GrowthKit AI
/Manifesto

The market is knowable.

Most go-to-market decisions are made on vibes — not because founders are careless, but because the rigor was never affordable. We think that era is over. What follows is what we believe, written down so you can hold us to it.

Founding document Filed — London, 2026 Seven articles
00Preamble

Every founder we've met can draw their architecture from memory. Ask them to draw their market, and you get three logos and a shrug.

Somewhere between the roadmap and the launch, the rigor runs out. The product is versioned, tested, instrumented. The go-to-market is a doc called growth ideas and a thread of screenshots from someone else's launch. Consultancies sold the missing rigor at six figures. Dashboards sold the data without the judgment. Everyone else guessed.

We started GrowthKit AI because guessing is now a choice — and we choose not to.

01The articles

Seven things we refuse to be vague about.

Five beliefs, two rejections. They shape what the engine measures, what the deliverables say, and who we choose to serve. When a product decision is hard, we read these and the decision usually stops being hard.

Art. 01 / Believe

Growth should be engineered, not guessed.

Your product ships from a roadmap. Your infrastructure has runbooks. Then the go-to-market gets decided in a brainstorm at 9pm. The same discipline that builds the product should choose its market — measure, dissect, name the gaps, then act. Vibes are not a methodology.

Art. 02 / Believe

Evidence beats opinion.

Every claim in a GrowthKit deliverable traces back to something we can show you — demand signals, mined complaints, pricing matrices taken apart line by line. If we can't source it, we don't say it. An investor's hot take is not a data point. A thousand angry reviews of your competitor are.

Art. 03 / Believe

Market fog is a tooling gap, not a fact of nature.

The signals exist: what buyers search, what they pay, what they complain about, where incumbents have gone soft. Nobody had assembled them for teams without a research budget. That's an engineering problem — and engineering problems get solved.

Art. 04 / Believe

Plans should produce decisions, not reports.

Research that ends in a deck is decoration. Ours ends in a sequenced 90-day plan — fourteen plays, each with kill-criteria — built to keep deciding for you in week four, when the launch adrenaline is gone and the calendar is winning.

Art. 05 / Believe

Intelligence has a half-life.

A market read from January is a guess by June. Competitors reprice, channels saturate, complaints move. That's why every GrowthKit plan refreshes monthly — not because dashboards are pretty, but because stale intelligence is just confident misinformation.

Art. 06 / Reject

Consulting-grade rigor at consulting-grade prices.

A strategy engagement runs six months and six figures — and most of that buys process, not insight. We deliver the synthesis without the theatre: about ninety seconds of compute, reviewed by operators, for the price of a SaaS seat. The margin consultancies defend is the gap we live in.

Art. 07 / Reject

Being for everyone.

Pre-idea founders need customer conversations more than market maps. Enterprises with a staffed intelligence team don't need us at all. We build for seed and Series A teams with a live product — the stage where one well-chosen wedge compounds hardest. If that's not you, we'd rather say so here than after you've paid.

/filed specimen №00 — founding document

Markets, dissected — not guessed.

The founding teamGrowthKit AI
London · 2026 · v0.4 — and counting
02Hold us to it

If this reads like how you already think — get in early.

The waitlist is where the first cohort comes from. The methodology page shows the machine these beliefs built.