I could not confirm exactly are counted as not-confirmed, which is the strict, conservative call).

• Predicted reachable: SureConnect predicts whether each number will connect, without dialing it. It is a prediction, not an actual pickup, and I report it separately, never folded into the funnel.

Right-person definition. Right person is an exact IPQS name match on a valid mobile, IPQS only, no second validator promoting partial matches. It is the strict definition, which is why these right-person rates read lower than a vendor’s own marketing number. The published dataset uses the same definition, so if you recompute it, you land on these numbers.

Limits. Single run, one snapshot in time. US, LinkedIn-sourced contacts only. The per-persona and per-size cuts are 100 to 200 contacts each, so treat those as direction, not gospel. The numbers move over time, so treat this as a snapshot.

Run it yourself

Filter the dashboard to your personas and company sizes, and every number recomputes for your exact segment. Sort all ten providers by coverage, by right-person accuracy, or by cost per right-person number. Put any two head to head. Read the best two-tool stack for who you actually call. Then download the full 1,400-row dataset and run the whole test against your own list.

Open the 2026 US B2B Mobile Data Benchmark:

Open the Benchmark

Hope this benchmark was useful!

See you in the next newsletter.

Elric

[Live Webinar] Why account scores tell you too late

Account-level ABM waits for the score to spike. By then the buying group has decided and the wrong messages have already landed. Forrester’s Kelvin Gee joins Influ2 CEO Dmitri Lisitski to break down what orchestrating around individual buyers looks like instead. June 3, 12 PM ET.

Save my spot

In the past 2 weeks, I saw 4 GTM leaders (CRO and SDR leaders) asking about how to increase show rate. And I saw this on LinkedIn:

If you are using Calendly or Chili Piper and sending automated reminders, that's exactly why your show rate is sitting at 30 to 50 percent. Just remove the automation, run the handoff manually, and the show rate climbs.

Which I disagree with the framing that automation is the problem, I worked at Chili Piper. So I've seen our internal data and with customers.

So if you have 30-50% show rate, it’s a meeting quality problem.

The meeting was bad before it got booked.

P.S. For paid members, the full meeting-quality diagnosis scorecard with if-yes / if-no branches is at the end of the newsletter, including what to fix for your outbound team.

Show rate is a defect rate

Think of show rate the way a kitchen thinks about a plate sent back from the table.

When a dish comes back, you do not fix it by training the server to apologize faster. You walk back to the line. You find the station where the dish was made wrong. You fix the process there. The complaint at the table is just where you noticed the defect. The lever is upstream.

The meeting got booked. That part worked. But what got booked? A prospect who said yes to escape a cold call. An invite titled “Intro call” with no agenda and no reason to attend.

Reminders confirm a meeting the prospect already wants to attend. They do not create the reason to be on the call.

Show rate is the receipt at the end of a meeting-building process. Fix the process, not the receipt.

Bad meetings are created before the calendar invite is sent.

Once you accept that show rate is downstream, the question changes. The right question is: where in our process are we building meetings prospects don’t want to attend?

So if an outbound team has a 30–50% show rate, I’d first look at the sales process:

Is the account in ICP?

Show rate is the first place a bad account shows up, but it is not the last. The same account that no-shows the first meeting also tanks opportunity rate, deal size, and close rate downstream. If your SDRs are booking outside ICP to hit activity quota, you are building defects on purpose. Audit the booked accounts against your ICP definition before you audit anything else.

What is the channel used for booking the meeting?

Meetings booked via cold calling versus email or LinkedIn have a different show rate. Don’t expect the same show rate.

Did the prospect name a problem?

Curiosity is what got the prospect on the call. That is outbound’s job. But curiosity is not enough to make someone show up next week. A stated problem is. If the SDR cannot repeat one clear sentence the prospect said about their pain, the prospect was only interested. They were not committed. That meeting is weak.

How far out is the meeting booked?

The farther the meeting is, the weaker it gets. Same week is strong. 14 days out is risky. After 14 days, you are betting that the prospect will still care about the same problem. They probably won’t.

Did the prospect accept the invite live, on the call?

A live accept is the cheapest, strongest commitment device you have. If your reps are not asking for it, you are leaving show rate on the table.

Does the invite explain why you are meeting?

“Intro call” is not a reason. The prospect’s stated problem is a reason.

The data points upstream, not at the reminder

At Chili Piper, where I worked, we ran 80 to 90 percent show rate on outbound meetings, around 200 meetings per month. We had automated reminders, sure. We also had the meeting-building process I just described. The reminders were a layer. The meeting-building was the foundation.

Gong data, cited in Cold Calling Sucks, breaks it down by rep tier.

• The average rep runs a 56.9 percent show rate on cold-call-set meetings.

• The top quartile runs 72.5 percent. 1.3x the show rate.

Tito Bohrt analyzed 6,414 meetings and found the same directional pattern on calendar distance. Same-week beats next-week beats two-weeks-out. The meeting was a stronger commitment at the time of booking.

• Booking meetings 0-1 days out results in the highest show rate, peaking at 80.79% for day 1.

• The show rate stays relatively strong for the first few days, around 75%, but starts to dip by day 5 and steadily declines to around 60% by day 14.

As always, compare apples to apples. If you sell to people on the ground, like restaurants, don’t expect to get 90 percent of show rate. Your show rate will look different from selling to office buyers who spend all day in their calendars.

So comparing show rates without segmenting by industry, persona, and channels can be misleading.

Show rate shows the result. Meeting quality is the lever.

Show rate is the metric you might check every Monday. Track it. If you want it higher, fix the meeting before you fix the reminders.

Outbound creates demand. We're here to give the best experience for prospects by creating curiosity on how we can help them fix their problems and change. If prospects won’t show up to the conversation, it was a politeness response on a cold call.

The fix is on the line.

That's it for today.

Hope this was helpful.

Elric

What’s coming

US Mobile data benchmark for 2026: Finally, wrapping this up, the benchmark will be sent this week.

For the paid newsletter

I’m still building The Pantry. It’s the next version of the paid newsletter.

I’ve been spending on how to design the next phase, don’t want to be just another “hub” just for content + templates like the paid newsletter. And working on the MVP.

Here’s what the membership will give you:

• Self-assessments and trainings that work like a kitchen audit. You walk your line, find the station slowing the whole pass down, and fix it this week.

• AI workflows, prompts, and skills you can run today. Example: scraping attendee lists from tradeshows and conferences.

• The methodology, built in the open. You’re on the journey with me as I turn what I do with customers into SOPs and frameworks.

• The system runs on any CRM, any SEP, any market. With AI you depend on tools less anyway. You build your own layer on top.

• Trained on real reps. What I run with customers, what I test, what top teams do. It gets sharper every month.

New: Pricing when Pantry launches: $79/month or $499/year.

To everyone who supported the paid newsletter until now: thank you. You made this possible. That’s why your current rate stays, a small way to say thanks.

• Already a paid subscriber? Nothing changes as long as you stay subscribed. (If you cancel and come back later, the new rate applies.)

• On the fence? Subscribe before The Pantry goes live to lock in the current rate.

Subscribe now

For Paid Members

Scorecard: The meeting-quality diagnosis

The full meeting-quality diagnosis scorecard with if-yes / if-no branches. Including what to fix.

Layer 1: Scope the problem before you fix anything

Read more

Free: The ABM That Starts With People Report

Most ABM targets accounts. But accounts don’t buy, people do. This report breaks down how to build ABM around the decision-makers who actually influence the deal.

Get the free report

An SDR leader I chatted with in Q1 had a problem:

He’d hired four BDRs the same week. Five weeks in, three had booked meetings. The fourth had booked zero.

Scaling SaaS org. Low patience. ROI watched closely. He wanted to let the rep go.

If you’ve led an outbound team, you’ve been there. The number is short. One rep is dragging it. Leadership wants to know what you’re doing. Firing the rep is the obvious answer. It also tells you the bar is high.

Before you pull the trigger, look at the data.

In 2025, 78% of B2B sellers missed quota (Ebsta). Right now, 44% of SDRs and 57% of AEs are missing their number (RepVue, May 2026, role-level).

Two different lenses, same direction.

Here is what most leaders do when a rep misses quota.

They look at the number. They check how long the rep has been ramped. They compare to the rest of the team. The rep is behind. The ramp window is closed. The rest of the team is on track. Decision made. PIP, then exit.

Shaunt Voskanian, CRO at Figma, said it on 20VC:

His point: quota is a lagging indicator. It’s downstream of territory, ICP, messaging, enablement, comp design, and a dozen other things you control. Pinning the miss on the rep is the cheapest read of the data you have.

It also protects the leader more than the team. Firing the rep closes the loop. Nobody asks if the territory was carved fairly. Nobody audits the ramp. Nobody pulls the call recordings to see if the booking reps are running a different play.

HBR has a name for this: the Set-Up-to-Fail Syndrome. Manzoni and Barsoux’s research found managers default to blaming the employee even when their own behavior, attention, and resourcing are the cause.

So diagnose first. Five checks, in order. Three on the rep. Two on you.

Here’s how it runs:

The Five-Check Diagnostic

Run them in order. Don’t skip ahead. Each check rules out a cause before you spend time and money on the next one.

Check 1: Effort

Pull the activity:

• Calls dialed

• Emails sent, LinkedIn touches

• Accounts worked.

• Prep for discovery calls.

Compare to the reps who are booking.

If the volume isn’t there, you have a motivation problem. Talk to the rep. Find out what’s going on. Sometimes it’s personal. Sometimes the rep checked out two months ago and you missed it. Either way, this is the cleanest call on the list. If effort doesn’t come back in two weeks, exit.

If the volume is there but results aren’t, move to Check 2.

Check 2: Knowledge

Read five cold emails. Listen to a discovery recording.

• Does the rep know the product?

• Can they articulate the problem you solve in the buyer’s words?

• Do they know the top three objections and have a response for each?

