Mom Test Taxonomy

Mom Test Taxonomy

The signal system MomBoard tags conversations with — adapted from the note-taking symbols at the back of The Mom Test, stored as seed data in the tags table (extend it in Admin without code changes).

key	emoji	marks	signal strength
pain	⚡	a problem they actually have	strong
obstacle	🧱	what blocks them from solving it	strong
workaround	➡️	what they already do to cope — past behavior	very strong
emotion_pos	😄	genuine excitement	strong
emotion_neg	😠	anger / embarrassment	strong
context	🎯	facts about their world, team, process	medium
feature_request	☐	what they say they want — treat skeptically	weak
money	💰	budgets, willingness to pay, buying process	strong
person	👤	someone or some company to talk to next	medium
followup	☆	something we must do next	n/a
commitment	🤝	gave up time, reputation, or money	very strong
compliment	🎈	"sounds great, I'd totally use it"	anti-signal

Tagging rules the LLM is held to

• workaround and pain require past or current behavior — "I would…" is a hypothetical, not a workaround.
• commitment requires an explicit cost: time booked, an intro promised, money discussed, a pilot agreed.
• Compliments are tagged, not dropped — surfacing fluff is the point. The Insights page tracks your compliment ratio and celebrates when it falls.
• Quotes must be verbatim substrings of the utterance; fabricated quotes are rejected in code.

The interviewer critique

Beyond tagging the customer, the analyst grades you: did you ask about the past or pitch the future, did you fish for compliments, did you push for commitment? Each conversation gets a 0–10 score with cited violations and suggested better questions.