JTBD and the Value Proposition Canvas

JTBD (Jobs To Be Done) is a theory that a customer “hires” a product to do a job. The Value Proposition Canvas sets jobs, pains and gains against the offer.

Not “what we do” but “why we are hired”

The earlier lessons were about how to write. This one is about what to write about. Before deciding how to tell the story of a project, it helps to ask: what does the customer actually want to do when they come to us? The answer is often not the obvious one.

A classic example is attributed to the marketing professor Theodore Levitt, and has been retold for decades: people don’t want a quarter-inch drill, they want a quarter-inch hole. The drill is only a means. Clayton Christensen, a professor at Harvard Business School and author of The Innovator’s Dilemma, developed the same thought, strictly and as a theory, in the early 2000s. The theory is called JTBD (Jobs To Be Done).

JTBD: the customer “hires” a service

Where from. It’s the idea of several authors, and it matters to know that, because they differ on details.

• Clayton Christensen (1952 to 2020) set out the language of “hiring a product to do a job” in The Innovator’s Solution (2003). His story of the milkshakes at a fast-food chain is well known: the team asked what a milkshake should be like, and when they began to ask why it was bought in the morning, it turned out that many people “hired” a milkshake to make a long, dull drive to work less tedious. The theory is summed up in the book Competing Against Luck (2016) and in an article in the Harvard Business Review, which is in the sources. (Tony Ulwick, below, argues that the milkshake story misses the point of the theory, and it’s worth hearing both.)
• Tony Ulwick (Strategyn) has worked since the 1990s on his own approach, “outcome-driven innovation”: he measures how well people achieve the outcome they want and looks for places where they fall short.
• Bob Moesta and Chris Spiek developed the “switch interview”: they ask people how they moved from one solution to another, and describe it through four forces.

They all agree on the main thing. A job isn’t the customer’s job title and not a property of your product. It’s the progress a person wants to make in a particular situation. The same person in different situations hires different things: in the morning a milkshake for the drive, in the afternoon one for the kids. That’s why, for describing a customer, age and gender work worse than the situation.

Types of jobs. Christensen distinguishes functional jobs (what to do: confirm a boundary), emotional ones (how to feel while doing it: stop worrying) and social ones (how to look in others’ eyes: not fall out with the neighbor). All three are present in any job, and a site that speaks only of the functional one misses two thirds.

Four forces. The decision to switch to something new is made of four forces. Push: what is wrong with the present one. Pull: what attracts in the new one. Anxiety: what the person fears about the new one. Habit: what holds them to the old one. The first two move people toward you, the other two hold them back. Unexpectedly but truly, most sites boost the pull and ignore anxiety and habit, though those most often stop people.

The job story. The “when…, I want to…, so that…” format appeared at the company Intercom around 2013, and Alan Klement described and spread it in detail. It replaces the “user story” (“as a customer, I want…”), which has a role but no situation. A job story starts with a situation, then comes the goal and the result. The main rule: there’s no solution in the wish. “I want a booking button” is a bad story. “I want to find out whether I can get my bike back today” is a good one: the solution is left to you.

An example with the default persona. “When my neighbor moved the fence and all I have is old documents, I want to find out where the boundary really runs, so that I can decide whether to go to my neighbor or to a specialist.” The functional job: find out the boundary. The emotional one: not feel cheated. The social one: keep things good with the neighbor, or at least not be the one at fault.

When it works. For positioning, for deciding what goes on the home page and which sections are needed, and for texts where you must show that you understand the customer’s situation.

Where it lets you down.

• A job is not made up at a desk. Real jobs are found out in conversations: how the person got to you, what they tried, what stopped them. Without conversations, a job story is the author’s hypothesis. It should be called that and tested.
• Jobs that are too general. “To be happy” isn’t a job. A good job is tied to a situation and can be checked: it worked or it didn’t.
• A solution hidden in the wish. “I want a customer portal” is already an answer. Ask what the person wants to achieve.
• The theory doesn’t solve everything. For routine goods where the choice is hardly thought about it works less well. And the different schools of JTBD (Christensen’s, Ulwick’s, Moesta’s) disagree, so don’t take every reference to JTBD as “scientifically proven”.

Check yourself: how does a “job” in JTBD differ from an “audience”?

An audience is who someone is (age, profession). A job is what a person is trying to do in a particular situation. The same person in different situations hires different things, so the situation matters more than a profile.

