Finding the job your customer is hiring you to do

Most marketing teams can describe their customer in some detail. Aged 35 to 54, homeowner, two children, shops at the weekend, likes the brand. What they often can’t say is what was going on the day that person decided to buy. That second description is the one that explains behaviour, and it’s the one jobs to be done goes looking for.

The idea comes from Clayton Christensen and colleagues, and it rests on one shift in framing. People hire products to make progress in a particular situation. The Christensen Institute describes a job as the progress someone wants to make in their circumstances, with functional, social and emotional sides to it. The circumstance is the important part. The same person hires a takeaway for a different job on a Tuesday night after a long shift than on a Saturday with friends round, and a description of the person can’t tell those apart.

The well-known example is a fast food chain trying to sell more milkshakes. Asking customers what they wanted from a milkshake produced the predictable answers about flavour and thickness, and acting on them changed nothing. Watching when and how milkshakes were bought turned up a group of morning commuters who wanted something that would last a dull drive and could be managed with one hand. The milkshake was competing with bananas and bagels, and the improvements that mattered were the ones that served that job. The story has been retold so often it has worn a bit smooth, and some JTBD practitioners argue about how it should be read, but the point holds up: the circumstance told them more than the customer profile did.

The method most associated with finding jobs is the switch interview, developed largely by Bob Moesta. You interview people who’ve recently switched, to your product or away from it, and reconstruct the timeline of how it happened. Moesta interviews customers rather than prospects, because prospects can only speculate about what they’d do, and customers can tell you what they did. The timeline runs from a first thought, the moment the old way started to feel inadequate, through passive looking, active looking and deciding, to using the new thing. You work through it in order, with dates and details. Where were you? What happened that week? What else did you look at? What did you do instead before this? The detail is what stops people giving you the tidy version they’ve told themselves since.

Along that timeline, you’re listening for four forces. The push is what made the current situation intolerable. The pull is what made the new option attractive. Against them sit anxiety about the new option, will it work, will I look foolish, what if I lose my data, and habit, the comfort of carrying on as before. Moesta’s point is that reducing anxiety and habit often matters more than adding pull, which is good news for anyone running a website, because the anxieties tend to be specific and fixable. The user interviews lesson covers the basic discipline of asking about the past instead of the future, and switch interviews are a particular, structured use of it.

This is where it differs from personas. A persona describes a type of person. A job describes the progress someone is after in a situation, and one person can have several jobs while very different people share the same one. NN/g’s view is that the two can work together, and that a good persona built on research with real users already contains much of what a job describes. The trouble is that many personas are built another way, as demographic segments with a name and a stock photo, assembled in a workshop from what the team already believed. NN/g notes that many so-called personas are marketing segments in disguise.

An assumption-built persona causes problems because it feels like evidence. “Sarah, 42, values quality” gets quoted in briefs and used to settle arguments, and nobody remembers that Sarah was invented. It also points tests in the wrong direction. If you think your customer is defined by being a busy professional who values quality, you test premium imagery. If you’ve found that people switch the week their old supplier lets them down, you test a page that speaks to that week.

Turning jobs into test ideas works best force by force. For each one you found, ask what on the site could strengthen or weaken it.

• A strong push suggests leading with the struggle people are leaving behind, in their words, on the pages where they arrive.
• A clear pull suggests making the specific progress visible early, ahead of features they have to translate themselves.
• An anxiety suggests answering it where it arises: what happens to my old account, how long the change takes, what if it doesn’t suit me.
• A habit suggests making the first step smaller, such as importing existing data, offering a trial that needs no setup, or letting people keep the old way running alongside.

Each of those can go into the if-then-because shape from the hypothesis lesson, with the force as the because.

When you brief an agency or a researcher, ask who they’ll interview and how recently those people switched. Recent matters, because the further back a purchase is, the more the timeline gets replaced by a neat story. Ask whether they’ll include people who switched away, because the forces that drove them out are often the same ones nearly stopping new customers. Ask to see the timelines, and not only the summary, so you can check the job came from what people described. And if the output is a set of new personas with names and photos, ask what evidence each line rests on.

The value of all this is that it gives you a reason to expect a change to work that comes from customers’ own accounts of their decisions. A test aimed at a real anxiety in a real circumstance has a mechanism you can check afterwards. A test aimed at an invented Sarah only tells you whether Sarah’s imagined tastes matched anyone’s.

Put it to work

Find three customers who bought from you for the first time in the last month. Ask each one to walk you through the day they first thought about solving the problem, and everything between then and the purchase. Write down what was pushing them, what pulled them to you, and what nearly stopped them.

Your answer

Show the worked example

There is no right answer. Write yours first, then compare.

Sources

1. Christensen, Hall, Dillon and Duncan, Know Your Customers' Jobs to Be Done, Harvard Business Review, 2016
2. Christensen Institute, Jobs to Be Done Theory
3. Bob Moesta, Jobs to be Done Live Case Studies, Business of Software
4. NN/g, Personas vs. Jobs-to-Be-Done