Research Jtbd Interviews

---
name: research-jtbd-interviews
description: "The user wants to understand why people buy, or you're looking at a persona that smells made-up. Load this whenever someone asks for \"the audience,\" \"the avatar,\" or \"the customer.\""
metadata:
  author: wayland
  version: "1.0.0"
  category: "research"
---

# JTBD switch interviews

## When to load this mode

The user wants to understand why people buy, or you're looking at a persona that smells made-up. Load this whenever someone asks for "the audience," "the avatar," or "the customer."

## The interview, walked back from the purchase

Forget asking "what do you want." Buyers can't predict the future. They can narrate a story that already happened. You're a journalist with a timeline.

Find someone who bought (or switched away from) something like the user's product in the last 60–90 days. Memory past that decays. Walk backward through five moments:

1. **First thought** — "Take me back to when you first realized you needed a different way to handle this. What were you doing? What had just happened?" You're hunting for the **trigger event**, almost always something concrete (a meeting, a moment of frustration, a comment someone made), almost never an abstract goal.
2. **Passive looking** — "After that moment, did you start noticing solutions you hadn't noticed before? Where? When?" The shift from invisible to visible. People often start paying attention months before they start searching.
3. **Active looking** — "When did you actually start putting time into figuring this out? What did you do? Who did you ask?" You want the verbs. Searched, asked, downloaded, tried.
4. **Decision** — "Walk me through the day you bought. What was the very last thing that made you pull the trigger?" The last domino. Almost never the feature the marketer thinks it is.
5. **First use** — "What did you expect would happen? What actually happened?" The gap between expectation and reality is where retention lives or dies.

Ask one question at a time. Sit with silence. Echo their words back; never replace them with your own. If they say "I just snapped," don't write "frustrated." Write "snapped."

## What you extract — the four Forces of Progress

After the interview, pull out:

- **Push of the current situation** — the friction in the old way. What made staying intolerable.
- **Pull of the new solution** — the appeal of the new option. Specific, not generic.
- **Anxiety of the new** — what made them hesitant. Switching costs, fear of being a sucker, fear of looking foolish, fear of the new thing not working.
- **Habit of the old** — inertia. Sunk costs. "It's not great but I know it."

Progress happens when push + pull beats anxiety + habit. If anxiety is high, you don't sell harder — you reduce risk. If habit is high, you don't push the wow — you make switching small.

## Decision rules

- **Use this method when:** the team needs to know who the buyer is, why they'd switch, what they'd be switching from, and what's stopping them. Always before copy, pricing, or channel selection. After three or more interviews you have a pattern; after five you have a segment; one interview is an anecdote.
- **Don't use this method when:** the user already has 50+ recorded sales calls or support tickets — read those first, then interview to fill gaps. Or when the product hasn't been built and no one has switched to it yet (in which case interview switchers from a near-equivalent).
- **Skip the method entirely if:** the question is "what color should the button be." Wrong tool. Route to Brand.

## Anti-patterns

- **Don't ask "what do you want."** You'll get a feature list shaped by the last marketing email they read.
- **Don't ask hypotheticals.** "Would you pay $50 for this?" returns noise. "What did you pay for the last thing like this?" returns signal.
- **Don't lead.** "Was it frustrating?" becomes "yes." Ask "what was that like?" and let them name it.
- **Don't summarize for them mid-interview.** Mirror their words. Your summary contaminates the data.
- **Don't interview only happy customers.** Churned users and shoppers who bought a competitor are where the gold lives.

## Before / after

**Before (generic survey question):**

> "On a scale of 1–10, how important is saving time when choosing a meal kit?"

You'll get an 8 from everyone. Useless.

**After (switch-interview question):**

> "Take me back to the week you signed up. What was happening in your evenings before that?"

> *"My wife was on call three nights that week and I was reheating frozen stuff at 9pm after the kids went down. I caught myself eating standing up over the sink and thought, this is grim. The next morning I saw an ad for HelloFresh on my phone."*

Now you have a trigger (eating standing up), a Push (grim, late, alone), and the moment passive looking started (the ad). Five more interviews like that and you have a segment.