The Value Proposition Canvas

Where from. The Swiss researchers and consultants Alexander Osterwalder and Yves Pigneur, with their colleagues Gregory Bernarda and Alan Smith, set it out in the book Value Proposition Design (2014). The canvas continues their own Business Model Canvas and unfolds one of its blocks in detail: the value proposition.

How it is built. Two halves that must fit each other like a puzzle.

• The customer profile. Jobs (what the customer is trying to do), pains (what gets in the way, scares or irritates on the path), gains (what they want and what would delight them). Everything is ranked by importance.
• The value map. Products and services (what you offer), pain relievers (how you remove specific pains), gain creators (how you bring specific gains).

When the map answers the most important pains and gains of the profile, there is said to be a fit. If not, there’s reason to change the offer or to be honest about its limits.

An example with the default persona.

• Job: confirm the boundaries, in order to build or sell.
• Pains: the documents are unclear; I don’t know how long it will take and what it will cost; I’m afraid the dispute will end up in court.
• Gains: build or sell with a quiet mind; a document to lean on.
• Products: boundary survey, expert witness report, review of another expert’s report, consultation.
• Pain relievers: I explain the documents in plain words at the consultation; price and time before starting; the order of work, step by step.
• Gain creators: a boundary plan ready to file with the county recorder.
• Left unanswered: “I’m afraid of court”. We can’t remove this pain: the court decides the outcome. An honest page doesn’t promise otherwise, and says how we can help: sort things out and set out the options.

This is the most important part of the canvas, and the one most often skipped: a pain you don’t cover has to be named. Work with the canvas of an invented workshop in the diagram: link the pains to the relievers and see which are left without an answer.

Job story Canvas

When I want to so that

When my tire went flat on the way and I need to ride to work this evening, I want to find out quickly what the repair costs and whether I can pick the bike up today, so that not lose a working day and not overpay.

• A story about the customer and their job, with no solution suggested. Good.

The sample is invented (a bicycle workshop). Replace the words with your own: situation, goal, result.

When it works. When you’re deciding what to include in the offer, how to arrange the blocks of a page, and what to say on the first screen.

Where it lets you down.

• The canvas is filled in without customers. The result is a beautiful hypothesis. It should be marked as a hypothesis and tested: talk to at least five people. Mark each entry: “known from conversations” or “assumed”.
• Invented “pain relievers”. If a pain is covered only on paper (“fast”, “high quality”), it’s a promise you won’t keep.
• Everything matters. Until you have set weights, the canvas is useless: the point is that three or four pains decide the matter, and the rest is small.

Positioning in one sentence

From the job and the canvas comes positioning: one sentence about who you are for, what you give and how you differ. A simple template was proposed by Geoffrey Moore in Crossing the Chasm (1991). In our own words: “For [whom], who needs [the job], [name] is a [category] that [the main benefit]. Unlike [the alternative], [how it differs and why it can be trusted].”

You needn’t use the template word for word; it’s there to force you to answer five questions in one sentence. If the answer doesn’t fit, you haven’t made up your mind yet. The test for honesty is the same as in plain style: no judgments without facts (“quality”, “unique”, “professional”), and every word can be backed up.

Check yourself: what do you do with a customer pain that your offer doesn’t cover?

Name it and say honestly that you don’t do this, or improve the offer. To pretend it’s covered is to make a promise you won’t keep.

Practice

Write down why a customer comes to you and check whether your offer answers it. The notes in the practice are gathered on a page that only you’ll see.

1. The customer’s job story in one paragraph: “When…, I want to…, so that…”. “When” is the situation, “I want to” is a goal with no solution suggested (without words such as “button”, “site”, “section”), “so that” is the result. There’s no “we” or “our” in the story.
2. The customer’s jobs: a functional one and an emotional one.
3. The canvas: three pains and what relieves each. That is either the name of an item of your offer from the brief in full, or the words “we don’t solve it”.
4. Two gains the customer expects.
5. Positioning in one sentence, without “quality” and “professional”.
6. What is known from conversations and what is a hypothesis. If there have been no conversations yet, say so, and add whom you’ll talk to.

The check looks at form: the parts of the story, no “button” in the wish, the link between pains and the items of the brief, the length of the positioning. Whether your story is true is for you to decide: ask yourself whether a real person from “for whom” would recognize themselves in it.