• Can they map a use case to a persona without thinking about it?

If no, you have a knowledge gap. That’s a 2-week fix. Onboarding refresh, shadowing top reps, certification on objection handling. If knowledge is there, move to Check 3.

Check 3: Skills

Knowledge is what the rep knows. Skills are what they execute under pressure. Pitching cleanly with a script in front of them is not the same as handling a buyer who interrupts in the first 15 seconds.

Listen for: opener delivery, qualifying questions, objection handling, control of the call, next-step ask. Compare to your top performer’s calls on similar accounts.

If the rep knows what to do but can’t execute it live, you have a skill gap. That’s a 4-to-8-week fix with deliberate practice. Role-plays, call reviews, live coaching on skill reps, not deal reviews.

If the rep is executing the skills, move to Check 4. This is where most diagnoses stop. Don’t stop here.

Check 4: Coaching transfer

This is the first check on you.

Pull your coaching notes from the last six weeks. What did you tell the rep to change? Did it show up in the next call? The one after?

Training transfer research is brutal on this. Across 89 studies, what reps actually apply from coaching depends on three things: the rep’s cognitive ability and conscientiousness, the rep’s own motivation, and the support they get from you and their peers after the session. If two of those three are weak, coaching doesn’t stick.

Two failure modes here. One: you’re giving feedback but not following up, so the rep hears five contradicting notes and applies none of them. Two: you’re “coaching” but it’s deal review dressed up. Frank Cespedes’s HBR data says most managers can’t tell the difference.

If the rep is coachable and you’ve been running deal review, the fix is on your side, not theirs.

Check 5: System

This is the second check on you, and the one most leaders skip.

If the rep is putting in the effort, has the knowledge, has the skills, and is applying coaching, and the number still isn’t there, the system is broken. Walk it backwards:

• Territory. Did this rep get the same account quality as the others? Run the ICP fit score on each book. If one rep is working a B-list, that’s not a rep problem.

• Data. Is the contact data on this rep’s accounts as good as the others? Bad data quietly tanks conversion before the rep ever opens their mouth.

• ICP and messaging. Sometimes one rep was handed the messaging deck that worked six months ago and the others figured out the new one in the Slack channel.

• Quota design. Aggressive quota set in a market-push environment, no safety net. Shaunt’s frame: that’s a quota design miss, not a rep miss.

• Enablement and process. Are the booking reps using an asset, a list source, or a workflow this rep doesn’t know exists?

The system check takes a week of your time. Most leaders skip it because it implicates them. That’s why you run it.

If the system is broken, firing the rep doesn’t fix it. The next hire walks into the same trap.

The Decision

After the five checks, you’re in one of three places.

One. Effort is missing, or coaching isn’t sticking and the rep isn’t coachable. Exit. Fast.

Two. Knowledge or skill gap, rep is coachable, effort is there. Give it 30 to 60 days of focused work on the specific gap. Not generic coaching. The gap you diagnosed in Check 2 or 3. If the leading indicators move, keep going. If they don’t, exit.

Three. The rep is doing the work and the system is broken. Fix the system. Keep the rep. This is the call most leaders won’t make because it means going to their boss and saying “the miss isn’t the rep, it’s how we built the territory, the data, or the play.” That conversation is harder than firing someone, and it’s the one that compounds.

Fix the system once, the next four hires ramp faster.

Back to the SDR leader from the intro.

He ran the checks. Effort was there. Knowledge was there. Skills were there. Coaching was landing. The rep was working the same play as the others.

So he gave it more time instead of pulling the plug.

By Feb 27, the rep hit two of three meeting targets. One of those meetings turned into €30K of pipeline. By the end of the quarter, the rep had generated €120K of pipeline across four SQLs. That was more than the rest of the 11-rep BDR team combined.

If he’d fired in week five on instinct under leadership pressure, that pipeline doesn’t exist. The leader doesn’t look like he hired well. The team loses a quarter of its number. The next hire walks into the same broken-ramp setup and the cycle runs again.

One last thing.

If you only remember one line from this:

quota miss is a symptom, not a diagnosis.

Firing feels like the cheap, decisive move. It isn’t. You eat ramp time, severance, recruiter fees, lost pipeline, and the team’s read on whether you have their back. Hiring well is expensive whether the next rep works out or not.

The five checks cost a week of your time. They tell you which reps to keep, which to coach, and which part of the system to fix before the next hire walks into it.

One more loop to close. If the rep does exit, run a hiring post-mortem.

• Did they have the right soft skills for the role and the stage of the company?

• Did your interview process actually surface those skills, or did you find out three months in?

• Feed every answer back into the next scorecard, the next reference call, the next interview.

That’s how you stop running the same diagnosis on the next rep.

Run the checks before the next exit conversation. If you’ve already had it this quarter, run them on the next rep before the pattern repeats.

See you in the next newsletter.

Elric

🔒 For Paid Members: Team Outbound Planner

The 5-check diagnostic above runs on one rep. The Team Outbound Planner runs the same logic on the whole team and tells you the two levers to pull this quarter.

It’s the same Root Cause Analysis I use with clients, packaged as a Google Sheet. Fill in your numbers, see where the team is leaking, get the two highest-ROI fixes.

P.S. The Pantry (the new product I’m building) will include an AI version of this. Same diagnostic, runs on your own data, no spreadsheet. More on that soon.

Grab the Google sheet here:

Read more

Free: The ABM That Starts With People Report

Most ABM targets accounts. But accounts don’t buy, people do. This Influ2 report breaks down how to build ABM around the decision-makers who actually influence the deal.

Get the free report

Most teams research accounts without a system.

They open a list, tab through LinkedIn and the careers page, copy a few facts into a doc, then start typing the email. Twenty reps means twenty different processes, and nothing a manager can coach against.

I run the same two steps on every account, with one framework underneath. Here’s how it works.

5 mistakes I see with account research

Mistake #1: no research step at all.

Most reps go straight from a list to a sequence. Skip Miller calls this out in Outbounding (my full breakdown here): research is part of the process, not optional. You don’t skip it any more than you skip writing the email.

Mistake #2: only researching the contact.

“Find 3 things to personalize.” Most contacts aren’t active online. Even when they are, none of it connects to why their company might need your solution.

Company data is always there. News, hiring patterns, funding, tech stack, leadership changes, team structure.

Research the company first. People second.

Mistake #3: no framework, no documented process.

If you’re running 20 SDRs, each one researches differently right now. Nothing to coach against. Nothing to QA. No standard for onboarding new reps.

You can’t improve what you can’t see.

Mistake #4: doing all of this manually in 2026.

AI pulls funding rounds, leadership changes, hiring patterns, and press coverage in seconds. If your reps are still tabbing through LinkedIn, Crunchbase, and the careers page by hand, you’re paying SDR salaries for work an AI does in 1-2 min.

Mistake #5: running prompts in chat instead of building a workflow.

One-off ChatGPT prompts don’t scale. Every rep writes their own version, outputs land in random docs, nothing is reusable. A workflow runs the same prompt the same way every time, lands the output in the same place, and lets the next rep inherit the system instead of rebuilding it.

These mistakes stack up when you scale. One rep wastes their own time. Twenty reps waste pipeline. And you can’t coach against twenty-five different research methods, so quality stays uneven and ramp takes longer than it should.

The Account Research System

The Account Research System sits between two things you’ve already done: defined your ICP and selected your target accounts. It comes before “what to say.”

Two steps:

• Step 1: Research. Gather the facts.

• Step 2: Reason. Turn facts into hypotheses.

One framework underneath both: SPICED.

One output that feeds every channel: cold email, cold calls, LinkedIn, and the discovery call if they take the meeting.

Why SPICED?

I picked SPICED (by Winning by Design) because it’s buyer-centric. Situation, Pain, Impact, Critical Event, Decision. Every letter is about the prospect’s business, not the rep’s process.

Compare that to BANT: Budget, Authority, Need, Timeline. BANT is rep-centric. It’s the rep’s qualification checklist. You can’t pre-research any of it. Budget and timeline only come up once you’re already talking. BANT works for inbound calls where the prospect raised their hand. It doesn’t work for outbound, where you have to earn the meeting first.

SPICED is different. Most of it can be reasoned from what’s already public.

What’s online: funding rounds, leadership changes, hiring patterns, tech stack, GTM motion, press, customer logos. That’s enough to build the Situation and infer the Critical Event.

What’s not online: which pain is highest priority right now, what it’s actually costing them, who internally owns the decision. That’s the gap. You close it by stacking the public facts into a hypothesis, then testing the hypothesis on the call.

That’s the whole reason this works before the meeting is booked. SPICED forces you to reason about the buyer’s business, and most of the inputs are already sitting in public sources.

That public/not-public divide gives you a natural two-step split:

• Step 1 = S (Situation). Pure facts.

• Step 2 = P-I-CE-D. Reasoning on top of those facts.

Step 1 is what AI does well. Step 2 is what the framework forces AI to do that it wouldn’t do on its own.

Step 1. Research: the S in SPICED

To make it simple, let's focus on a net-new account. No prior conversations. No CRM notes. Just what you can find online.

Pure fact-gathering. The job here is data. Save the interpretation for Step 2.

• Employee count and growth trajectory

• Tech stack: CRM, sales tools, marketing automation

• Funding stage and recent rounds

• Leadership changes: new CRO, CMO, VP Sales

• Hiring patterns: 5 AE openings, 0 SDR postings

• GTM motion: PLG, sales-led, or channel

• Customer base and market positioning

• Account mapping: who’s on the team, their backgrounds, what tools they’ve used before. If there’s a CRO and 2 reps, the CRO is your go-to person. (First part of D, Decision in SPICED.)

One caveat. Not all of this data lives in the same place.

Some is easy to find with AI: funding rounds, job postings, leadership changes, press coverage. An AI model pulls that in seconds.

