Key Principle
Innovation fails at a rate of 80-90% because companies have no shared definition of what a customer "need" actually is. "95% of managers say there is internal disagreement on what a need is" (p. 41). Without a common ontology, every downstream function -- value proposition, segmentation, positioning, product development -- inherits this definitional corruption.
Jobs-to-be-Done Theory resolves this by providing an exhaustive taxonomy of six precisely defined need types, all anchored to the customer's core functional job:
- Core Functional Job -- the overarching task the customer is trying to accomplish
- Desired Outcomes -- 50-150 measurable metrics customers use to judge success at each job step
- Related Jobs -- adjacent functional jobs the customer needs done alongside the core job
- Emotional and Social Jobs -- how customers want to feel and be perceived
- Consumption Chain Jobs -- lifecycle jobs (installing, maintaining, disposing, purchasing)
- Buyer's Financial Desired Outcomes -- the buyer's economic metrics
"Jobs-to-be-Done Theory provides a framework for (i) categorizing, defining, capturing, and organizing all your customer's needs, and (ii) tying customer-defined performance metrics (in the form of desired outcome statements) to the job-to-be-done." (p. 48)
The critical distinction: "While a job describes the overall task the customer is trying to execute, an outcome is a metric the customer uses to measure success and value while executing a job." (p. 49) Jobs are qualitative anchors; outcomes are quantitative metrics. This conversion from wishes to metrics is what makes needs actionable.
Why This Matters
Most companies operate under the ideas-first paradigm: generate solutions, then test them. Roughly 68% of large businesses use stage-gate development that begins from ideas, not needs (p. 33). The structural flaw is that filtering cannot compensate for untargeted input. With 50-150 customer needs per job and 5-80% potentially unmet, the combinatorial space for random ideation is astronomical -- "the mathematical probability of someone coming up with an idea that satisfactorily addresses all the customer's unmet needs without knowing what they are or whether or not they are satisfied is close to zero" (p. 36). Ideas-first companies "struggle to achieve success rates greater than 10-20%" (p. 38).
Even needs-first companies fail when they lack definitional precision. Standard methods (VOC, focus groups, ethnographic research) capture vague desires in the customer's own words rather than measurable outcome statements. "Obtaining inputs in the customer's own words will more often than not result in the wrong inputs" (p. 42). The JTBD framework fixes this by specifying exactly what a need is, what types exist, and how to state each one -- enabling complete capture, quantitative prioritization, and predictive scoring of concepts before development begins.
Good Examples
Bosch circular saw: Given a precisely defined segment (25% of end users) and 14 specific unmet desired outcomes, the product team conceptualized a winning solution in just 3 hours. "Without these insights, innovation remains a game of chance; having them changes everything." (p. 51)
Wireless presenter (related jobs): A telescopic pointer evolved into a wireless presenter by enabling the presenter to advance slides, time the presentation, and shut off the projector. "Its value increased as it enabled the presenter to get more related jobs done." (p. 58) This illustrates how the related-jobs need type opens platform strategy.
Dyson bagless vacuum (consumption chain jobs): By addressing the underserved consumption chain job of dirt disposal -- a lifecycle step most competitors ignored -- Dyson created a differentiation advantage outside the core functional job entirely. (p. 60)
Counterpoints
"Just generate more ideas." More ideas do not improve targeting. With 15 unmet needs and 3 competing ideas per need, the combinatorial space is 3^15 = 14 million possible combinations (p. 36). Volume of ideas is the wrong variable to optimize; absence of targeting at the input stage is the actual constraint. "Doing something bad faster does not lead to better results." (p. 36)
"Customer personas define our target." Demographic personas create "phantom targets" -- demographic similarity does not predict needs similarity. "Customer personas that are built around demographic and psychographic data and claim to represent customer 'segments' are highly misleading as they usually create phantom targets." (p. 52) Products built for persona-based segments satisfy no one particularly well because the segment itself was never real.
"Customers have latent needs they can't articulate." This is a symptom of the company's definitional failure, not a property of customers. "Companies routinely try to satisfy customers' needs without a clear definition of what a need even is. It is like trying to solve a word puzzle without knowing what a 'word' is." (p. 45) Fix the definition, and customers can articulate their needs through structured outcome interviews.
Key Quotes
"The 'ideas-first' approach is inherently flawed and will never be the most effective approach to innovation. It will always be a guessing game that is based on hope and luck." -- Tony Ulwick, p. 31
"How to get a handle on customer needs is an unsolved mystery -- and that mystery is killing innovation." -- Tony Ulwick, p. 46
"The core functional job is the anchor around which all other needs are defined." -- Tony Ulwick, p. 53
"While defining the functional job correctly is important, uncovering the customer's desired outcomes... is the real key to success at innovation." -- Tony Ulwick, p. 57
Rules of Thumb
- A customer has 50-150 desired outcomes per core functional job; 100-200+ total needs across all six types. If your needs list is shorter, it is incomplete.
- The core functional job is stable across time, geography, and solution category -- research on it does not depreciate with technology shifts.
- Products that get the job done 20% better or more are "very likely to win in the marketplace" (pp. 48-49). Use this as a concrete go/no-go threshold.
- Each of the six need types opens a different competitive lever. Capturing only functional needs misses emotional positioning, adjacent jobs, lifecycle friction, and buyer economics.
- Quantitative validation is non-negotiable. Qualitative methods discover what the needs are; only statistical research determines which are unmet and how they cluster into segments.
- If the core functional job is defined incorrectly, every downstream need captured against it is misanchored. The job definition is the single highest-leverage decision.
Related References
- Desired Outcome Statements - Structured syntax for capturing needs
- Why Ideas-First Innovation Fails - Why the alternative approach fails