Skip to content

Success Criteria: The Customer's Win Condition

Definition

Success criteria are the concrete, often highly specific standards a customer uses to judge whether a job has actually been accomplished well — separate from the job itself and separate from the business's own success metrics (first-year sales, breakeven time, market share). Knowing the job tells you what to build toward; only the success criteria tell you how good is good enough, and along which dimensions the trade-offs should be made.

In the Book

Chapter 4 opens with a cautionary tale: a large cell phone company asked Stephen to apply Jobs to be Done to a next-generation device, but knowing the jobs alone said "too little" about concrete trade-offs — screen resolution versus battery life, size versus durability. Only by digging into success criteria (how fast must a new employee master the device? how well must it surface a specific conversation thread?) could those decisions be made. The chapter's central case is Big Heart Pet Brands, which found cat owners forced into a trade-off between dry food (cheap, convenient) and wet food (better taste and texture, but messy) to satisfy the emotional job of feeling their cat's affection through feeding. Rather than pick a side, Big Heart's Meow Mix Tender Centers — dry exterior, wet meaty center — dissolved the trade-off and generated over $100 million in combined sales in two years. The chapter also uses the Tesla Model S, which scored 103 out of a possible 100 on Consumer Reports' rating scale, to show that a genuinely new solution can break an existing success-criteria framework by satisfying jobs (acting on environmental values) that the old measurement system was never built to capture.

Why It Matters

A job well understood still leaves open how a customer will actually judge the outcome — and companies that skip straight to business metrics (sales, share) build products that hit their own targets while missing the customer's. Naming success criteria up front turns "make customers happy" into testable, often surprisingly specific design requirements, and reveals when an existing scoring system itself needs to be replaced rather than beaten.