How GTMcraft works

From GTM problem to repeatable operating system.

GTMcraft helps you answer four questions.

  • What is actually blocking growth?
  • What should we do about it?
  • How do we turn that decision into execution?
  • What can AI run once the process works?
That becomes one operating loop
Diagnose→ Decide→ Execute→ Compound
Find what to fix next →
Why this exists

Most GTM problems aren't knowledge problems.

  • You probably already know that pipeline matters.
  • You know conversion matters.
  • You know your ICP should be tighter.
  • You know your managers should coach.
  • You know AI should make the team more productive.

Knowing isn't the bottleneck. Turning the right decision into consistent execution is. That's what GTMcraft is designed to solve.

Step 1 · Diagnose

Find the constraint behind the number.

A missed target is a symptom. Before changing strategy, adding people or buying software, understand what is underneath it.

GTMcraft works backwards
Business outcome→ Metric→ Capability→ Behavior→ Play
Instead of
Problem→ random tactic
The result is a much clearer question: what is the smallest constraint we can remove that materially changes the outcome?
Step 2 · Decide

Know what to fix, and what to ignore.

Once the constraint is clear, GTMcraft connects it to the relevant operating patterns and plays. The intelligence layer combines:

  • Current GTM signal.
  • Patterns from leading operators.
  • 14+ years of field experience.
  • Your stage, market and motion.

The goal isn't to give you ten recommendations. It's to identify the move worth making now.

Step 3 · Execute

Turn the decision into a play.

A GTMcraft Play isn't an article. It's an operating system for a specific GTM problem. Depending on the play, that can include:

  • Diagnostic.
  • Step-by-step implementation.
  • Templates.
  • Workbooks.
  • Metrics.
  • Manager coaching.
  • Operating cadence.
  • AI workflows.
  • Agent instructions.
  • Governance.
The goal is simple: move from "we should do this" to "this is how we run it."
Step 4 · Compound

Once the process works, make it cheaper to run.

This is where AI becomes useful. Not before the process exists. Once you know:

  • What good looks like.
  • What context matters.
  • What steps repeat.
  • What requires judgment.
  • What can be measured.

You can give repeatable parts of the workflow to AI.

Process first. System second. AI third. Humans stay accountable for judgment.
The architecture

One intelligence layer. Three execution layers.

Intelligence

100,000+ GTM signals from leading operators, founders, podcasts, newsletters and field experience.

↓
Judgment

What matters for your stage, motion and constraint.

↓
Plays

The operating system for solving the problem.

↓
Human Skills

What your team needs to learn, practice and improve.

↓
AI Skills + Agents

What AI can execute repeatedly once the system is clear.

↓
Metrics

The evidence that tells you whether the system is working.

Human vs AI

Don't automate judgment. Automate repeatability.

Keep human
  • Strategy.
  • Customer conversations.
  • Negotiation.
  • Leadership.
  • Coaching.
  • Context-heavy decisions.
Systemize
  • ICP.
  • Positioning.
  • Methodology.
  • Processes.
  • Metrics.
  • Cadences.
  • Standards.
Give AI
  • Research.
  • Preparation.
  • Analysis.
  • Documentation.
  • Monitoring.
  • Administration.
  • Repeatable workflows.
Start here

Start with the constraint, not the solution.

If you already know exactly what is broken, find the relevant play. If you don't, start with the diagnostic. Either way, the objective is the same: find what to fix next, then build the system to fix it.

Start with GTMcraft →