Other data isn’t available via web search. Headcount growth over the past 12 months. Sales team size compared to 24 months ago. For that, you need enrichment tools: Clay for headcount and team changes over time, BuiltWith for tech stack.

This is the input for every step after this, including who to contact.

Example: Replit researched through TrackRec’s lens

If you’ve been reading the past few newsletters, you’ve seen me return to TrackRec → Replit. TrackRec, a company that helps organizations hire sales talent.

Without a system, a rep finds: Replit raised $400M, they’re hiring 12 roles. They send:

“Congrats on the raise! We help companies hire sales talent. Open to a 15-min call?”

Congrats, then pitch. No connection between the two. The rep found the facts but never connected them.

Here’s what the Account Research System produced instead.

Target account: Replit. Researched through the lens of TrackRec, a company that helps organizations hire sales talent.

Perplexity output for this step:

P.S. Based on my recent benchmark, for this step I recommend Perplexity or GPT-5.2 Thinking (as of April/May 2026).

Step 2. Reason: the P-I-CE-D in SPICED

Step 1 gave you the Situation. The facts.

Step 2 is where you reason through them using the rest of SPICED: Pain, Impact, Critical Event, Decision.

A pile of facts without a framework leaves you guessing. SPICED organizes those facts into hypotheses about why this company might need you.

This is the move I call layering flavors.

Step 1 gives you the raw ingredients. Step 2 stacks them so each one changes the meaning of the next. A funding round alone is just a fact. A funding round plus 12 open roles plus a new CRO plus no recruiting leader is a different dish entirely.

Now they’re hypotheses.

Pain.

Look for facts that imply a constraint: roles that stay open, leaders who scale through methods that don’t scale at the next stage, capacity missing where growth requires it. The pain isn’t in any single fact. It’s in how the facts together describe the bottleneck.

Impact.

The framework asks: what does this constraint cost them, and on what clock? Revenue at risk per month. Hiring milestones missed against a board commitment. Time burned during a leader’s first 90 days. Impact is always a number times a time horizon.

Example for Replit:

Critical Event (identified from Situation).

Leadership changes are critical events. Funding rounds are critical events. Multiple openings going up at once is a critical event. These create the time pressure that makes “later” stop being an option for the buyer.

Example for Replit:

Decision (hypothesized from the org map).

You already mapped the account in Step 1. Now you reason through it.

Who’s the economic buyer? Who would feel the pain most? Look at a leader’s first hires. Anyone they brought in from a company that uses your product, or a competitor’s, is a champion hypothesis.

Company size gives you process clues. 50 employees means the CEO is in every deal. 500 means department heads. 2,000 means procurement.

Example for Replit:

Blockers:

Two steps. Four connected hypotheses on top of the Situation.

Outreach Angles

From this research, three angles surface: the hiring-philosophy angle (passion vs. motion at 200 hires), the recruiting-capability angle (the gap, without attacking the team they’ve built), and the geography + ramp angle (SLC SDRs and NYC enterprise AEs running simultaneously). Each one opens a different conversation. None of them lead with “congrats on the raise.”

You can use this info on the phone, in a cold email, or on LinkedIn. And for your discovery calls when you book the first meeting.

That’s the Account Research System. If you want to start using it, try with 10 accounts to control the outputs, then scale with Clay, n8n, or any other orchestration tool.

Hope this is helpful.

Talk soon,

Elric

P.S. If you need help setting this up in your outbound systems, let me know.

For Paid Members - Your Account Research System

Paid members get the complete system: 3 prompts that personalizes this entire framework to YOUR company.

You can use this inside your orchestration app (Clay, Claude Code, etc).

The Prompts:

Prompt 1: Company Context Builder. Give it your website, ICP, conversations with customers, or G2 page. It builds your company context: ICP, value props, how buyers solve this today, critical events that create urgency. This is the lens for everything else.

Prompt 2: Account Research Generator. Paste your company context. It generates a customized research prompt with your competitors, buyer titles, and signals hardcoded in. Save it. For every target account, just add the company name and run.

Prompt 3: Reasoning Generator. Paste your company context. It generates a customized hypothesis prompt with your pain points, proof points, and outreach angles baked in. Save it. For every target account, just paste the research and run.

Prompt 1: Company Context Builder

Copy everything between the lines below.

Read more

This episode was originally recorded in French for Cognism’s Prospect podcast with Laetitia Fall. The English audio you’re hearing is an AI-translated dub generated with ElevenLabs. Original French version.

Listen on YouTube, Spotify, Apple Podcasts, and Snipd.

We discuss:

• Why outbound isn’t dead, but pipeline got roughly 10x more expensive (1% conversion in 2015 to needing 1,000 emails per opportunity in 2023, source: Winning By Design)

• Why global SDR headcount grew over 20% in a year while public layoffs dominated the LinkedIn feed

• The Owner.com benchmark: 40 appointments per BDR per month, 25% close rate, $70-80K sourced revenue per BDR per month

• When to hire experienced SDRs vs juniors, and the 12-meetings-per-month threshold that signals your playbook is ready

• Why ICP work beats tool selection (Tier 1 closes at 25%, Tier 2 at 5%, you need 5x the volume to compensate)

• Snowflake’s 140 ICP data points and why 10-15 is the realistic target for most teams

• Multichannel orchestration when only 30% of your prospects are reachable on LinkedIn

• The data-quality test most teams skip: coverage rate per persona, per market

• AI in outbound (January 2026): what actually works, why most LinkedIn AI messages fail, and the foundation AI needs (market, CRM, message principles, signals)

Referenced:

• Cognism’s original French episode

• Host: Laetitia Fall

  • LinkedIn

  • YouTube

• Sources cited in episode: Winning By Design, Insight Partners (Jeremy Donovan’s portfolio data), Owner.com BDR benchmarks, Snowflake ICP methodology

My biggest takeaways from this conversation

The State of Outbound in 2026

Read more

[Live Webinar] Why GTM Orchestration Breaks Without Contact-Level Execution

Every contact in a buying committee takes a different path. Most ABM nurtures them identically. Devin Reed (ex-Gong) joins Influ2 to break down where orchestration fails and how contact-level execution fixes it. May 6, 10 AM CST.

Save your spot

This newsletter is also available in video and audio version.

You can listen on YouTube, Spotify, Apple Podcasts, and Snipd.

A few weeks ago I saw a post from the CRO of Vanta.

Someone sent her a Quake rocket launcher replica, a reference to her time as a professional Quake player. She posted about it. She took the meeting.

Then a similar post: the CMO at Incident.io got a signed vinyl of their favorite album of all time. Same reaction. Same outcome.

Both worked for the same reason. The research is the system. The gift is the output.

I went down a rabbit hole figuring out how to turn this into a repeatable system for a GTM team running ABM on tier-1 accounts.

Here’s the math. If your top accounts are worth $200k–$1M in ARR, a $300 gift is cheaper than the cold email sequence you’d run instead. You’re not adding to the budget. You’re moving acquisition spend from one channel to a better one.

Cold outreach is harder every quarter (not dead, harder). Generic gifting is the same problem in a new channel, a branded notebook is just another ignored touch. One of my SDR team already sent bottles of champagne or tequila to celebrate promotions and new jobs. Better than nothing. Not personal enough to break through on your top accounts.

The whole system scales with Clay and Claude Code (or any similar tool) once you’ve tested it by hand on 10–20 accounts. Then run it on your top 200 with your GTM team.

Prompts at the bottom.

The 2-step play

Step 1: Find the clue + score the prospect. A specific personal detail the prospect has publicly declared about themselves.

Each prospect lands in one of four buckets:

• Identity: they declared a past identity (”ex-pro Quake player,” “former touring drummer”)

• Passion: they declared a specific, narrow obsession (”favorite album of all time,” “seen Radiohead 7 times”)

• Milestone: no clue, but a recent trigger in the last 90 days (funding, promotion, book launch)

• No gift: nothing strong. Revisit in 90 days. Expect this 70% of the time.

Step 2: Turn the clue into a gift + note. Specific to the clue, not the broader category.

If Step 1 fails, Step 2 doesn’t happen. Most of your prospects won’t have a strong enough clue. That’s the filter working. And realistically, not all prospects will share specific clues that you can use

P.S. I’m not talking about personal stuff from Facebook or Instagram. I mean public business stuff.

Step 1 - The clue

Not every personal detail qualifies. There’s a specificity ladder.

Only Passion and Signature clues qualify.

The rule: focus on the statement, not the source. A Signature clue in a LinkedIn About section is as valid as one on a 2-hour podcast. What matters is specificity and that the prospect said it themselves.

Look for Signature language in their own words: “favorite of all time,” “changed my life,” “in a past life I was...,” “I grew up doing X.” Most common in podcasts, long-form posts, and keynote intros.

Where to look:

• LinkedIn (About section, posts, articles)

• Podcast appearances and interviews

• Conference talks, YouTube keynotes

• Their own blog, Substack, Medium

• Published articles, book forewords

• Company page “About” and “Founding story”

Here’s the output when I ran this on Kyle Norton (CRO of Owner.com):

Interesting thing:

When I was testing the prompts, Perplexity was the one returning the best output. Just to let you know, last time I tested, Perplexity and ChatGPT were both equally finding stuff. This time, ChatGPT failed, so just for FYI. This a good reminder to always keep testing the AI models + benchmarking them.

The decision

Run the research on your prospects of your top accounts. For each one, the output is a tier.

Step 2 - The gift + the note

Once you have the clue, the gift has to map to the specific niche, not the broader category.

What to send for each bucket:

Identity gift: an artifact from the past identity. Quake CRO got a rocket launcher replica. A former touring drummer gets a signed pressing of the album they toured behind. A former chef gets the cookbook from their old kitchen. The rule: it has to be the specific artifact, not the category. Not “a guitar pick”, the pick from the show.

Passion gift: a high-signal item inside the niche they named.

• “Huge Radiohead fan” → the limited Kid A Mnesia box.

