Clayton Christensen famously asked McDonald's why people buy milkshakes at 7 AM. The answer wasn't "because they're delicious." It was: "to kill thirty minutes of a boring commute, with one hand free for the steering wheel, and to not be hungry until lunch." That's a job. The milkshake was hired to do it. Same liquid, completely different framing — and a 7× improvement in product strategy.
1 job
Per customer segment
10
Power-user interviews
3 forces
Push · pull · habit · anxiety
7×
Strategy improvement
What Jobs-to-Be-Done actually means
JTBD is a research lens, not a methodology checkbox. It assumes that when a customer "buys" something, they're really hiring it to make progress on a specific job they need done. The job is functional, emotional, and social — all three at once.
"Customers don't want a quarter-inch drill. They want a quarter-inch hole. And really, they want the picture hung on the wall before their guests arrive."
The mistake most founders make: they describe their product in features instead of jobs. "We have a faster onboarding flow" is a feature. "We help operations leads ship the new hire's first task on day one, so they don't look bad to their boss" is a job.
The structure of a good job statement
| Component | What it captures | Example |
|---|---|---|
| Action verb | The work the customer is trying to do | Help me kill, save, send, prove… |
| Object | What the action is performed on | …thirty minutes, an investor's doubt, a forecast… |
| Context | The situation that makes the job urgent | …during a boring commute, before the board meeting… |
Bad: "Improve user experience for marketing teams."
Good: "Help a marketing manager prove campaign ROI to their CFO before the quarterly review."
The 4 forces of progress
Every job lives at the intersection of four forces. Get them on paper before you ship anything.
Push (the pain)
What is making the customer's current situation untenable?
Pull (the promise)
What better future is the new product offering them?
Habit (the friction)
What about the current way is hard to give up?
Anxiety (the doubt)
What will make adopting the new thing feel risky?
Most founders obsess over Push and Pull. The startups that win usually have lower Habit and Anxiety than the alternative — that's the unsexy work that decides who closes the deal.
The 5-step playbook
Step 1 · Recruit your power users
Not your average users. Not your top-of-funnel signups. Your top 10% by retention or revenue. They are the only customers whose job is real enough to learn from.
Step 2 · Run "buying timeline" interviews
Anchor on the moment they switched to you. Walk them backwards: "What was the first time you thought about this problem? What were you using before? What changed? What almost stopped you from switching?" Past behavior is the only honest signal.
Step 3 · Identify the fired product
Customers don't adopt — they switch. Whatever you replaced is a clue. If five interviews tell you they fired Excel, your real job is "make me feel less like I'm one merge away from a public mistake," not "improve productivity."
Step 4 · Write the job statement
One sentence. Action + object + context. If you can't fit it on one line, you don't have a job — you have several jobs that need to be split.
Step 5 · Validate against new prospects
Take the job statement to people who haven't bought from you. If they nod the moment you say it, you found the right job. If they say "interesting, I see what you're doing", you found a feature.
Run a JTBD-shaped analysis
Pin down your real job before the next product call.
NexTraction's project analysis breaks your idea into the JTBD framework automatically — pains, gains, jobs, and signals — and produces a structured memo your team can argue with. Stop guessing which job you're hired for; run the analysis.
Try the project analysis →The 3 most common JTBD mistakes
Confusing job and persona. The job isn't "the marketing manager." That's the persona. The job is what the marketing manager is trying to make happen.
Listing too many jobs. If you have eight jobs, you have one company per job. Pick one job per segment and double down.
Treating JTBD as a workshop, not a habit. JTBD is a continuous discipline. Re-run it every 6 months on your top users. Jobs migrate as customers grow.
FAQ
How does JTBD differ from "personas"?
Personas describe who. Jobs describe why. The same persona can have multiple jobs depending on the moment — and the job is what actually drives buying behavior.
Can JTBD apply to B2B?
Especially to B2B. The "social" dimension (avoiding embarrassment in front of the boss, looking competent in front of peers) often matters more than the functional dimension in B2B sales.
How many JTBD interviews do I need?
10 power users is the floor. After 10 you'll see the same job phrased three different ways — that's the signal you've found it.
Conclusion
Jobs-to-Be-Done is the lens that turns "we have a product" into "we know exactly what our customer is firing to hire us." Every roadmap call, every pricing decision, every onboarding email gets sharper once you have the job written down. Run a structured panel to find yours before you ship the next feature.
