The famous name is often the wrong one.
Bring one live decision. We find the mechanism that actually applies, grade the evidence honestly, and run a test that could prove the idea wrong.
The costume name
“The café queue is the 25% tipping point”
The actual mechanism
Information cascades
I have a decision to make
Tell us about one live decision. A person reads every intake and replies within 48 hours.
Browse the library
Correct names, honest evidence, ideas from other fields. Free, no sign-up.
See what we tested
Every result, including the ones that failed.
Costume vs correct name
“The café has a queue, so it must have hit the 25% tipping point.”
The famous name goes in the title. The correct name goes in the answer.
Read page 2Guess first.
Before you read the answer, pick the mechanism. Quick, accurate feedback is exactly what a founder's week lacks.
Case question
“Customers keep telling you 'we love your product', but almost nobody pays. Which mechanism is most likely at work?”
Your answers stay in your browser.
How a Decision Review works
Intake — Seven questions and two numbers.
Seven questions about one live decision, plus two numbers: monthly users at that step and the date the decision is due. A person reads every intake and replies within 48 hours.
Refusal screen: fake reviews, false urgency, implied official status are declined.
About six to eight weeks from intake to readout (our estimate; it depends on when you run the test).
See prices and detailsCan your traffic detect a realistic effect?
Relative lift required: +18.1% from baseline (5% → 5.91%).
For reference: about 1.4 percentage points, the average effect of government nudges at scale (DellaVigna & Linos 2022: 126 trials by two US government nudge units, against 8.7 points in published papers). Using it for a product decision is our extrapolation, not their finding.
A full test can likely detect a realistic effect.
From the library
Pages 1–4 are drafts awaiting review. The rest are planned.
Is the white coat a halo effect? Authority cues and official signals
The café queue is not the 25% tipping point
Five models and a judge: an ensemble, not Bayes
Weekly vision recitals and the build trap
Wicked learning environments: why a founder's experience teaches the wrong lesson
Default effects: the published size and the size at scale
Survivorship vs selection bias: Wald without the fable
Oreo didn't win on its name
The Registry, honestly
Every Decision Review ends in an outcome on the public record: survived, killed, inconclusive or abandoned. Results are published in wording the client approves; Private Reviews appear only in the totals. A killed idea is never quietly dropped.
Not enough Reviews yet. The first results appear after the pilot Reviews.
Who it is for
- Founders, growth and product leads at funded Indian startups (seed to Series A; D2C, fintech, edtech, consumer apps, marketplaces) with live users and one decision due within weeks.
- Pre-launch teams: a priors memo.
- Accelerators and VC platform teams: cohort Reviews, from month 7.
- Readers: the library is free.
Who it is not for
- Deception: fake reviews, seeded queues, false urgency or implied official status.
- Generic frameworks: SWOT, templates, canvases or “25 biases” workshops.
- Treatment: for addiction or mental health.
- Hidden tests: on exams, hiring, credit or health without telling the people tested.
Have a pricing page, an onboarding default or a renewal reminder to decide this month?
A person reads every intake. No sales calls you didn't ask for.