• “Deep into pour-over coffee” → a Lance Hedrick-recommended grinder.

Milestone gift: tied to the trigger.

• New Series B → a framed print of their funding announcement.

• Promotion to CRO → a signed copy of The Qualified Sales Leader.

• Book launch → buy 10 copies, send to their team with a note.

No gift: don’t force it.

Here are some recommendations I got for Kyle:

Personalized Solid Oak Martial Arts Belt Display & Trophy Rack on Etsy:

1. Go to https://www.etsy.com/listing/667908227/martial-arts-belt-display-personalized.

2. Select ‘Oak’ finish and your preferred number of belt slots (1-slot is cleanest for a single black belt display).

3. In the ‘Personalization’ field enter: Top plaque line 1: ‘Slow is Smooth. Smooth is Fast.’ | Line 2: ‘Kyle Norton’ | Line 3: ‘Black Belt — OpenMat MMA’.

4. Add to cart. Production + shipping is approximately 7–14 days (seller ships from the US with free shipping).”

The note is the other half.

Here’s the actual handwritten note sent with the Quake replica to Vanta's CRO:

Two things make this note work:

First, the note stacks two references. “Professional gamer” is the surface clue. “Killcreek” is her old gamer handle, and the Vanity Fair quote about the rocket launcher is the flex. That’s the detail that proves the note wasn’t AI-written or outsourced.

Second, the pitch is one sentence with proof. Two recognizable customer logos (Ashby, Amplitude) + one hard number (25% convert).

Here’s the template you can use behind every note:

1. Research proof - the surface clue, in the prospect’s own language

2. The rabbit hole - a deeper detail that required real effort to find

3. Gift bridge - one line connecting research to why you’re sending this

4. Pitch + proof- one sentence on what you do, with logos or a number

5. Soft ask - specific, low-friction (”chat next week?” not “would love to connect”)

6. Name + direct email - not a calendar link

~60–90 words. Handwritten on a real card.

Here's an example for Kyle:

And that’s it!

Hope this is helpful.

Talk soon,

Elric

The 3 prompts

Three prompts:

PROMPT 1: Research & Classify

Read more

Here’s everything I published in Q1 at Outbound Kitchen, plus what’s coming in Q2.

This newsletter is also available in video and audio. Watch or listen on YouTube, Spotify, Apple Podcasts, or Snipd.

What we cooked in Q1

March 2026

• Podcast - OK28: How ElevenLabs is Scaling Outbound From 5% to 46% With Human SDRs Jonathan Chemouny (GTM at ElevenLabs) on rebuilding SDR motion and moving reply rates from 5% to 46%.

• Podcast - Best AI for Account Research? I Tested 7 AI Models
  March 2026 benchmark. Seven models, same account research test. Which one wins and where each breaks.

February 2026

• Newsletter — I tracked 232 outbound teams. Here’s what I found.
  47% of the 232 outbound teams I tracked grew their SDR headcount in 2025.

• Podcast — OK26: 5 AI Cold Call Training Scenarios Every Outbound SDR Team Should Run
  Troy Johnson (SDR Enablement at Jumpcloud) on the 5 AI roleplay scenarios.

• Newsletter — ClickUp built THIS before scaling dials
  The call prep system behind 3x pipeline and 150 dials/day.

• Newsletter — 9 ways I use Claude Code for outbound
  Call prep, tradeshow workflows, campaign building. The 9 workflows I run before touching Clay.

January 2026

• Podcast — OK25: How to Build a Profitable Outbound SDR Team
  Jay Glenn on SDR P&L math, headcount decisions, and why the payback window is now 9 months.

• Newsletter — How to prove outbound is working (to your CEO)
  The exec dashboard that answers the questions your CEO is actually asking.

• Podcast — OK24: How DoorDash Scaled Outbound from $291M to $8.6B
  Diane Ring (ex-DoorDash) on full-cycle field reps, city-level ownership, and scaling to $8.6B without SDRs.

• Newsletter — How to Divide Your Outbound Market Into Territories (7-Step Guide)
  TAM → SAM → ICP → territories. Seven steps for splitting a market cleanly.

• Podcast — OK23: How to Build SDR Enablement from Scratch
  Troy Johnson on when to hire your first enablement lead and the 3 things every outbound playbook needs.

• Newsletter — 100 emails → 1 meeting (2015). Now it’s 1,000+
  Cost per opportunity is at an all-time high. What actually moves reply rates in 2026.

• Podcast — OK22: How to Cold Call for Higher Connect Rates
  Jason Bay (CEO, Outbound Squad) on higher connect rates, better conversations, and meetings that show up.

What’s coming in Q2

The Pantry (Launch: May 20th)

I’m converting the paid newsletter into The Pantry. Building it now. I’ve been saying “soon” for months. Should have shipped it in Q1. Finally committing.

When I recorded the video yesterday, I had not finalized the launch yet. I worked on it this morning, and the date will be May 20.

How it splits:

• Free newsletter: keeps running. Free advice, frameworks, playbook walkthroughs.

• Paid (Pantry): the full system. Systems, prompts, files, Claude Code setups (agents, etc), AI workflows. Everything I build with customers or my previous full-time roles, turned into something you can run yourself.

What’s already built and landing inside:

I’ve shipped a lot of internal tools for customers recently using Claude Code. Templatized versions land in The Pantry so you can run them on your own data.

Three worth calling out:

• Tradeshow attendee scrapers: workflows for pulling attendee lists from the apps and sites where they actually live.

• Account-based outbound campaign system: the one I didn’t plan to build, and the one I’m most impressed with right now. Account list → contact discovery → outbound message, with a feedback loop baked in. Works across cold email, cold calling, and LinkedIn outreach. Built for sales teams running real multi-channel motions, not solo reps.

• Rep call-shadowing system: records call transcripts with reps, then turns them into recommendations. Flags where time is being lost, what could be done differently, and where reps can add more value. Built to sharpen team workflows.

The campaign system isn’t ready yet. I’ll share progress.

Pricing when Pantry launches: $79/month or $499/year.

To everyone who supported the paid newsletter until now: thank you. You made this possible. That’s why your current rate stays, a small way to say thanks.

• Already a paid subscriber? Nothing changes as long as you stay subscribed. (If you cancel and come back later, the new rate applies.)

• On the fence? Subscribe before The Pantry goes live to lock in the current rate.

Subscribe now

Q2 podcast lineup

• Growth Marketing director of Gorgias

• Co-founder & CRO of Reachdesk

• Growth Outbound working with a team of 20+ rep outbound team and building their own internal sales app

Systems and newsletters in the pipeline

• Account research system

• How to scale outbound teams in 2026

• How I use Clay for outbound

• Account-based gifting system

• Cold calling system

• How Snowflake uses AI for outbound

• Event system: how to maximize tradeshow output as a sales team

• How to build AI outbound systems for your team

• Trend watch: top outbound teams building their own apps for SDRs and AEs

Help I’m still looking for

Finalizing my US mobile phone number benchmark for 2026. I have datasets from Clay data providers (Upcell, Leadmagic, etc), Wiza, Apollo.io, Lusha, and LeadIQ. Still need ZoomInfo (normally should be good) and Cognism.

If you work at a lead gen agency or in RevOps and have access to either, reply to this email. Will pay for this.

Happy to be back!

Let’s get cooking,

Elric

Listen on YouTube, Spotify, Apple Podcasts, and Snipd.

Jonathan Chemouny is GTM at ElevenLabs, where he scaled outbound from 5% to 46% of pipeline in one year with a small distributed team. He previously built SDR teams at Intercom and Lattice. His team runs with no job titles, no standups, and no activity metrics, operating on async Slack and trust-based autonomy.

Listen on YouTube, Spotify, Apple Podcasts, and Snipd.

This episode was first published on outbound.kitchen.

We discuss:

• Why an AI-native company at 95% inbound decided to hire human SDRs instead of automating everything

• How ElevenLabs defines outbound -- pure cold vs. PLG-assisted -- and why the distinction changes the entire playbook

• Why finance and VC backgrounds outperform career SDRs, and the 4-round hiring process that screens for AI fluency and pressure tolerance

• The operating system: no standups, no pipeline reviews, async Slack (”Slack militaire”), and AI agents replacing management overhead

• SDRs running demos, flying to meet prospects, and attending events solo -- one rep skipped President’s Club for CCW Berlin and returned with 60+ SQOs

• The AI SDR handling multilingual inbound qualification (78% of decisions require no human intervention)

• WhatsApp as a prospecting channel in Europe and why LinkedIn is outperforming email

Referenced:

• ElevenLabs VP of Sales, Carles Reina: Inside ElevenLabs $330M ARR Sales Machine & The 20x Sales Comp Plan Reps Must Hit (20x Base-to-quota model)

• Jonathan Chemouny on LinkedIn: linkedin.com/in/chemouny/

• ElevenLabs: elevenlabs.io

My biggest takeaways from this conversation:

Read more

Subscribe now

By the way, if you’d rather watch or listen to this newsletter, you can find it on YouTube, Spotify, Snipd or Apple Podcasts.

Six months ago, I tested 6 AI models for account research.

GPT-5 Thinking took the crown.

Since then, the landscape has shifted again:

• OpenAI: ChatGPT 5.2 Thinking

• Anthropic: Claude Opus 4.6

• Google: Gemini 3 Pro

• xAI: Grok 4.2 (Beta)

Time for round 3.

What I’m testing today: 7 models:

• ChatGPT 5.2 Thinking (NEW)

• Claude Opus 4.6 (NEW)

• Gemini 3 Pro (NEW - The 2.5 model hallucinated last time)

• Grok 4.2 Beta (NEW)

• Perplexity Sonar Deep Research (returning)

• Grok 4 (returning)

• Argon by Clay (NEW - first time testing a Clay-native model)

Why I keep doing this:

Two things keep changing: new models are released, and existing models keep improving. The best tool for account research six months ago might not be the best one today. I don’t assume, I test.

