How to Run Jobs-to-Be-Done Interviews

Jobs-to-be-done interviews reconstruct the exact timeline of a real purchase or switch, so you learn the underlying job a customer hired your product to do rather than the features they say they want. You run one by recruiting a recent switcher, walking them backward through the moment they decided to change, and mapping the four forces that pushed and pulled them across the line.

This is a method guide, not a tools roundup. If you want software to record, transcribe, and analyze these sessions, see our roundup of the best platform for JTBD customer interview studies. Here we focus on how to actually run the interview and turn it into product decisions.

What a jobs-to-be-done interview is

The jobs-to-be-done framework comes from Clayton Christensen and his collaborators, popularized in the Harvard Business Review article Know Your Customers’ Jobs to Be Done. The core idea: people do not buy products, they hire them to make progress in a specific circumstance. The milkshake, the mattress, the project management tool. Each one is hired to do a job.

A JTBD interview, sometimes called a Switch interview or a timeline interview, is the primary research method for uncovering those jobs. Instead of asking customers what they want, you rebuild the story of a decision they already made. Behavior that already happened is far more reliable than predictions about future behavior, which is why this technique produces sharper insight than most abstract user interviews.

The distinction matters. A standard user interview might ask, “What features would make this better?” A JTBD interview asks, “Walk me back to the day you realized your old approach was not working. What happened?” One collects opinions. The other collects evidence.

For a broader view of where this sits among discovery methods, see our complete walkthrough to product research methods.

The four forces of progress

Every switch is a tug of war between four forces. Two push the customer toward a new solution, and two hold them back. Demand only exists when the forces of progress outweigh the forces of inertia.

Force	What it is	What it sounds like
Push	Frustration or a problem with the current situation	”Our spreadsheet finally broke when the team hit 20 people.”
Pull	The attraction of a new solution and the future it promises	”I saw a demo and thought, that is exactly how I want this to work.”
Anxiety	Fear and uncertainty about the new solution	”I was worried it would take weeks to migrate and my team would revolt.”
Habit	Attachment and comfort with the old way	”Honestly the old tool was annoying, but everyone already knew it.”

Your job in the interview is to make all four forces visible. Push and pull explain why someone moved. Anxiety and habit explain why they waited, and why so many prospects stall out before ever reaching you. Reducing anxiety is often the single highest-leverage change a product or growth team can make, because the demand is already there.

Step 1: Find and recruit recent switchers

The quality of a JTBD study is decided before the first question. You need recent switchers: people who bought, adopted, or changed solutions in roughly the last 60 to 90 days. Recent memory is detailed and honest. Older recall gets smoothed into a tidy story that never happened.

For B2B products, this is where most studies stall. The people you want, the decision-makers and hands-on users who just went through a switch, are hard to reach and rarely sitting in a generic panel. Recruiting the wrong people, or waiting weeks for the right ones, quietly kills more research than bad interviewing does.

This is exactly the gap CleverX was built to close. It is a B2B research platform with a panel of more than 8 million professionals verified by work email and LinkedIn, across 150-plus countries, with participants typically delivered in about 2 to 5 days. You can screen for people who changed vendors in the last quarter and hold a real conversation with them instead of guessing.

For recruiting tactics by audience, these help:

Aim for 8 to 12 interviews per segment. You are hunting for repeated forces and moments, not statistical significance. When new conversations stop surprising you, you have reached saturation.

Step 2: Set up the timeline interview

A JTBD interview is a documentary, not a survey. You are directing the customer to replay a movie of their decision in slow motion. Keep the session to 45 to 60 minutes, record it with consent, and resist the urge to talk.

Anchor the entire conversation to one concrete purchase or switch. Not “how do you think about tools like this” but “the specific time you bought this one.” Every question loops back to that event and its timeline.

Four moments structure the story:

First thought. The very first time the customer sensed the old situation was not good enough. Often months before any purchase.
Passive looking. They notice alternatives but take no real action.
Active looking. A trigger flips them into searching, comparing, and asking around.
The decision. The final push over the line, and the anxieties they had to overcome to sign.

Your questions walk the customer through these four moments in order, digging for the emotional and situational detail at each one.

Step 3: The question sequence

Start at the purchase and work backward. The most powerful opening question in JTBD research is some version of: “Take me back to when you first bought this. When did you first realize you needed something different?”

From there, follow the timeline. Use questions that ask for events, not opinions:

“What was going on in your work that week? Set the scene for me.”
“What was the first thing that made you think the old way was not working?”
“When you started looking, where did you look first? Who did you talk to?”
“Was there a moment you almost did nothing? What held you back?”
“What finally made you decide to commit?”
“When you signed up, what were you nervous about?”

Notice how each one points at a specific past moment. That is the discipline of the method. The instant a customer starts speaking in generalities (“I usually like tools that are simple”), gently steer back: “Was there a specific time that came up?”

