# Jobs-to-Be-Done Interview Questions: Find the Work Behind a Feature Request

Jobs-to-be-Done interview questions focused on real work, with a feature-request example, source evidence and a counterexample.

- Canonical URL: https://www.harmate.com/en/blog/jobs-to-be-done-interview-questions-feature-request
- Author: Harmate Team
- Published: 2026-10-25
- Updated: 2026-10-25T22:47:12.166083+00:00
- Language: en

## Content

# Jobs-to-Be-Done Interview Questions: Find the Work Behind a Feature Request

A Jobs-to-Be-Done interview can help you understand what someone is trying to accomplish beyond the product they use. When a customer says “We need an export,” the task is not to immediately turn that request into a universal need. Reconstruct an episode: what they were doing, the steps involved, the intended result and how they judged that result.

This guide focuses on work being performed. Its output is a revisable description of the job, connected to source excerpts and a case that may challenge it. It does not claim to reveal hidden motives or predict whether an innovation will succeed.

## Be explicit about the kind of JTBD interview

Jobs to Be Done covers different practices. A switching interview explores movement from an old way of doing things to a new one. An interview focused on a functional job examines what a person is trying to accomplish independently of a particular tool.

Professional accounts make this distinction visible. [Dscout describes an approach centered on switching and desired progress](https://dscout.com/people-nerds/the-jobs-to-be-done-interviewing-style-understanding-who-users-are-trying-to-become). [Mike Boysen, an Outcome-Driven Innovation practitioner, distinguishes the functional job from the buying journey](https://jobs-to-be-done.com/jobs-to-be-done-interviews-79623d99b3e5). These are practitioner frameworks, not evidence of one method validated for every situation.

Here, the focus is the work performed. The guide is not a complete ODI protocol. If your question is primarily about vendor selection, design research around the purchasing decision rather than combining both goals in one interview.

## Recruit the person who performs the work

Describe the activity, even provisionally: preparing a committee briefing, checking a delivery or organizing feedback from training. Speak to someone who has actually done it and choose an episode they can describe in detail.

The buyer, manager and end user may all have useful perspectives, but they do not know the same steps. Ask who does what. When a participant delegates part of the work, mark that boundary instead of asking them to invent an account of what colleagues probably do.

Avoid recruiting only people who have mastered your solution. Depending on the research question, people using another method or combining tools may reveal steps your product does not support. That does not automatically create a representative sample. Preserve a description of the situations actually included.

## Start with the last episode and follow the work

Use a flexible guide. This [semi-structured interview guide](https://www.harmate.com/en/blog/semi-structured-interview-guide-method-steps-and-15-examples) explains how to prepare themes without reciting a rigid questionnaire. The prompts below are proposed for this situation, not offered as a standardized instrument.

**The episode.** “Tell me about the last time you prepared that briefing. What was expected, and by whom?” Look for a specific situation and recipient without assuming that the need begins when your software opens.

**The starting point.** “What did you already have? What else did you need before starting?” This may reveal collection, approval or document reconciliation that happens before product use.

**The sequence.** “What did you do next?” At significant steps, ask: “What were you trying to achieve at that point?” Distinguish an action, such as copying rows, from its purpose, such as preparing a comparison the committee can understand.

**Handoffs between people and tools.** “What did you pass on, to whom, and what did you expect back?” An offline step is still part of the work. Do not remove it from the account because your offer does not cover it.

**The result.** “When did you consider the work finished? What showed you that?” The answer may concern a delivered document, approval or a decision made possible. Preserve the participant’s own criteria before rewriting them.

**Difficulties and variations.** “What needed reworking in this episode? Does it sometimes happen differently?” Allow for an uneventful attempt and seek a different situation. Avoid “How much time do you waste because of this tool?” which presumes both waste and its cause.

[AAPOR’s best practices](https://aapor.org/standards-and-ethics/best-practices/) recommend clear questions, one concept at a time and pretesting with people similar to the intended participants. In particular, check what words such as “briefing,” “result” and “finished” mean to them.

## A worked example: the export is not the whole job

This example is fictional. A training manager asks to export responses into a spreadsheet. The team immediately considers building another button.

During the interview, the manager explains:

> “Before the committee meeting, I bring together feedback from three sessions. I need to explain why I’m proposing a change to an exercise, and find the comments again if someone asks what supports it.”

The account reveals work beyond exporting, without making the export unnecessary.

| Level | Working description | Evidence in the episode |
|---|---|---|
| Requested feature | Export responses into a spreadsheet | Initial request |
| Possible broader job | Prepare an evidence-supported training adjustment proposal for the committee | Bringing three sessions together and preparing the proposal |
| Intended result | Present the proposed change and retrieve the comments supporting it | Passage about justification and source comments |
| Open question | Does the committee require a spreadsheet, or is it the manager’s personal working choice? | Not established in the account |

Next, check that understanding with the participant and investigate comparable episodes. It would be premature to conclude that all trainers want a decision record or that exporting no longer matters.

## Keep a counterexample in the job description

In a second fictional example, another person requests the same export solely to feed an internal system. They are not preparing for a committee and do not need to argue for a training change.

The requested feature is identical; the described work differs. Do not merge both accounts into a statement so broad that it can no longer guide a choice. Keep two hypotheses or specify the conditions under which each applies.

A provisional statement might be: “Prepare a training adjustment proposal whose supporting evidence remains retrievable during the committee meeting.” Add the known boundaries: type of meeting, the person’s role and available sources. This statement is a summary to challenge, not a psychological truth about the participant.

An [evidence-based buyer persona](https://www.harmate.com/en/blog/buyer-persona-build-it-from-evidence-not-assumptions) addresses another need: organizing relevant contexts and groups. A job description does not replace that information, and a shared job title does not establish that people perform the same work.

## Turn the account into a research decision

The final record should preserve the episode, steps, intended result, stated criteria, proposed job description and disagreements. Do not add an opportunity score based only on a few interviews. They can generate a hypothesis; they do not by themselves measure market size or rank every customer need.

[Open-question analysis in Harmate](https://www.harmate.com/en/product/open-questions) can help examine themes and findings linked to source excerpts, then select evidence for a decision record. Defining the job and choosing the next check remain human judgments.

Before turning the account into a feature, write down the unresolved question. Do customers need easier handoffs, clearer justification, retrievable evidence or several separate steps? The interview becomes useful when it makes that decision more precise, even if the conclusion is to keep the current solution.