Jobs to Be Done interviews: how to analyse the events behind a choice
Use the transcript to reconstruct a Jobs to Be Done interview: situation, first trigger, search, doubts, choice and outcome. Includes a worked example and table.
Start with the sequence of real events
“I need a convenient service” is too broad. It tells you nothing about what happened before the search, what alternatives the person considered or why the decision became urgent. Reconstruct a recent case from the first difficulty to the outcome.
A transcript preserves the order and the precise conditions. One story doesn’t establish a rule about the market. Conclusions about a segment require several interviews and a look at cases that contradict them.
Analyse what happened at each step
Choose one decision
Ask the interviewee to recall the most recent specific case. Clarify the time, context and task. Separate different purchases if they come up in the same conversation: each may have its own trigger and criteria.
Reconstruct the old method
Record how the person dealt with the task before looking for a new option, what worked and what event changed the situation. Don’t label their experience a “pain point” before you understand its practical consequence.
Record alternatives and doubts
Note what they compared, what risks they saw and whose opinion mattered. A competitor need not be named for there to be alternatives. Doing it by hand or doing nothing may have been options.
State the job and its limits
Describe the progress the person wanted in that situation. Keep the conditions under which the story applies and the facts you still lack. Compare your wording with other interviews.
A fictional story about preparing a report
This is a fictional training example. The names, circumstances and outcomes are included solely to explain the method.
I used to listen to the calls again in the evening. After meeting three clients in a row, I realised I wouldn’t have the reports ready by morning. I compared typing them up myself with automatic transcription, but I was worried about losing names and amounts.
The trigger was three meetings and a morning deadline. It wasn’t a general desire to save time. The old method worked when the workload was lighter, so the case doesn’t support “manual typing is always bad.”
The criteria were preparation speed and the ability to check names and numbers. A practical output should therefore keep timestamps and allow important passages to be checked separately, without promising a flawless transcript.
| Excerpt | What to record | What not to infer |
|---|---|---|
| Situation | Context and time limit | That the problem is universal |
| Trigger | What changed the old routine | That the person had long wanted to buy |
| Comparison | Options and criteria | All options on the market |
| Outcome | What changed after the choice | Long-term effect |
Decision story worksheet
Fill it in after reviewing the full sequence. One memorable quotation is not enough.
Participant / date: Specific situation: Old method: Triggering event: How the search began: Options considered: Decision criteria: Doubts and risks: Who influenced the decision: What was chosen and why: Outcome after the choice: Quotes and timestamps: What needs clarification: Limits of the conclusion:
Check the result before using it
- There’s one concrete episode, not a mixture of stories.
- The trigger is separate from general dissatisfaction.
- Alternatives include the old method and doing nothing.
- The conclusion stays within the verified events.
A closer look
Should I use “Jobs to Be Done” in the questions?
You don’t have to. Plain questions about the last case, the old method, the search and the choice usually make the interview clearer.
How many interviews are enough?
There’s no universal number. Don’t stop just because several stories resemble each other. Look for new patterns and contradictions until you have a reason to end the research.