5. Jobs to Be Done
Jobs to Be Done reframed the unit of analysis from the user to the progress they are trying to make. People hire products to do a job; competition is anything that does the job; features should map to jobs. It gets need exactly right and says too little about urgency, which for a startup is the difference between a job that exists and a job someone will pay to have done this quarter.
- What is Jobs to Be Done and how does it change product discovery?
- What does it mean that competition is anything that does the job?
- What does JTBD miss for startups?
§ 5.1What it gave us#
§The milkshake. Clayton Christensen's story, retold in the 2016 Harvard Business Review piece and the book that followed, is the most efficient teaching device thinking has produced. A fast-food chain wanted to sell more milkshakes and had tried every improvement to the product. Then someone watched, and noticed that half the milkshakes sold before eight in the morning, to solo commuters, who took them to the car. The milkshake was not competing with other milkshakes. It was competing with bananas, bagels and boredom, for the job of making a long drive bearable with one hand free.
§That is the contribution in one image. People do not buy products; they hire them to do a job. The unit of analysis moves from the user, with their demographics and preferences, to the progress the user is trying to make in a specific circumstance. Three consequences follow, and the original guide listed them well: understanding the job matters more than understanding the user; competition includes anything that does the job, not just products that look like yours; and features should map to jobs, so a feature that maps to none is decoration.
§ 5.2What it assumes#
§ assumes the job exists and is stable enough to study. The method's tools, the job statement, the switch interview that Bob Moesta developed for understanding why someone fired one solution and hired another, the outcome statements Ulwick uses to score unmet needs, all presuppose a person who has already tried to get the job done and can tell you about it. That is a fair assumption in a mature market with incumbents to be fired.
§It also assumes that understanding the job deeply is the scarce thing, and that once understood, a company has what it needs to build. This is where it is quietest about a startup's real question.
§ 5.3Where it breaks at one to five#
§The break is urgency. A job can be real, well understood and shared by millions of people, and still not be one anyone will pay to have done differently this quarter. JTBD tells a founder what progress people want. It says less about how badly they want it now, from someone new, at a price that funds a company. A startup with a beautiful job statement and no urgency has a market that will arrive eventually, which is another way of saying it will arrive after the runway.
§The second break is the switch interview's dependence on incumbents. In a nascent market there is nothing to fire. The AI products of 2023 to 2025 mostly created jobs people did not know they had, and a method built on studying past switches has less to say about a job that has never been done.
§ 5.4What to keep#
§The reframe, entirely. Ask what job the person is hiring for and in what circumstance. Treat anything that does the job as a competitor, especially the spreadsheet and the intern. Map every feature to a job or cut it. Then add the question JTBD leaves open, which Product Sprint's conversations are built to answer: how much did this cost you last week, and what did you do about it?
Understand the job better than anyone. Then find out how badly the job needs doing this month.
- Clayton Christensen, Taddy Hall, Karen Dillon, David Duncan, Know Your Customers' Jobs to Be Done, Harvard Business Review (September 2016). hbr.org/2016/09/know-your-customers-jobs-to-be-done
- Clayton Christensen et al., Competing Against Luck (2016). www.harpercollins.com/products/competing-against-luck-clayton-m-christensentaddy-hallkaren-dillondavid-s-duncan
- Anthony Ulwick, Jobs to Be Done: Theory to Practice (2016), on outcome-driven innovation. jobs-to-be-done-book.com
- Bob Moesta, Demand-Side Sales 101 (2020). www.demandsidesales.com