1. Each model is great for specific use cases.

2. I test the model before I scale it with Clay, Cargo or n8n.

3. Account research is where model quality matters most. Your message quality and discovery call prep depend on the quality of your research.

1. Today’s Test: TrackRec + Replit

TrackRek:

Today I’m testing this for a company called TrackRec. They help companies hire sales talent faster, verifies quota attainment, ranks candidates on data, cuts time-to-hire.

If you’re a rep at TrackRec, you need to know: Is this company hiring salespeople? Who’s the decision-maker? What’s their employer brand? What tools are they using? Is there a reason to reach out right now?

Replit:

The company I’m testing on: Replit. $2.8M revenue to $150M+ ARR. $250M raise in September. 10+ open sales roles. CRO was on a podcast talking about hiring through his network.

I gave all 7 models the same prompt. 12 specific instructions. Same company, same prompt. Here’s what they found.

Disclaimer

These are my results. One company, one prompt, one test. Your results might be different. The point isn’t “use Perplexity.” The point is: here’s a framework for testing which model works best for YOUR account research.

Important

All models were tested in the web app, except Argon which was tested directly in Clay. Web app vs API vs Clay will give different results, different token limits, different output formats. Next time I will add API testing to the benchmark.ons.

2. What I Changed vs the last test

Last time: 3 criteria (web access, reasoning, additional insights).

This time: 6 weighted categories. Added Accuracy, Business Relevance, and Research Usability. Business Relevance asks: did it find what the BUYER needs? Research Usability: can a rep actually act on this?

Added an instruction completion table.

The prompt asks 12 specific things. Did you do what I asked? Yes or no. Perplexity did 10/12. Gemini did 1/12.

3. The Scores: Top 3 Deep, Bottom 4 Compressed

1. Perplexity Sonar — 9.2/10

Found the most. VP of SDR no one else found. Found both funding rounds. Found the CRO’s podcast and pulled quotes from it. One false claim: said RepVue had no profile when 5 other models found it. Best researcher, with one blind spot.

2. GPT 5.2 Thinking — 7.5/10

Zero false claims. 24 citations. Everything it said was true. But missed the January funding round, missed the podcast, missed 7 of 10 tech stack tools. Nothing wrong, but nothing unique. The safest model. Not the deepest.

3. Grok 4.2 Beta — 7.2/10

Fastest model, 56 seconds. Best RepVue depth, found sub-ratings no one else pulled. Honest about uncertainty, hedged on the $400M round instead of stating it as fact. But only found 1 tech stack tool.

Scoreboard (March 2026)

Bottom 4

• Grok 4 (6.1/10) — Found the $400M round but generic reasoning. Reliable, not insightful.

• Claude Opus 4.6 (5.6/10) — Best reasoning in the test. Worst champion identification. Missed the CRO despite checking 11 sources. Use for analysis, not finding people.

• Argon (Clay) (5.3/10) — Found the CRO and the podcast from just 4 pages. Most efficient model. But too thin for standalone research.

• Gemini 3 Pro (4.2/10) — Worst researcher. 1/12 instructions completed.

4. What I Found When I Verified Everything

After scoring, I manually verified every claim the models disagreed on. Six disputed facts. Checked LinkedIn, the articles, Replit’s website.

Zero hallucinations.

No model made stuff up. Last time, Gemini fabricated two executives. In this one, nobody invented anyone. That’s progress.

Three false claims

Claude said “no visible CRO.” Here he is. Ghazi Masood, CRO at Replit. Five other models found him.

Perplexity said “no RepVue profile.” Here it is. Overall rating 4.7, quota attainment 73%. Four models found it.

Only Perplexity found him. Nate Robbins, VP of Sales Development at Replit.

REPORTED vs. VERIFIED.

Press says the $400M round happened. Replit’s own website only shows $250M. Should your SDR reference funding the company hasn’t announced? Grok 4.2 hedged — said “follow-on talks.” That might be the most honest answer in the whole test.

Verification Summary

Zero hallucinations. 3 false claims total.

5. How I’d Actually Use These

Perplexity or GPT for the research, and if you want to go 1 layer deeper use Claude to analyse the findings.

• If you’re running 500 accounts: GPT is the safest. Zero false claims. You trade depth for safety.

• If you’re researching 10 high-value accounts: Perplexity. Spot-check the “not found” answers.

6. Where You Can Actually Use Each Model

Every model was tested in its web app (or Clay for the Argon Model). If you want to run these at scale in Clay or via API, here’s what’s available:

The scores in this test are web app scores. If you run GPT 5.2, Claude, or Gemini through Clay’s native integration, the output might be constrained by Clay’s token budget. Perplexity and Grok 4 are available directly in Clay. Grok 4.2 is beta and web-app only, you can’t use it in Clay yet. Claygent IS Clay’s model (Argon), so that score is exactly what you’d get.

In the next benchmark, I’ll test the models in the web app vs. Clay vs. API to measure the gap.

Here’s the prompt, If you want to get the prompt I used:

P.S. I’m also building a prompt builder for paid members so you can create your own account research prompt customized to your product.

Read more

• Instant access to 40+ paid newsletters, including:

  • 6 AI prompts to find and reach accounts 10x faster

  • The Guide to Building an Enterprise Prospecting System

  • [Starter Kit] Step 1: Build your ICP + Persona Creation v1

• My Outbound Chef Kit: 130+ resources (20+ email templates, cold call scripts, AI prompts, top tools, and more). Value: $1,000+

And the best part? You can try it free for 7 days. Cancel anytime.

But I think you’ll love it!

👉 Start your free trial today.

If you only read LinkedIn, you’d think outbound is dead.

AI is replacing SDRs. Cold calling is dead. Email is dead. The SDR role is dead.

If you stopped there, you’d think outbound is in trouble.

Last year, I started tracking top teams to see what they’re actually doing. I also looked at global SDR numbers by country and company. I’ve been tracking 232 companies since March 2025.

I cross-referenced my research with data from Insight Partners, The Bridge Group, SaaStr, RepVue, and PeerSignal.

Here’s what the data shows for 2025:

• Global SDR headcount increased by 12%.

• 47% of the 232 companies grew their SDR teams in 2025.

• Top GTM teams like Rippling, Ramp, Vanta, and Owner.com expanded their outbound teams.

• AI companies are investing in SDR teams (OpenAI, Anthropic, ElevenLabs, Clay).

• Traditional industries, not just tech, are investing in SDRs too: logistics, banking, paint, marketing agencies, and more.

1. The market split into three tiers.

Global SDR headcount:

• 170,000 in March 2025.

• 190,000 by December, 2025.

That’s 20,000 new SDR roles in nine months.

And the hiring isn’t slowing down.

In December 2025, there were 11,259 open SDR roles globally, across 4,944. companies. By February 2026, that number hit 14,135 postings across 6,248 companies.

By Country (top 10)

The US holds 37% of global SDRs, but the growth is happening everywhere. Mexico grew 20%. South Africa grew 33%. Ireland grew 33%. The Philippines has 5,000 SDRs and grew 11%.

I have 213 companies with full March-to-December 2025 data. Here’s what I found:

Here’s what I found:

47% of companies grew their SDR teams. 12% cut. And 41% did neither, they added a head here, lost one there, averaging +1.5 SDRs per company across the full year.

And 41% did neither, they added a head here, lost one there, averaging +1.5 SDRs per company across the full year.

Aggregate Stats

• Net SDRs added across the 213 companies: +2,772

• Growers added +3,171 headcount (avg +32 per company)

• Decliners lost -532 headcount (avg -20 per company)

• Flat band: +133 across 87 companies (effectively flat)

• Median company grew +8.9% (+8 heads)

• Mean growth was +19.8% (pulled up by explosive growers)

The market is splitting into movers vs. the stuck middle.

By Broad Category

By Funding Status

54% of VC-backed companies meaningfully grew, but 13% meaningfully declined. No other funding category has this spread.

Biggest Growth Stories (Absolute Gains, March-Dec 2025)

The movers:

Cybersecurity went all-in. ThreatLocker (+60%), KnowBe4 (+52%), CrowdStrike (+12%). Six out of eight cyber companies in my data grew SDR teams. The security buyer is hard to reach, outbound is the play.

• CrowdStrike 280 (+12%)

• Fortinet 141 (+11.9%)

• Palo Alto Networks 113 (-1.7%),

• SentinelOne 113 (+27%)

• ThreatLocker 83 (+59.6%)

• Zscaler 42

• Rapid7 80 (+23.1%)

Traditional industries

Another thing I learned from this research: Traditional industries are also investing in SDR teams, not just SaaS companies anymore.

• HDFC Bank in India has 446 SDRs.

• Sherwin-Williams, a paint company, has 220.

• POOLCORP, which sells pool supplies, has 156.

• Labatt Breweries grew its SDR team by 67%.

• Accenture: 320 SDRs

• TP (Teleperformance), BPO/Outsourcing: 241 SDRs

The outsourcing market: 92% of SDR agencies in my dataset grew. Mid-tier agencies grew 20-49%. Companies want outbound, many just don’t want to build it in-house.

SaaStr and Emergence Capital surveyed 560+ companies and found 36% of B2B software companies cut SDR headcount in 2025. SDRs had the highest reduction rate among all sales roles.

Big Tech: Flat, Not Pulling Back

They’re maintaining, which for companies this size means they’ve decided the current headcount is roughly right. Amazon (AWS) is the exception worth watching.

Why the growth? My hypothesis:

More companies entering more markets means more sellers competing for the same buyers. Inbound alone can’t sustain growth when your category has 47 competitors instead of 12. Outbound is how companies create demand when demand doesn’t come to them.

If outbound were dying, the numbers would be going the other direction.

2. AI companies are hiring SDRs

39% of top AI companies with open GTM roles are actively hiring SDRs, data from Adam Schoenfeld at PeerSignal.