Silence is your best tool. After an answer, wait. The second and third sentences, the ones that come after the pause, are where the real forces live. For a deeper bank of prompts, borrow from our 50 user interview questions that uncover real insights and this collection of qualitative research questions.

JTBD interviewing sits close to two adjacent techniques worth studying if you want to sharpen it: churn interviews and the work of identifying switching triggers through customer interviews. Same backward-timeline logic, different moment in the customer lifecycle.

Step 4: Common mistakes to avoid

The method is simple to describe and easy to botch. The usual failures:

Asking about the future. “Would you use a feature that did X?” People are terrible at predicting their own behavior. Stay in the past, where behavior is real.

Accepting the first answer. The first reason a customer gives is rarely the real one. “It was cheaper” often hides “my old vendor made me feel stupid on a support call.” Keep pulling the thread.

Leading the witness. If you ask, “So the integration was the deciding factor?” you will get a yes and learn nothing. Ask what the deciding factor was and let them name it.

Interviewing non-switchers. Talking to happy long-term users about a purchase they made three years ago produces fiction. Recent memory only.

Skipping anxiety and habit. Teams love push and pull because they flatter the product. But the forces of inertia are where the growth blockers hide. Chase them just as hard.

For a fuller list, our guide to common user interview mistakes that ruin your research applies directly here.

Step 5: Synthesize jobs into product decisions

Interviews are worthless until they change a decision. After your 8 to 12 conversations, work through synthesis in four passes.

Write the timelines. For each interview, lay out the four moments and note what happened at each. This forces you to separate the story from your interpretation of it.

Tag the forces. Go through every transcript and label each quote as push, pull, anxiety, or habit. Patterns emerge fast when you see the same anxiety appear in eight of ten interviews. Our guide to analyzing user interview data covers the mechanics.

Write job statements. Distill each recurring job into a single sentence in the form: “When [situation], I want to [motivation], so I can [expected outcome].” For example: “When my team outgrows a shared spreadsheet, I want to move to a real system without a painful migration, so I can scale without losing everyone’s trust.”

Map jobs to your product. Score which jobs your product serves well and which it serves poorly. Then prioritize the jobs with strong push and pull but high anxiety. The demand already exists there; you just need to lower the barrier through onboarding, messaging, or a targeted feature.

Those job statements feed directly into positioning and pricing work. They also sharpen your buyer personas and give you the customer language to run a value proposition test with buyers.

Making it repeatable

One study of switchers is a snapshot. The teams that compound their advantage run these conversations continuously, so that every new segment, launch, and churn spike feeds fresh jobs into the roadmap. Our guide to building a continuous user interview program walks through the operating model.

The bottleneck is almost never willingness to do the work. It is getting the right recent switchers in front of you before the memory fades. CleverX handles that side: you post a screener for verified professionals who match your segment, run moderated sessions yourself, or use AI Interview Agents to run structured JTBD conversations at scale, and pay for participants with pay-as-you-go credits rather than a platform subscription. When you can reach the right people in days, JTBD interviewing stops being a special project and becomes a habit.

When you are ready to talk to real switchers, start recruiting verified participants on CleverX.

Frequently asked questions

What is a jobs-to-be-done interview?

A jobs-to-be-done interview is a structured conversation that reconstructs the timeline of a real purchase or switch, from the first thought that something needed to change through to the moment the customer chose a new solution. The goal is to uncover the underlying job the customer hired the product to do, not their stated feature preferences.

How many jobs-to-be-done interviews do I need?

Most teams reach useful patterns after 8 to 12 interviews with recent switchers in the same segment. You are looking for repeated forces and moments, not statistical significance. If new interviews stop surprising you, you have likely reached saturation for that segment.

Who should I recruit for JTBD interviews?

Recruit recent switchers: people who bought, adopted, or changed solutions in the last 60 to 90 days. Recent memory is far more accurate than older recall, and the switch itself contains the forces you want to study. For B2B products, target the actual decision-makers and users involved in the change.

What are the four forces of progress?

The four forces are push (frustration with the current situation), pull (attraction of the new solution), anxiety (fear of the new), and habit (attachment to the old). Push and pull drive a customer toward change, while anxiety and habit hold them back. Demand exists only when push and pull outweigh anxiety and habit.

How is a JTBD interview different from a normal user interview?

A normal user interview often asks about needs, opinions, and desired features in the abstract. A JTBD interview anchors every question to a specific past event and rebuilds the timeline of that decision. It studies behavior that already happened rather than predictions about future behavior, which makes the findings far more reliable.

How do I turn JTBD interviews into product decisions?

Write each job as a job statement, group the forces that recurred across interviews, and map which jobs your product serves well and poorly. Prioritize the jobs with strong push and pull but high anxiety, because reducing that anxiety through onboarding, messaging, or features is usually the fastest path to growth.