The Claude for GTM resource | Follow for Claude Code GTM skills founders use to build predictable pipeline | Get full access👇🏼 | DM for coaching/training.

Joined June 2022
17 Photos and videos
You're paying $800/mo for lead enrichment tools that a terminal command can replace. I wrote a step-by-step guide for non-technical founders to build GTM lead lists with Claude Code. Install to first output in under 10 minutes. --> Grab it for free here: foundersgtm.com/p/claude-cod…
1
5
2,358
Clay’s enrichment moat died with MCP. Monts of running without it confirmed that. Clay’s orchestration moat is still alive but shrinking. Routines is closing the scheduling gap. By May 2027, the founding teams that learned Claude Code now will be running their full GTM motion from Skills and Routines. The ones who did not will still be paying $495 a month for something a prompt can do.
1
5
Everyone expected AI to make outbound faster, and it did. Speed was never what held results back though. Relevance was, and relevance still depends on a judgment call about who to talk to and why they would care this week. That part stayed exactly where it was.
1
2
Stay on Clay if: - You write into HubSpot or Salesforce at volume and the guardrails matter - You have an ops person who is not the founder running the workflows - You run more than 1,000 rows per workflow per week - You have twenty or more conditional branches in a single workflow and you need them visible Move everything to Claude Code if: - You build the workflows yourself - You have hit the row cap and you are looking at the price jump anyway - You want to read the unstructured layer: call transcripts, deal notes, the language your buyers actually use
1
8
A great network hides an empty pipeline the same way a great close rate hides a thin one. Both feel like winning until the source dries up. Here’s the prompt that finds the middle you’re missing. open.substack.com/pub/founde…
1
7
When anyone can build the same stack in an afternoon, owning the stack stops meaning anything. What you understand about your market survives that shift. A competitor can buy your subscriptions by Friday and still have no idea why your customers actually say yes.
1
2
8
There’s a kind of outbound that treats the prospect as a target and a kind that treats them as a person with a problem you happen to understand. They can look almost identical on the surface. The prospect can tell the difference instantly, which is why one gets archived and the other gets a reply.
1
12
The reason a non-technical ops person can run Clay is the visual table. Fifty parallel columns and you can see what is failing. That is the moat that lasts longest. But it is also the moat that matters least if you are a bootstrapped founder running your own outbound. You are the operator. You did not have an ops person to hand the table to.
1
13
Here’s how to build a lean outreach campaign in 15 minutes, with Claude Code lead enrichment tool. Get the full breakdown here: substack.com/home/post/p-201…
1
15
Where Claude Code wins for GTM: - Writing cold emails that sound like a person who understands their buyer’s situation - Defining a sharp ICP by pattern-matching across customer interviews and win/loss notes - Prioritising accounts based on judgment rather than just a score - Skills ecosystem for outbound - MCP patterns for lead enrichment tools Where Codex wins: - CRM integrations - Lead scoring models - Webhook setup - Data pipelines, automation infrastructure, and design-to-code workflows from screenshots or wireframes.
1
32
The founding teams who win the next two years of GTM will not be those with the most sophisticated tech stack. Instead, they will be those who understood their motion well enough to know exactly what to automate and what to keep human. The stack is a consequence of the clarity. Never the other way around.
1
16
Open any LinkedIn inbox and you’ll see the same polite noise: “Great to connect,” “Appreciate the add,” “Loved your latest post.” People send these thinking they’re being friendly, but all it signals is that they have nothing to say next. These openers don’t start conversations. They end them.
3
4
72
Instead of lowering the bar for outbound, AI raised it. When everyone can generate a personalised-looking message in seconds, the personalised-looking message stops working. What still works is the thing AI cannot fake on its own: a founder who actually understands the second-order consequences of the problem they solve.
3
17
Thank you @Substack, means a lot! Join foundersgtm.com to get the Claude Code GTM skills founders use to build predictable pipeline, instead of outsourcing it. [includes Claude Code Starter Guide upon sign-up]
3
39
Before MCP, a typical GTM lead enrichment flow looked like this: You → Clay → Data Provider → Clay → You You paid Clay at every point in that chain. Their platform fee, their credit markup on top of the underlying data, and your time managing a visual interface that required attention at every step. After MCP, the flow looks like this: You → Claude Code → Data Provider → Claude Code → You
2
24
There’s a difference between a prospect who cannot afford you and one who has not yet understood the cost of their problem. The first is a qualification issue. The second is a positioning issue. Dropping your price solves neither, and usually signals you were not sure of the value to begin with.
1
14
"We forget that we're at the edge of the spear." A founder running a software dev agency told me this when I asked where his clients were with AI. "Most people are not using it. Nobody cares. When you talk about AI, they think ChatGPT, and that is it. Unlike us, who notice when OpenAI ships something the day before. That’s not the case in enterprises." The bubble is real. If you spend your days inside the AI conversation, it’s easy to assume everyone else is too. They are not. Which is either a problem or the entire opportunity.
1
26
The most expensive thing in early-stage GTM is not a wasted ad budget or a bad hire. It is the six months you spend running a motion you never properly measured, convinced it was working because you were busy. Activity feels like progress right up until you check the numbers.
1
22