OpenAI is building its first SDR organization. Anthropic is hiring an 8-12 person BDR team. Clay, LangChain, ElevenLabs, and CoreWeave are all building SDR functions from scratch.

These are the companies building the AI tools that are supposedly replacing SDRs. And they are hiring SDRs.

There are currently 2x more Enterprise AE roles than SDR roles posted at AI companies, which means SDR hiring is coming, because those AEs need pipeline.

If AI were going to replace the SDR role, the AI companies would know first. Instead, they are investing in it.

3. What the winners built (and what the cutters got wrong).

Insight Partners surveyed 150 companies: top performers were 1.6x more likely to expand SDR teams. Only 2.9% of the best GTM teams eliminated their SDR function -- versus 9.8% of the broader market.

What I found in my dataset is similar to the data from Insight Partners.

The best GTM teams are investing more in SDRs.

• Rippling: 300 to 360+ SDRs (+20%)

• Ramp: 100 to 140+ (+40%)

• Vanta: 64 to 82 (+28%)

• Gong: 61 to 84 (+37%)

• Owner.com: 20 to 30 (+50%)

Since 2024, I study the best outbound teams:

Snowflake. Ramp. ClickUp. Owner.com. Rippling.

They don’t have a silver bullet. They invest heavily across the board:

• Systems: data infrastructure. AI and automation.

• Enablement.

• Talent: Hiring the right people.

Think of it like a Michelin-star kitchen. Each person is specialized. Each part of the system is optimized. They succeed because everything works together.

That’s what top outbound teams look like right now. Most companies are still looking for the one tactic that will fix everything. It doesn’t exist.

And the output tells the story.

• Owner.com and Ramp generate 3–4× more pipeline per rep than their competitors.

• ClickUp has shared publicly that they generate 3× more pipeline with the same number of SDRs.

• Verkada built a system where their MDRs (SDRs) are 4–6× more productive than traditional SDRs.

What’s the difference?

Productivity.

They built systems that let their team spend maximum time on high-quality prospecting, not preparing to prospect. If your reps are spending time on manual research, finding and cleaning data, writing individual emails from scratch, toggling between five tools, you’re burning budget on low-value work.

Top teams build backend systems that push data into tools reps already use.

AI

The best teams are using AI:

• Snowflake’s 15x’d their reply rate using AI to write outbound messages. They iterated 140+ times to get their prompts right.

• Gorgias cut 200 sequences down to 10 modular sequences and uses prompts to write targeted emails at scale.

• Ramp had a fully automated AI outbound motion and killed it a few months ago,they hit a ceiling. Now they’re investing more in the team, not automated outbound.

But here’s what most companies miss: most bought ChatGPT seats and told their teams to use it manually. That’s just adding another tool to the stack.

The real unlock: build AI workflows that run in the background and push outputs directly into your CRM or sales engagement platform, where your team actually works.

The cutters were cutting bad outbound.

Companies that cut treated SDR as a cost line. Hired bodies. Measured activity. Called it outbound. Companies that expanded treated SDR as a system. Invested in enablement. Measured outcomes. When the cost experiment got expensive, they cut. When the system produced returns, they scaled.

Most only had headcount.

Salesloft, a sales engagement platform, now has zero SDRs. They cut their SDR team. They’re PE-backed and focused more on efficiency than growth.

Outreach and Apollo also cut their SDR teams in 2024/2025 and in 2025 started rebuilding them from scratch.

I know three companies that tried cutting SDR teams entirely to replace them with AI. One early-stage company eliminated the team in 2024. By 2025, they hired a Head of BDR and eight BDRs on the floor. Two years without outbound pipeline, then a rebuild from scratch, because AI couldn’t replace the conversations that generate pipeline.

The other two, both around 1,000 employees, are in the same cycle. Cut. Trying to automate.

The SDR role is harder. That doesn’t mean it’s dead.

The role is evolving.

• Manager spans are shrinking, 6.4 reps per frontline lead, down from 8. That’s more coaching time per rep, not less.

• Ramp time hit 3.0 months, the fastest ever, down from 5.7 in 2020.

• Tenure rebounded to 1.9 years from a low of 1.4 in 2022. 97% of orgs now advertise advancement tracks.

GTM engineer

The GTM Engineer role is forming around it.

In November 2024, Brendan Short counted roughly 200 GTM Engineers globally, technical operators who build the infrastructure around SDRs (data pipelines, enrichment workflows, tool integrations) and other GTM teams.

He asked in one of his newsletters: The Rise of the GTM Engineer.

“Will this number be higher or lower this time next year? My bet: it will be much, much higher.”

Indeed, 15 months later:

• 2,125 people hold the title on LinkedIn.

• 67% got it in the last 12 months.

• One in four works at an agency.

4. The economics changed. The bar is higher.

The cost of outbound is rising from every direction.

Winning by Design documented a 10x conversion decline over 8 years: in 2015, 100 cold emails generated one opportunity. By 2018, it took 200. By 2023, it takes 1,000+. The volume required per opportunity has grown tenfold. And it keeps climbing.

Insight Partners advises 500+ B2B SaaS companies. Jeremy Donovan, their EVP of Sales, puts the breakeven at $40-50K ACV: below that, human outbound doesn’t pencil out. That threshold keeps rising. Companies with $20K ACV are redeploying SDR budget into inbound and channel partners.

The exception is extreme velocity, Owner.com runs 8-10K ACV with a 2-call, 6-day sales cycle and generates $70K/month per BDR in closed-won.

But that’s the exception, not the model.

SDR salaries

Base was flat at $55K for three years, then jumped to $60K in 2026. OTE stepped up from $80K to $85K in 2025 and held. Top performer ceiling peaked in 2024 ($131K) and has slightly declined since.

More attempts per opportunity. Higher cost per SDR (2015 → 2026). The economics of outbound have fundamentally shifted.

And yet the best teams are beating those economics. Bridge Group found pipeline per SDR up 26%, from $3M to $3.78M annually. Phone-centric teams generate 1.4x more qualified conversations than email-first teams. Insight Partners’ portfolio data puts it even starker: 70% of meetings are booked by phone. Not email. Not LinkedIn.

The gap between the top and the average is widening, not because of talent, but because of infrastructure & enablement.

You can’t outspend bad systems. You have to build better ones.

What this means for you

Someone posts “SDRs are dead” and it gets 500 likes. A survey says 36% of SaaS companies cut SDR teams. A vendor promises their AI tool can replace your whole team. And suddenly you’re in a board meeting defending the existence of your SDR org.

It depends who you’re learning from.

If you’re taking strategy cues from people who say outbound doesn’t work, you’ll make decisions based on that narrative. If you look at what Rippling, Ramp, Owner.com, and ClickUp are actually doing, building, expanding, investing in systems, the picture is different.

Being bad at outbound doesn’t mean outbound doesn’t work. It means you were bad at it.

Outbound works. Even in 2026.

And in competitive markets, it does something inbound can’t, it manufactures timing. While your competitors wait for hand-raisers, outbound lets you reach the right companies before anyone else knows they’re in-market.

Audit your system. Data infrastructure. AI workflows. Training and enablement. The right people. The winners aren’t doing one of these well, they’re doing all of them. If your system has gaps, fix those before adding headcount.

In March, I’ll share the full dataset (SDR headcount, salaries, GTM Engineer role, and more).

Until next time!

Elric

Sources:

• Original research via LinkedIn Sales Navigator (190K SDR headcount, 232 companies, 119 countries, March-December 2025).

• 2025 SDR Metrics and Trends Report: Bridge Group (SDR metrics, 351 companies)

• SaaStr/Emergence Capital (SDR team cuts, 560+ companies)

• Adam Schoenfeld/PeerSignal (AI company hiring)

• RepVue (SDR compensation)

• Winning by Design/Revenue Architecture (conversion trends)

• Jeremy Donovan/Insight Partners (Revenue Leadership Podcast E61)

As soon as you’re ready, here’s how I can help:

1. Outbound Consulting Call: Need quick help? Book a 40-min session here.

2. Outbound Private Chef:

   1. Part-time, I’ll build, optimize efficiency, and execute your outbound systems.

   2. Coaching/Advising: get tailored advice to fix your strategy, boost efficiency, and drive results.

3. Interested? Reply to this email or hit me up at bonjour@elriclegloire.com

Listen on YouTube, Spotify, Apple Podcasts, and Snipd.

Troy Johnson is the SDR Enablement Program Manager at JumpCloud, where he runs the SDR Enablement program for the Americas. His team books 80-90% of meetings through cold calls. He has over eight years in sales development, including six years managing SDR teams. This is his second appearance on the show, Part 1 covered SDR enablement foundations, and Part 2 dives into how they use AI to make reps more confident on the phone.

Listen on YouTube, Spotify, Apple Podcasts, and Snipd.

This episode was first published on outbound.kitchen.

We discuss:

• How JumpCloud uses AI role-play to certify SDRs on cold calls during onboarding, 100+ practice scenarios before going live

• Running a March Madness-style qualification competition to drive AI adoption and peer coaching

• The monthly rhythm for rolling out updated cold call talk tracks with live practice and manager demos

• Using Gong’s theme spotter to identify the most common phone objections and build targeted AI drills

• Building an SDR-to-AE progression program with AI call scenarios aligned to AE managers

• Why managers should practice AI role-play themselves to lead by example

• Tools used: Hyperbound, Orum, Gong, and ChatGPT Voice Mode as a free alternative

Referenced:

• Hyperbound: hyperbound.ai

• Orum: orum.com

• Troy Johnson on LinkedIn

• Gong

• ChatGPT

My biggest takeaways from this conversation:

Read more

• Instant access to 40+ paid newsletters, including:

  • 6 AI prompts to find and reach accounts 10x faster

  • The Guide to Building an Enterprise Prospecting System

  • [Starter Kit] Step 1: Build your ICP + Persona Creation v1

• My Outbound Chef Kit: 130+ resources (20+ email templates, cold call scripts, AI prompts, top tools, and more). Value: $1,000+

And the best part? You can try it free for 7 days. Cancel anytime.

