GTM research, built like a decision document.
An AI analyst delivers a structured GTM brief — sourced, opinionated, ready to defend. Not another optimistic AI score. Not a hundred ChatGPT prompts and forty browser tabs. The document that backs the call you’re about to make.
Built by Alexander Sass, founder · Berlin
€299, not €19. Us, not ChatGPT.
Every founder asks the same two questions before paying. Here are the honest answers — because if you’re going to bring this conversation to a co-founder, an advisor, or an investor, the document you bring matters.
It’ll happily validate any idea you bring it.
ChatGPT is a coach. It says yes. The brief you’d get from it is shaped by what you ask, in the order you ask it, forgetting context after fifty messages. That’s a conversation, not a document.
- Hallucinated citations and made-up URLs.Live web search; every claim links to a real source you can click.
- Optimistic by default — “here are some considerations…”Refuses when the evidence says no. Conditional Go / Pivot / Pass with explicit gates.
- You have to be the analyst — knowing what to ask, in what order.We bring the structure (8 sections, dependency-ordered). You bring the idea.
- A chat thread. You can’t attach it to an email.A 1,100-line document. Print it, share it, mark it up.
A “Score: 87/100” isn’t a decision document.
Cheap AI validators sell shareability — a number you can screenshot. Their business model is making you feel validated enough to tweet. Ours is making you decide. Different shelf, different buyer.
- A 3-paragraph summary plus a confidence score.An 8-section structured brief: market sizing, ICP, competitors, channels, pricing, unit economics, risk, next steps.
- Generic risks (“competition is tough”).Geography-specific traps caught — UWG, GDPR, Kleinunternehmer VAT cliff for German founders, regulatory context the AI training data doesn’t cover.
- Sandboxed — summarises training data, doesn’t research the live web.Hits the actual web. Pulls competitor pricing from their pricing pages, not from 2023 training snapshots.
- Won’t refuse — that’d hurt the “share your score” loop.Frequently lands on Pivot or Pass. Saving you months of burn is the whole point.
The right comparison isn’t ChatGPT or DimeADozen. It’s the freelance researcher you’d otherwise pay €500–2,000 to spend two weeks producing roughly the same document. We deliver it in thirty minutes for a fraction of that — because the AI does the legwork and the structure does the thinking.
Three steps. Thirty minutes. One defensible call.
Tell us about the idea
Five required fields, optional context for whatever you have. The more specific you are, the sharper the brief — five minutes to fill out.
AI analyst writes the brief
Six (Essentials) or eight (Pro) sections. Live web searches, citations on every claim, sized against named comparables. Verdict at the top: Conditional Go, Pivot, or Pass — with the gates that would change it.
You receive the document
About thirty minutes after Stripe Checkout. A 1,100-line brief in your dashboard, printable, shareable, defensible. 14-day quality guarantee — re-run or refund if a section is missing, citations don’t resolve, or the verdict contradicts cited evidence.
Eight sections. Each one earns its place in the document.
The same eight sections analysts and operators reach for when they pressure-test a market from scratch. Essentials covers the six research sections; Pro adds channel strategy and pricing & unit economics. Both tiers carry the executive-summary verdict and the 30-day action list.
- 01
Executive Summary
All tiersThe page you’d hand to an investor. Verdict (Conditional Go / Pivot / Pass), confidence level, the conditions that would flip the verdict, and the top three actions for the next thirty days. Written last; stands alone.
- 02
Market Sizing — TAM, SAM, SOM
All tiersTop-down estimates with named industry sources, geography splits, and Year-1 / Year-2 / Year-3 capture against named comparable companies. Regional slices labelled when extrapolated rather than directly sized — so you know what’s measured versus modelled.
- 03
Customer Segments & ICP
All tiersPrimary segment, secondary segments, buying triggers, and reachability. Pro adds the ICP narrative — day-in-the-life, decision criteria, the sharp end of the buying motion. The slide you’d build out for a customer-development plan.
- 04
Competitive Landscape
All tiers3–10 competitors with current pricing pulled from their live pages, positioning map, and feature matrix. Pro adds moat assessment — honest about whether the moat exists yet, with the “No moat yet” verdict on the table when it’s warranted.
- 05
Channel & Distribution Strategy
Pro only3–5 ranked channels with CAC estimates anchored on named benchmarks, distribution-model recommendation (self-serve / sales-assisted / hybrid), and a 90-day launch plan with go/kill thresholds at day 30, 60, and 90.
- 06
Pricing & Revenue Model + Unit Economics
Pro onlyRevenue-model comparison, competitor pricing benchmarks, willingness-to-pay analysis, plus a per-unit P&L with LTV:CAC, payback period, sensitivity scenarios, and path-to-break-even. The numbers you’d defend in a fundraising conversation.
