Applying the Mom Test when you don't have a research team

Rob Fitzpatrick's The Mom Test comes down to one rule: don't ask people what they think of your idea. Ask about their life, their problems, and their past behavior, because those are things they can't accidentally lie to you about, even when they're trying to be nice.

It's a great rule, and it's hard to apply when you're both the person emotionally invested in the idea and the person writing the questions. Below are the traps that catch solo founders and the rewrites that fix most of them.

Why the Mom Test exists

People are kind. When you ask them to evaluate your idea, most respondents imagine a helpful future version of themselves and tell you what that person would do. That person doesn't exist and will never sign up.

The Mom Test points every question at facts. Instead of "would you use a tool that does X," you ask "tell me about the last time you needed to do X." The first invites a flattering guess, while the second pulls up a memory.

The trap every solo founder falls into

You write a question like "Would you find a tool that does X useful?" and it feels neutral. It asks the respondent to picture a hypothetical and be helpful about it, which they will, because most people are polite.

The version that tests something is "Tell me about the last time you tried to do X. What did you use? What happened?" The wording barely changes, and the quality of what you hear back changes enormously.

The rewrites that fix most bad questions

Opinion to behavior

• "Would you pay for this?" becomes "What are you using today, and what does it cost you in money or time?"
• "Do you think this is a good idea?" becomes "Walk me through the last time this problem came up for you."
• "Would you switch from your current tool?" becomes "What would have to happen for you to switch away from what you use now?"

Future to past

• "How often would you use this?" becomes "How many times did this come up in the last month?"
• "Would this save you time?" becomes "How long did it take you the last time you did this manually?"

Leading to open

• "Is late payment a big problem for you?" becomes "What's the most annoying part of getting paid?"
• "Don't you hate doing X in a spreadsheet?" becomes "How do you handle X today?"

Every rewrite takes your idea out of the question entirely. Once the respondent can tell what you want to hear, the answer is worth less.

The follow-up matters as much as the question

Mom Test interviews live and die on follow-ups. "That's interesting, why did that happen?" gets you further than any cleverly worded opening. The opening question starts the story, and most of what you learn comes in the follow-ups.

These three almost always earn their place:

• "Why was that a problem?"
• "What did you do about it?"
• "What happened next?"

If you're running an asynchronous survey instead of a live call, adaptive follow-ups that react to a specific answer do the same job at scale.

Why this is harder alone

A research team catches bad questions in review before they go out. A solo founder writing their own screener is grading their own homework, and motivated reasoning creeps in right where it does the most damage: the question that would confirm what you want to be true.

Some outside structure that enforces Mom Test discipline before questions go out will do more than willpower. That can be a co-founder, a research-minded friend, or a tool that rewrites leading questions for you. Whatever you use, the check has to happen before the responses come in.

Common questions

Does the Mom Test work for B2B as well as consumer?

Yes, and arguably better. B2B buyers can usually point to concrete past spend, existing tools, and a budget line, which are exactly the facts the Mom Test is designed to bring out.

How many Mom Test interviews do I need?

The method is about question quality more than sample size. Ten well-run conversations with the right segment tell you more than fifty leading ones. See how many customer interviews you need.

Can I run the Mom Test as a survey instead of live calls?

Partly. You lose the improvised follow-up, so you have to build likely follow-ups into the survey branching in advance, or use a tool that generates them as people answer. The core rule, asking about behavior instead of opinion, carries over unchanged.

What if someone volunteers that they love the idea?

Thank them, then set it aside and go back to behavior: "That's kind of you. When did you last deal with this problem yourself?" Enthusiasm isn't evidence.

Mom Test discipline, built in

Every Hypothis research round uses Mom Test discipline by default. Screeners filter out the wrong respondents, questions ask about behavior, and adaptive follow-ups dig deeper so you don't have to catch every leading question yourself. It's free during early access. See how it works, or read how to write a startup hypothesis first.