But I think you’ll love it!

👉 Start your free trial today.

The live sessions are back, first one in 12 months.

The podcast format has changed. No more talking-head conversations. We build live, show screens, walk through real examples.

First session: 5 AI Role-Play Use Cases for Outbound Teams and Cold Calling. Part 2 with Troy Johnson.

Built for GTM leaders developing cold calling skills at scale.

Register here.

You’ve seen this pattern before. Ramp, Owner.com, and now ClickUp.

Outbound teams producing outsized results.

Not just through headcount. Through systems.

This week, Kyle Coleman posted this. Same team size. 3x the pipeline.

In the post, he called out four systems driving the results:

1. XDR activity went from ~35 to ~150 calls per day through infrastructure changes

2. Automated signal monitoring that pushes intel into rep workflows programmatically

3. Every tool and process audited for friction, with XDRs co-authoring every change

4. AI embedded at every layer, context stitching, workflow automation, feedback loops

I already broke down ClickUp’s outbound strategy in a previous issue:

How ClickUp Scaled to $300M With 2 Outbound Teams

Since then, ClickUp’s Growth Ops team shared one of their recipes.

Today we focus on one specific system, the AI-powered call prep that turns a parallel dialer into a context machine. It’s the specific workflow under #4 that directly enables #1. When you use a parallel dialer like Nooks and go from 35 to 150 calls a day, you need a way to maintain quality at that volume. The AI-powered call prep is what makes that possible, it turns every dial into a context-loaded conversation.

Andy Cabasso (Growth Ops Manager) broke down the system architecture, what failed, and the one thing that made it stick.

P.S. key Results from Clickup:

• 11x LTV lift when customers move from self-serve to sales-assisted​

• 3x CAC reduction (50% from culture/team alignment, 50% from incrementality testing)​

• $1M pipeline/month from their inbound AI-SDR in the first month of piloting​ (via Retool)

• $300M+ ARR announced September 2025

• Same team size. 3x pipeline.

• Top SDRs booking 200+ meetings in a quarter.

The Recipe

ClickUp uses Nooks, a parallel dialer that lets reps dial six prospects simultaneously. When someone picks up, you have zero seconds to prepare. Before this system, reps had Salesforce, LinkedIn, Pocus, Outreach, and multiple dashboards open. Scrambling for context mid-conversation.

The Growth Ops team built a system that killed the scramble.

Here’s the exact architecture.

The system runs as an automated workflow in Retool. You could build something similar in Workato, n8n, Make, or Zapier.

Step 1: The trigger. Every time a prospect enters a sequence in their sales engagement platform, that platform sends a webhook to Retool with the prospect details. That webhook starts the workflow.

Step 2: Data pull. A set of SOQL and SQL queries pulls data from their data warehouse tables. They bring in CRM data about the contact and account, recent logged sales activities, details on the current email sequence the prospect is in, and product usage highlights: free vs. paid, signup date, key events.

Step 3: Enrichment. If there are gaps on the contact or company, an enrichment flow runs to update and fill in missing fields.

Step 4: Three parallel AI queries. Full context loaded. Three AI models fire at the same time:

• Query 1 (Perplexity): Looks up the company online and finds recent news relevant to ClickUp’s offer. Summarizes the context and how it ties into their pitch.

• Query 2 (OpenAI GPT) Summarizes the company context, the prospect’s activity history, and their product usage.

• Query 3 (Gemini): Uses everything they know about the prospect plus internal data to develop a full talk track for the rep.

Step 5: CRM write-back. A POST webhook writes these outputs into custom fields on the prospect record in their CRM. Those fields sync to their dialer.

The result: Prospect picks up. Rep already has background, activity summary, and a talk track on screen.

The rep sounds like they’ve been studying the account for 20 minutes. “Hi, I noticed your team has been using ClickUp for project management but hasn’t activated the automations, curious if that’s intentional.” That’s the difference between a hang-up and a conversation.

No training deck does that. Live context does.

That’s the recipe. Not more dials. Better conversations on every dial.

What Broke

The workflow worked on day one. The AI outputs were unusable.

Three things kept breaking:

1. Bad data queries from the warehouse.Wrong fields, stale records, incomplete data, everything downstream breaks. Andy Cabasso’s advice: "if the AI output is wrong, check your SOQL and SQL first. The data is usually the problem”.

2. Poor prompt engineering. Even with clean data, poorly constructed prompts produce generic or misleading outputs. If an AI output looks off, start by improving the prompt, tightening the instructions, and giving clearer examples.

3. Misalignment between what reps needed and what the system delivered. This one is the least obvious. The system might be technically sound, clean data, good prompts, but if it delivers information the rep doesn’t actually need in the moment, it’s noise.

Here’s what that looks like in practice:

“A prompt might tell a rep the prospect is on a free plan when they’re actually a paid customer. That’s a credibility-killer. And in a cold call, there’s no recovery from getting basic facts wrong.”

One wrong output and the rep never opens it again.

They tested everything. Different models for different jobs:

• Gemini performed best for structured, rule-based tasks

• GPT handled creative prompts and flexible phrasing

• Claude had decent real-time performance, but still lagged

• Perplexity was essential for up-to-date web context

No single model does everything well. If you’re running one model for every task, you’re getting mediocre output on most of them.

What Actually Made It Work

Their SDRs told ops when the outputs were wrong. Every time.

• The talk track was off → they said so.

• The data was stale → they flagged it.

• Format wasn’t useful → they pushed back.

Without that loop, the system would’ve looked good in a demo and died in production.

Most companies never get here. They invest in the architecture, the workflows, the integrations, the AI models. They ship something that looks impressive in a demo. And then reps quietly go back to their old way within a week.

The feedback loop itself is simple, a Slack channel or a weekly review. The hard part is reps who actually say “this output is wrong” and ops teams who actually fix it.

ClickUp got this right because they started with the culture before they scaled the technology. One of their SDRs flagged the original problem and then helped shape every iteration of the solution.

Talk tracks got more accurate. Reps stopped checking Salesforce manually. Call conversion improved. Each iteration compounded because the people using the tool were the same people shaping it.

The Takeaway

This is one system. ClickUp’s 3x didn’t come from the AI call prep alone. The first breakdown covers the rest of their outbound strategy. Read my first deep dive here.

Three questions to audit your own AI implementation:

1. Is the data correct? Audit your SOQL/SQL queries before blaming the model.

2. Are the prompts specific enough? Vague instructions produce generic results. Tighten the input, tighten the output.

3. Do the people using it have a voice in improving it? If reps can’t flag bad output and see it get fixed, they’ll stop using it.

ClickUp took reps from ~35 calls a day to ~150, with a top performer booking 200 meetings in a quarter. We’re still finding the ceiling on what a rep can do when the systems keep getting better.

Follow Kyle Coleman (Global VP Marketing, ClickUp) and Andy Cabasso (Growth Ops Manager) for the source material.

Until next time, build systems, not shortcuts.

Elric

What’s coming in February:

• Benchmark for U.S. mobile phone numbers (the first of several)

• 2025 outbound benchmark and recap

• How I use Clay

Thinking about upgrading to the paid newsletter? Here’s what you’ll get:

• Instant access to 40+ paid newsletters, including:

  • 6 AI Prompts for Researching Accounts & Cold Outreaching Them (10x Faster)

  • The Guide to Building an Enterprise Prospecting System

  • [Starter Kit] Step 1: Build your ICP v1 + Persona Creation v1

• My Outbound Chef Kit: 130+ resources: 20+ email templates, cold call scripts, AI prompts, top outbound tools, and more, everything the best outbound teams use. Worth $1,000+

And the best part? You can try it free for 7 days. Cancel anytime.

But I think you’ll love it!

👉 Start your free trial today.

As soon as you’re ready, here’s how I can help:

1. Outbound Consulting Call: Need quick help? Book a 40-min session here.

2. Outbound Private Chef:

   1. Part-time, I’ll build, optimize efficiency, and execute your outbound strategy.

   2. Coaching/Advising: get tailored advice to fix your strategy, boost efficiency, and drive results.

3. Interested? Reply to this email or hit me up at bonjour@elriclegloire.com

• Instant access to 40+ paid newsletters, including:

  • 6 AI prompts to find and reach accounts 10x faster

  • The Guide to Building an Enterprise Prospecting System

  • [Starter Kit] Step 1: Build your ICP + Persona Creation v1

• My Outbound Chef Kit: 130+ resources (20+ email templates, cold call scripts, AI prompts, top tools, and more). Value: $1,000+

And the best part? You can try it free for 7 days. Cancel anytime.

But I think you’ll love it!

👉 Start your free trial today.

I’ve been using Claude Code since December. It hasn’t replaced Clay, but it’s changed how I use Clay.

Claude Code now comes before Clay in my workflow. It’s become my prep layer, the place where I do the thinking work before Clay does the enrichment and sequencing.

Here are 9 ways I use Claude Code:

1. Call prep that shows value (Before and during the call)

2. Tradeshow attendee workflow

3. Account-Based Campaign Builder

4. ICP/TAM analysis

5. Account list building

6. Merging and cleaning CSVs before Clay

7. Web scraping

8. For one client: a custom CRM lookup app because their CRM’s API was too limited for what I needed

9. Building a knowledge layer from customer conversations

For anyone who hasn’t used it:

Claude Code is Anthropic’s command-line tool that lets you give Claude tasks and it executes them, writing scripts, pulling APIs, scraping websites, processing files, building workflows. Think of it like having a technical co-pilot that can actually do things on your computer, not just answer questions.

It’s built for developers. I’m not a developer. But it turns out you don’t need to be one to get serious value out of it for outbound.