- 07
Risk Assessment
All tiers5–12 risks ranked by likelihood × impact, with mitigation paths and load-bearing assumptions named. German regulatory traps caught — UWG cold-outreach, GDPR, the Kleinunternehmer VAT cliff at €22k.
- 08
Recommended Next Steps
All tiers30-day action list grounded in §3–§7, plus validation experiments with explicit success thresholds. Pro adds a 90-day roadmap with go/no-go gates by month — the operating plan the brief funnels into.
Three ex-ante briefs. What we would have written, before the outcome was known.
Each one is a full Pro brief produced for a famous startup at its founding moment — using only sources, market data, and signals available at the time. No hindsight, no narration around the outcome we now know. The point is to show how this kind of analysis reads when the answer isn’t obvious yet.
AirBed&Breakfast
August 2008Verdict: Conditional GoPre-YC, ~60-day runway, three event runs (IDSA, SXSW, DNC), no cold-traffic validation. Cereal-box fundraising is months away.
Read the brief →Personio
Early 2015Verdict: Conditional GoFour CDTM grads with no HR-tech experience, ~20 private-beta customers in Bavaria, the dominant prior is “DACH SMBs don’t buy SaaS.”
Read the brief →Figma
December 2015Verdict: Conditional GoPrivate beta, Sketch dominates the SF design community, “browsers are too slow for vector design” is the consensus prior.
Read the brief →
Each brief uses the same eight-section template as paid Pro briefs. Methodology notes call out where no primary source was available at the framing date — intellectual honesty is the entire point.
Founders building a real business case.
- →A solo or technical founder with an early idea, building the business case before you commit months of cycles or talk to investors.
- →Pre-seed and weeks away from your first investor conversation — you need a structured document you’d be willing to attach to an email or read off a screen in a pitch.
- →An operator evaluating idea #2 after a plateaued v1 — and you want outside pressure on the new one before committing.
- ×A serial founder with a warm investor network — you validate through conversation, not reports, and you should.
- ×Looking for a €20 AI report that tells you the idea is great. Briefs frequently land on Pivot or Pass — we don’t soften the verdict.
- ×Seeking ongoing advisory hours. This is a one-shot structured deliverable, not a retainer — though the optional expert tier in the works will offer a debrief call alongside the brief.
The things people ask before paying.
- Why pay €299 instead of using ChatGPT?
- ChatGPT will produce something — but it’s a chat, not a document. It says yes to most ideas, hallucinates citations, and needs you to be the analyst (knowing what to ask, in what order, with what depth). You’re paying us for the structure (8 sections in dependency order), the rigor (live web search with real source links on every claim), the honesty (refuses with Pivot or Pass when the evidence demands), and the artefact (a 1,100-line brief you can attach to an investor email). Plus the time — about thirty minutes versus a Sunday afternoon of prompt-wrangling.
- What about the €19–49 AI validators (DimeADozen, ValidatorAI, etc.)?
- Those tools run a prompt against your idea and return a 3-paragraph summary plus a confidence score. The score sells — it’s shareable. But your investor doesn’t read scores; they read documents. We give you the document. Different shelf, different buyer. The right comparison for what we deliver isn’t the €20 AI tool — it’s the €500–2,000 freelance researcher you’d otherwise pay to spend two weeks on the same artefact.
- What if the brief says “don’t build this”?
- That’s often the most valuable outcome. The brief frequently lands on Conditional Go or Pivot, and founders who act on a real verdict save months of burn. The 14-day guarantee isn’t a “change your mind” refund — it covers product defects: a missing section, citations that don’t resolve to real sources, or a verdict that contradicts the brief’s own evidence. Re-run for fixable quality issues, refund for the rest.
- Is there expert review on top of the AI brief?
- Not on the live €299 / €499 tiers — those are AI-only with web-search citations and the same 14-day quality guarantee. We’re evaluating a separate expert-review tier — a named operator’s verdict, top risks they’d push back on, and a 30-min debrief call. If demand justifies recruiting a vetted panel, we’ll launch it. Interest signup lives on the pricing page; no money changes hands until we ship.
- What happens to my idea and my data?
- Intake data is processed by the Claude API under a signed DPA; Anthropic does not use the data to train models and retains it for up to 30 days for abuse review only. Your brief is private to your account. Full details in the Datenschutzerklärung, linked below.
- How long does it really take?
- About thirty minutes from the moment Stripe Checkout clears. The AI analyst runs in the background; watch live status on your run page or close the tab and come back to an email when the brief lands.
Bring a real document to your next conversation.
Pick a tier, submit your intake, pay through Stripe Checkout. The AI analyst starts within seconds; the brief lands in your dashboard about thirty minutes later. 14-day quality guarantee — re-run if a section is weak, refund if the brief is broken.