AI-assisted, not AI-built
The engine accelerates the measurement. The database, the judgment, and the accountability are human. Every report carries a name, not a model number.
For AI GTM founders
Your GTM needs an eval, too.
We measure which buyer, which price, and which words make you your buyer's obvious choice: every score with its evidence. HUMAN-SIGNED
Built on fifteen years of positioning practice at NVIDIA, SAP, and venture-backed startups. HIGH · verifiable
Every claim on this page carries its own confidence tag: the same honesty system the instrument applies to your business.
CLAY · PUBLIC CASE
BENCHMARKS AS OF Q3 2026
Sound position
71 / 100 · minor drift
Above 60 means the position is sound: what needs fixing is the wording, not the strategy.
Verdict: the line nobody can copy, "GTM development environment", sits one click too deep.
Buyer match7/10MED
Price unit7/10MED
Buyer language8/10MED
Product vs message7/10HIGH
Competition6/10MED
Reviewed & signed: C. Han
2 gaps · 3 moves · read the case study →
The market you're selling into
Every AI sales tool claims 3× pipeline. Buyers stopped believing any of it.
The AI GTM category is consolidating in public, and its buyers have been trained to discount every quantified claim. Positioning, not product, is deciding who survives.
50–70%
Annual churn on AI SDR deployments: roughly double human SDR turnover. Pilots convert; renewals don't.
MED · attributed1$850–$10K
Monthly price range across near-identical AI SDR offerings. A category that hasn't priced its own value.
HIGH · sourced26+ of 20
AI SDR vendors that shut down, pivoted, or were acqui-hired within five quarters. The shakeout is live.
HIGH · sourced3By the market's own jobs index, 95% of GTM spend executes decisions made by the 4.5% of jobs companies staff worst. Not one of the 43 jobs in that index is checking those decisions. That job is the GTM eval. MED · attributed4
The instrument
The five areas
Five areas. Dozens of checks under each. One score out of 100.
1Does your buyer match your value?Your current buyer, scored against the buyers you could be selling to: how much the problem hurts, what they can spend, how badly served they are, and how easily they can prove a result. One weak answer sinks a buyer.
2Is your price on the right unit?The value you deliver versus the price you charge, and whether the thing you bill for matches the thing your buyer is judged on. Seats sold to a pipeline buyer is a mismatch at any price.
3Do you say what you actually do?What you built versus what your message sells. The strongest thing you own is often the thing your homepage never mentions.
4Does your copy sound like your buyer?Every load-bearing word, checked against how your buyer actually talks. "Replace your SDRs" scares the exact person who runs your evaluation.
5Who else is competing for the same buyer?How crowded your fight is: counting startups, the platforms your buyer already pays, and the ops team that could build it themselves.
The recurring failure patterns behind these questions are named, published, and free to run against your own copy: see the free eval.
How it works
Machine measures. Human judges. Nothing ships unsigned.
Evidence in
A structured intake built on artifacts, not opinions: your last three closing emails, real lost-deal reasons, live pricing. 60–90 minutes of your time.
You
The engine measures
Five areas, over 90 checks, scored against dated benchmarks: what your buyers can sign, what your category charges, how your buyers actually talk. Run five times; the median holds.
Machine + database
Charlotte judges
Every finding reviewed, every score challengeable, weak evidence flagged. What the machine can't do happens here: believing, doubting, spotting the non-obvious buyer.
Human
Verdict + moves
Which buyer, which price, which words: an alignment score, confidence tags on every answer, and your first three moves. Yours to act on, with or without us.
You again
Pricing
Four ways in. No retainer behind any of them.
The free eval
€0
instant · no email wall
For: any founder, any stage
- Your site, read by the instrument
- One named finding + your band
- Honest about what it can't see
The full report
€2,400
5 business days from completed intake
For: funded or revenue-stage teams
- The full diagnostic: five areas, evidenced.
- Buyer comparison + pricing headroom
- Language map with replacements
- Reviewed & signed by Charlotte
- 45-minute debrief call
The intensive
€12,000
2 weeks · 3 sessions
For: teams making the decision
- Everything in the full report
- Three sessions that test your evidence live: calls, lost deals, win rates.
- A decision you're confident in: which buyer, which price, which words
- Your positioning model, yours to keep: what every future check of your copy runs against
- Day-30 check-in
The subscription
€300
per month · after a report
For: teams that keep shipping
- Check any new asset against your positioning model
- Competitor drift alerts
- Quarterly model refresh
- Cancel anytime; the model stays yours
The report stands alone. No retainer behind it, nothing engineered to require the next purchase. Most clients take the report and act on it themselves. That's the point.
The posture
Honest instruments admit uncertainty.
Confidence tags on everything
HIGH means evidenced. LOW means the instrument is telling you it's guessing, and where more evidence would change the answer. Precision without honesty is theater.
Dated benchmarks
Every number cites its source and its date. When data goes stale, the reports say so on their own. This market moves monthly; a report that doesn't is fiction.
Who signs the verdict
Charlotte Han
Fifteen years positioning technology for enterprise buyers: from AI silicon to developer platforms to autonomous systems. The framework encodes that judgment; the review applies it to your evidence, one company at a time.
NVIDIA · SAP · BITRISE · FERNRIDE
HIGH · verifiable on LinkedIn
"The insider knows more about the company. The instrument knows what the company can't see about itself."
WHAT THE MACHINE CAN'T DO
Believe or doubt your evidenceHUMAN
Spot the buyer nobody pitchedHUMAN
Force the decision in the roomHUMAN
Sign the verdictHUMAN