JTBD (Jobs To Be Done)

People don't buy a drill - they buy a hole in the wall. It sounds like a worn-out quote, but this is exactly what Jobs to Be Done is built on. A customer doesn't want your product for its own sake - they want to get a specific "job" done.

And the more precisely you understand that job, the more precisely you build the product, phrase the questions in your surveys, and interpret the answers. JTBD is not a framework full of matrices - it's a way of looking at customer behavior.

Jobs to Be Done (JTBD) - a product-research concept stating that a customer "hires" a product or service to get a specific "job" done - a functional, emotional, or social task. Instead of asking "who is our customer?", JTBD asks "what job is the customer trying to get done with our product, and in what context?"

The concept was formulated by Clayton Christensen. His classic example: McDonald's wanted to sell more milkshakes. They ran a standard survey - improved the flavor, changed the packaging, and nothing changed. Then they started asking people standing in line at 7 a.m.: why do they need a shake right now? Most bought it for a long commute to work, to keep from being bored behind the wheel. The shake was "hired" for a different job than the marketers thought. They made it thicker and added bits of fruit - and sales went up. That's JTBD in action.

The three types of "jobs"

Functional job - the practical task that needs to be done. "Get feedback from customers", "evaluate employees", "find the reason for churn". This is what the customer articulates on their own - usually through a verb: measure, find out, compare, speed up. The functional job is the visible part of the iceberg.

Emotional job - how the customer wants to feel in the process. Confident in the decision they've made. Protected from risks. Professional in the eyes of management. The emotional job is often more important than the functional one - it's what explains why a customer picks one product out of several that are equally functional.

Social job - how the customer wants to look in the eyes of others. "Show the team that they ran serious research", "report to the investor with data", "look like someone who makes decisions based on facts". To ignore the social context is to misunderstand half of the motivation.

All three types work at the same time. An HR manager launching an employee engagement survey functionally wants to get data (the functional job), to feel like a competent specialist (the emotional one), and to show the director that they work systematically rather than on intuition (the social one).

JTBD competitors: not the ones you think

One of the most useful JTBD insights is rethinking who your competitors are. In traditional marketing, a competitor is a company with a similar product. In JTBD, a competitor is anything the customer might "hire" to get the same job done.

For an online survey service, the competitors in the traditional sense are other platforms. In the JTBD sense, they're Excel with manual processing, Google Forms, a corporate NPS once a year through an outside contractor, and also "we're managing fine without surveys for now". These are completely different competitors with completely different switching barriers.

Understanding the real competitors changes the questions in your research. Not "why did you choose us over X?" - but "what did you do before, when you needed to solve this task?". The answers are often surprising.

How JTBD changes the questions in a survey

A standard survey asks: "How satisfied are you with the product?" or "What do you like and dislike?". The JTBD approach changes the angle: "What task were you solving when you first used us?", "What did you do before to solve this task?", "What was the last straw that pushed you to look for a new solution?"

Answers to questions like these give you a different quality of information. You learn not satisfaction with product features, but the context of the hire. This also changes your segmentation: two customers with identical demographics may hire the same product for completely different jobs - and they need different features, different messaging, different onboarding surveys.

For quantitative surveys, JTBD helps you build precise screening questions - filtering for exactly those who hired the product for the relevant job. Without this, you mix people with different goals into one sample and lose the signal in your data.

JTBD and CustDev: the difference and how to combine them

JTBD and CustDev are often confused or lumped together - they are different but highly compatible approaches. CustDev checks whether a problem exists and how painful it is. JTBD explains in what context and why a customer looks for a solution. Together they give the full picture: first you confirm that the pain exists, then you understand under what circumstances a customer "hires" something to solve it.

With personas, JTBD works as a refinement: a persona describes "who", JTBD describes "why". A persona described as "marketing manager, 30 years old, B2B sector" doesn't explain behavior. The JTBD job "convince management of the value of research with numbers in order to secure a budget for next quarter" explains it quite concretely.