The problem with AI tools for outbound is the same problem with Notion. If someone hands you Notion and says “go,” you’re staring at a blank page. You need to see specific examples of how other people use it before it clicks. That’s what this newsletter is, the specific ways I’m using Claude Code for outbound right now, so you can see what’s possible and steal the ideas that make sense for you.

How I Use Claude Code for Outbound (On Top of Clay)

1. Call prep that shows value (Before and during the call)

I recently listened to Jeanne DeWitt Grosser, COO of Vercel (and former CBO at Stripe), talk about treating Go-To-Market (GTM) like a product, obsessing over the customer experience at every touchpoint. The line that stuck with me:

“Make every sales interaction great, whether customers buy or not.”

That made me rethink how I show up to prospect calls. I started asking myself: what am I actually bringing to this call? Because if the answer is just a list of qualifying questions, I’m not giving the prospect any reason to care.

So I started building things for them before the call even happens. Not just to show value, to educate them on how they should be thinking about their market:

• Here’s how you should think about your TAM.

• Here are companies showing specific pain signals online right now that could be great accounts for you.

• Here’s what the opportunity actually looks like.

And then I hand them a sample of what I’ve already done for them, so whether we end up working together or not, they walk away with something useful.

I’ve trained a Claude Code workflow on SPICED and Gap Selling, the same frameworks I use with clients. When I give it an account, it looks for:

• Situations that signal pain

• Problems I can tie to my services

• Critical events that might accelerate a deal

• Gaps between current state and where they want to be

And then I use that research to build actual deliverables I can share on the call:

• TAM analysis: for one of their markets, showing where the opportunity is

• In-market accounts: companies showing specific pain signals online that could be strong prospects for them, with the reasons why

• Account mapping: showing business units within a target account

• Account-Based Campaign Examples

The key decision I made here was to build these as visual dashboards instead of just handing over CSVs. I could give them a spreadsheet and let them scroll through rows. But when you show someone a visual breakdown of their market, where the opportunity is, which accounts are showing pain, how the segments break down, it lands completely differently. They can actually see what the opportunity looks like for their business. A CSV is data. A visual is a story.

Last week, I sent a message to a prospect before our call:

“Hey, I prepped a TAM analysis for the market you mentioned. Super excited to walk through it.”

During the call, they were already engaged. Because I’d already given them something useful before we even started talking.

Here's an example for the TAM analysis:

I think applying the idea of GTM as a product experience to outbound means exactly this: when someone gets on a call with you, they should get value whether they buy or not. Claude Code is what’s letting me do that without spending half my day on building this.

2. Tradeshow attendee workflow

This one’s still a work in progress, but it’s already saving hours of manual work.

Here’s what my old process looked like:

1. Go through the tradeshow website manually

2. Study the website to figure out what scraper I need (Apify, etc.)

3. Scrape the information

4. Bring it into Clay to cross-reference against the CRM

5. Look up contacts in another Clay table

6. Build a separate table to find contact info for people who aren’t in the CRM yet

7. Package the output and send it to my customer

That was multiple Clay tables, multiple tools, and a lot of manual decision-making at each step.

Now, depending on what the tradeshow gives you access to, there are a few ways I handle this with Claude Code:

Scenario 1: Tradeshow website (sponsors, speakers, exhibitors)

The most common one. I give Claude Code the tradeshow URL. It scrapes exhibitors, sponsors and speakers, figures out for each whether it’s a person or a company, checks if they fit my client’s ICP, looks them up in the CRM, finds the right contacts, and pulls contact info for anyone missing.

Scenario 2: Web app with full attendee list

Some tradeshows give you access to the attendee list through a web app. Sometimes Claude Code can scrape it directly via Apify or other tools. Other times I download the HTML and feed it to Claude Code so it can analyze the full list. Either way, it handles the enrichment from there.

Scenario 3: Mobile app only (what’s next)

Some tradeshows lock the attendee list behind a mobile app with no web version. Jordan Crawford recently showed a clever workaround for this: screen share from your phone to your computer, then use Claude Code to take screenshots and extract the information. I’m not doing this yet, but that’s my next step when a client needs it.

In all three scenarios, Claude Code handles what used to be steps 1 through 6 of my old process. The output is a CSV ready for Clay, which then pushes to the CRM. I send my client a Google Sheet with everyone prioritized — who to talk to first, who’s already a customer, who’s net new.

What used to take hours now takes under two. And I’m not done, the next version will include photos of each person and a talk track on why they’re worth talking to, so the team knows who to find on the floor. I’ve seen Jordan Crawford do something similar, and I want to take it further by pulling CRM context into that view.

3. Account-Based Campaign Builder

This is the one I’m most excited about, and the one I’m still actively building.

Right now, when I build a campaign for a client in Clay, it takes a long time. I’m creating multiple tables, clicking through enrichment steps, mapping contacts to accounts manually, writing messages one pattern at a time. The setup work before a single email goes out is significant.

Here’s what I’m building with Claude Code instead:

I start with two CSVs. One is the list of companies I want to prospect. The other is the contacts at those companies. I feed both to Claude Code, and from there, it connects the dots.

It maps which contacts belong to which companies. It looks at each account and finds the angles, who at that company could be a name-drop, who’s a recent hire, what’s happening at the company that creates an opening. It does the account research for each company on the list: what they do, what pain signals they’re showing, what’s relevant to my client’s value prop.

Then it pulls from my library of POV, talk tracks, cold emails and LinkedIn messages, templates I’ve built and refined across clients, and matches the right message to each situation and each contact. A recent hire gets a different message than a VP who’s been there three years. A company that just raised funding gets a different angle than one that just laid off their SDR team.

The output can go a few different directions depending on what I need:

• A CSV ready to go with the full emails or messages written out, plus a POV talk track for the reps so they know why each person is getting that message

• Or even pushing this directly to the systems where the rep are spending their time

• A prompt built specifically for Clay, so I can import it directly and Clay handles the enrichment and sequencing from there

Instead of building multiple Clay tables and configuring each one, I’m doing the thinking and the mapping in Claude Code first, and Clay becomes the execution layer.

I’m still building this out. The workflow isn’t fully systematized yet, right now I’m doing pieces of it across different sessions. But I can see where it’s going. When it’s done, I should be able to take a campaign idea, feed it two CSVs, and get back a fully researched, fully written campaign ready to load into Clay or hand to a rep.

I think this could be a game changer, not just for my own business, but for anyone running account-based outbound. The end goal is building this into a system that anyone can plug into. That’s what I’m working toward.

What Else I’m Running Through Claude Code Weekly

• ICP/TAM analysis: for clients and for my own prospect research

• Account list building

• Merging and cleaning CSVs before Clay

• Web scraping (or building Apify scrapers for harder sites)

• For one client: a custom CRM lookup app because their CRM’s API was too limited for what I needed

• Building a knowledge layer from customer conversations, I’m storing all my conversations with customers and prospects (call notes, pain points, objections, what they reacted to) and layering additional research on top (company data, market context, competitive landscape). Every conversation feeds it. Right now I’m using Clarify as my CRM and working on connecting it to Claude Code via MCP so this flows in automatically. The more context Claude Code has, the better everything else gets, call prep, account research, campaign setup. And when you store enough conversations across enough prospects, you start spotting patterns that sharpen your outbound messaging and targeting.

Why I use Claude Code over other AI tools?

It’s faster than Clay for certain tasks.

I love Clay. But for some things, merging CSVs, quick data analysis, prepping account lists, doing lookups, Claude Code is just faster. In Clay, you need to connect your accounts, create a table, set up your columns, configure your enrichments. That setup time adds up. In Claude Code, I describe what I need and it runs. For the stuff that needs to go into Clay, I prep it in Claude Code first so the Clay table is cleaner and simpler.

It connects to your stack.

Claude Code can connect directly to APIs and MCPs, Lemlist, EXA, Apify, client CRMs, so it can pull data, run lookups, and build lists on its own. That said, I only use it for reading right now. I don’t let Claude Code write to CRMs. I don’t trust it to not break something yet. Clay handles the write layer, which gives me a checkpoint before anything hits a client’s system.

It compounds.

Most AI tools are one-shot. You ask, you get an answer, you start over. With Claude Code, the workflows I build for one client, call prep, tradeshow scraping, ICP analysis, I reuse and improve for the next one. The frameworks I’ve trained it on (SPICED, Gap Selling) carry forward. Each client engagement makes the system better for the next client, even though I’m not sharing any data between them. The skills and the knowledge compound over time.

I’m just getting started.

I’ve been doing this since December. Most of what I showed you is V1 or V2. I can already see where this is going, and it’s way bigger than what I’m doing today.

I’m not saying any of this is perfect. The tradeshow workflow is still V2. The campaign builder is still coming together. But that’s the point, I’m building in public, and this is where I am.

The shift from “prepping to qualify” to “prepping to educate and show value” has changed how prospects show up to my calls. And Claude Code is what’s making it possible.

Elric

Thinking about upgrading to the paid newsletter? Here’s what you’ll get:

• Instant access to 40+ paid newsletters, including:

  • 6 AI Prompts for Researching Accounts & Cold Outreaching Them (10x Faster)

  • The Guide to Building an Enterprise Prospecting System

  • [Starter Kit] Step 1: Build your ICP v1 + Persona Creation v1

• My Outbound Chef Kit: 130+ resources: 20+ email templates, cold call scripts, AI prompts, top outbound tools, and more, everything the best outbound teams use. Worth $1,000+

And the best part? You can try it free for 7 days. Cancel anytime.

But I think you’ll love it!

👉 Start your free trial today.

As soon as you’re ready, here’s how I can help:

1. Outbound Consulting Call: Need quick help? Book a 40-min session here.

2. Outbound Private Chef:

   1. Part-time, I’ll build, optimize efficiency, and execute your outbound strategy.

   2. Coaching/Advising: get tailored advice to fix your strategy, boost efficiency, and drive results.

3. Interested? Reply to this email or hit me up at bonjour@elriclegloire.com