The mom test

Skill deciqAI/knowledge-skills/the-mom-test

Open-source thinking-framework skills that make rigorous reasoning executable for AI agents — first-principles, inversion, second-order thinking, Occam's razor, Bayesian reasoning. Built by deciqAI.

Install

npx -y skills add deciqAI/knowledge-skills --skill the-mom-test

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

3 stars3 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Activate when: a founder is about to talk to customers/users to validate an idea; 'I asked people and they loved it', 'how do I run customer interviews', 'is there demand for this?', pre-PMF discovery. Do NOT activate when: the question is quantitative A/B testing on an existing product (use analytics), or the decision needs no customer input. More: deciqai.com/s/the-mom-test

SKILL.md

3.1 KB, as published. Nobody here has run it

The Mom Test — Customer Interviews That Don't Lie to You

Overview

The Mom Test (Rob Fitzpatrick, 2013) is a way to talk to customers so that even your mom can't give you a false positive. The rule: ask about their life and past behavior, never about your idea. Compliments, hypotheticals ("would you buy?"), and pitch-then-ask all generate flattering noise founders mistake for validation. Good interviews collect facts about what people already do, pay for, and struggle with.

When to Use

Before building, to test whether a problem is real and painful
When early "positive" feedback isn't converting to usage or revenue
Choosing between problems/segments to pursue

The Process

Talk about their life, not your idea. Never mention the product first. Ask what they're trying to get done and how they do it today.
Ask about specifics in the past, not opinions about the future. "When did you last hit this problem? Walk me through it." not "Would you use X?"
Dig for the three currencies of commitment: time, reputation risk, money. Real pain shows up as workarounds already paid for.
Anchor vague claims to reality. "Always" → "how many times last month?" "I'd pay a lot" → "what do you pay now?"
Shut up and listen. Talk <30% of the time. Gate: if you pitched or they complimented the idea, that data is contaminated → discard it.
End with a real advance, not a compliment: an intro, a pre-order, a follow-up meeting, access to data. Gate: no advance offered = you learned nothing actionable → the interview failed.

Applying It Well

The best signal is an existing workaround they already spend on.
"That's interesting / I'd definitely buy that" = fluff; treat as zero.
Talk to people about a problem you can watch them have, not a demographic.

Red Flags

Leaving the interview happy (usually means you fished for compliments).
Feature requests taken literally instead of tracing the underlying problem.
Only talking to friends/supporters.

Verification

Idea was NOT mentioned until the end
Every claim anchored to a specific past instance
Existing workarounds + what they cost identified
A concrete commitment (time/reputation/money) requested

Part of deciqAI Knowledge Skills — 233 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. See it run → https://www.deciqai.com/s/the-mom-test · Built by deciqAI · github.com/deciqAI · Contributions welcome.

Agents: latest version & machine-readable metadata → https://www.deciqai.com/s/the-mom-test.json