How to run a JTBD study: the core structure

A JTBD interview is built around the moment of switching - when the customer decided to look for a new solution. Not "what do you like about the product", but "tell me how you ended up here in the first place". The basic sequence of questions:

1. The context of the hire. "Tell me what was going on during the period when you decided to try our product?" The goal is to understand the situation, not a characteristic.
2. The previous solution. "What were you doing before that to solve the same task?" JTBD competitors aren't only direct equivalents, but also Excel, notebooks, a colleague from the next department.
3. The switching trigger. "What exactly wasn't working for you? What was the last straw?" This is the most valuable part - the reason for switching is often not what you think.
4. The selection criteria. "By what signs did you evaluate the options?" This is where the functional, the emotional, and the social all surface.
5. The moment of decision. "What exactly convinced you to choose us specifically?" The gap between expectation and reality is gold for product decisions.

5-10 interviews like these reveal patterns. After that, the patterns are validated quantitatively: you build a survey with closed questions about the identified jobs on a large sample.

Example: an onboarding survey in JTBD logic

A classic onboarding survey: "How did you hear about us?", "What position do you hold?", "How many people are in your company?". All of this is demographics - it doesn't explain behavior. The JTBD version of the same survey: "What happened that made you decide to try us right now?", "What task do you want to solve first?", "How did you solve it before?"

Suppose that after 200 responses it turns out: 55% came with the job "collect feedback about the product quickly and without IT", 30% with "replace an inconvenient corporate tool", 15% with "launch NPS for the first time". These are three different segments with different needs, different pains during onboarding, and different churn triggers. Without JTBD questions they would all look the same: "companies that run surveys".

Common mistakes when applying JTBD

Confusing the job with a product task. "Create a survey" is not a JTBD job - it's an action inside the product. The job is "get data to make a decision within 2 days, before the presentation to the board of directors". The difference is fundamental: the job includes context and constraints, the action does not.

Stopping at the functional. Companies easily hear functional jobs and ignore the emotional and social ones. As a result, the product covers the function but loses to a competitor that better covers the feeling of confidence or status.

Doing JTBD through a closed-question form. You can't hand people a list of jobs to choose from and ask them to mark their own - you'll impose your own categories. JTBD questions for a quantitative survey are built only after the qualitative stage, once the jobs have been articulated in the customers' own words.

Not updating the jobs over time. JTBD is not a one-off exercise. The market changes, the target audience evolves, new switching triggers appear. Companies that ran a JTBD study three years ago and consider it still current risk building the product for outdated jobs.

JTBD in market research

In market research, JTBD is often used for market segmentation: instead of dividing by demographics or psychographics, you divide by jobs. Such segmentation is more stable, because jobs change more slowly than preferences or behavior.

A concrete scenario: a company enters a new market and wants to understand which segments exist. The traditional approach is a survey with questions about features, prices, channels. The JTBD approach is first 15-20 in-depth interviews with representatives of the market, identifying 3-5 key jobs, and then a quantitative survey to estimate the size of each job-segment. The result is not "the 25-34 segment with above-average income", but "companies that need to collect NPS quickly and without a research budget". The second description immediately suggests what to offer, how to communicate, and through which channels to look.

For more on applying qualitative methods at the start of a study, see our blog article on JTBD research.

How to collect data for JTBD in SurveyNinja

In-depth JTBD interviews can't be automated - they're conducted in person or over video call. But SurveyNinja fits well into the quantitative stage: once the jobs have been identified and you need to validate them on a large sample.

Use open-ended questions at the start of the survey - this gives you the customers' own wording for subsequent coding and for building closed answer options. Logic jumps let you guide the user through different scenarios depending on the job they named. Answers in the reports section are easy to group by text responses and export for JTBD clustering.

JTBD flips the logic of research: instead of "who is our customer" - "why did the customer hire us". When you know the job precisely, you ask the right questions, build the features you need, and stop being surprised by why customers use the product differently than you planned.