| Symbol | Meaning |
|---|---|
| :) | Excited |
| :( | Angry |
| :| | Embarrassed |
| ⚡ | Pain or problem |
| 🥅 | Goal or job-to-be-done |
| ☐ | Obstacle |
| ⤴ | Workaround |
| ^ | Background/context |
| ☑ | Feature request / purchasing criteria |
| $ | Money / budget / purchasing process |
| @ | Mentioned specific person or company |
| ★ | Follow-up task |
Use these in real-time during conversations. Faster than writing words; preserves emotional valence.
Founder: Hi, I'm building a tool that helps marketing teams automate their reporting. Would that be useful to you?
Customer: Oh yeah, that sounds great!
Founder: How much would you pay for that?
Customer: I dunno, maybe $200/month?
Founder: Would you use it if I built it?
Customer: Definitely!
Founder: Awesome, I'll send you a link when it's ready.
[6 months later: customer ignores the launch email]
What went wrong: every question was a Mom Test violation. Hypothetical, opinion, future-tense, no commitment.
Founder: Hey, I noticed you're a marketing director. What's your week look like?
Customer: Honestly insane. Tuesdays I lose 4 hours just pulling reports.
Founder: 4 hours? Walk me through what that looks like.
Customer: I export from Google Analytics, then HubSpot, then Salesforce, mash them together in a spreadsheet, format for the board...
Founder: What happens if you don't get it done?
Customer: My CMO yells at me. Last quarter I missed it and almost got fired.
Founder: How are you handling it now?
Customer: I have a contractor who does it but he's expensive and slow.
Founder: How much do you pay him?
Customer: $1500 a month for like 8 hours.
Founder: Who else on your team has this problem?
Customer: My counterpart at [other company] is dealing with the same thing. Want me to introduce you?
Founder: That'd be amazing. Also - if I had something working in 2 weeks, would you be willing to try it for the next quarterly report?
Customer: Yeah honestly send it over, I'll pay for early access.
What went right: past behavior, real costs, current solution, intro committed, time + money commitment offered. This is signal.
Subject: 20-min chat about [their domain]?
Hey [Name],
I'm working on making [problem] easier for [target segment]. (VISION)
We're not selling anything yet - just trying to understand
if we're solving a real problem. (FRAMING)
I'm stuck on [specific thing you don't understand]. (WEAKNESS)
You've been doing [their relevant experience] longer than
just about anyone I could talk to. (PEDESTAL)
Got 20 minutes in the next two weeks for a casual chat? (ASK)
[Your name]
Subject: Thanks - and a small ask
Hey [Name],
Thanks for the chat yesterday. The thing you said about
[specific painful detail] is going to change how we
think about this.
Two quick things:
1. Would it be okay if I came back to you in 4 weeks
to show you what we built?
2. You mentioned [colleague's name] also runs into this -
would you be open to making an intro?
Either way, super grateful for the time.
[Your name]
Subject: Real version of [thing]?
Hey [Name],
We have a working version of [the thing we discussed].
Three options for what's next:
1. I send you a link, you try it on a real report,
and we talk about what works.
2. I do the first report for you so you can see it,
and your team uses it for the next one.
3. You skip it - no hard feelings, would just love to
know which option to pull off the table.
Let me know which one.
[Your name]
The third option matters: it makes "no" easy and reveals lack of commitment without the polite-yes trap.
When the conversation drifts into hypothetical territory, redirect:
| They Say | You Say |
|---|---|
| "I would always..." | "When was the last time? Walk me through it." |
| "Most people probably..." | "Tell me about you specifically." |
| "It would be cool if..." | "What's the worst part of how you do it now?" |
| "Sounds like a great idea." | "Thanks - how are you handling this today?" |
| "I'd definitely buy it." | "Could I send you an invoice today for early access?" |
| "Add feature X." | "Why? What would that let you do?" |
After each conversation, score yourself:
- Did I learn something I didn't already know?
- Did I get specific past behavior, not generalizations?
- Did I get a real commitment (time/reputation/money)?
- Did I keep my mouth mostly shut?
- Was I terrified of any answer I might hear?
- Did I avoid pitching?
5+ checks = good conversation. <3 = redo it.
Push the offer earlier than feels comfortable:
- Not "I'm thinking about building..."
- Instead: "Would you put down $500 today to be the first user?"
Their reaction is data. A flinch, a wandering excuse, or a "let me check with my team" all = no. A "yes, where do I send it?" = real lead.
This feels rude. It's not. It saves both of you from a 6-month dance ending in nothing.
Raw interview notes don't become product direction on their own. This protocol converts them into shared, tracked beliefs.
Step 1 - Write up within 24 hours. Each interviewer writes up their notes within 24 hours of the conversation. Include direct quotes, emotional signals (use the symbols from the note-taking table), and one sentence on what you personally believe changed or was confirmed.
Step 2 - Weekly team review. The team reviews all write-ups from the past week together. Each person calls out surprises, contradictions, or patterns. No decisions yet - just shared understanding.
Step 3 - Track belief changes on a shared doc. Maintain a living belief tracker. Before each batch of interviews, record what the team currently believes. After the batch, record what (if anything) changed and why. This makes drift visible and prevents the team from unconsciously moving the goalposts.
Step 4 - Only update product direction on convergence. A single interview, no matter how compelling, is not signal. Only update roadmap or positioning when multiple interviews, from different people with no reason to coordinate, point at the same problem or behavior. One data point is an anecdote. Three is a pattern.
Belief Tracker Template
| Belief | Evidence For | Evidence Against |
|---|---|---|
| [State the belief clearly] | [Quotes, behaviors, patterns that support it] | [Quotes, behaviors, patterns that contradict it] |