Problem Framing
Problem Framing
What is problem framing?
- Problem framing sets the range of solutions
- Four elements: phenomenon, scope, criterion, perspective
- Rewrite it 3 to 5 ways and compare
What problem framing actually changes is not the quality of ideas but the budget and the owning department. Frame 'sales performance is poor' as a capability problem and the training team gets it; frame it as a lead shortage and marketing gets it; frame it as a pricing problem and product planning gets it. Even with the same numbers, who takes accountability and how performance is measured is decided by this one sentence. If you can't agree on the definition in the first 30 minutes of a meeting, the remaining two hours are spent with everyone solving a different problem.
Sorting framing statements into types makes them easier to work with. There are symptom statements ('18 claims a month'), gap statements ('12 percentage points below target'), and opportunity statements ('we have data our competitors don't'). For a document you will hand to an execution team, gap statements work best, because the baseline and the target are built into the sentence, so what counts as done is already settled. Test the scope too. Restate the same problem one level higher and one level lower, and the level where you can realistically act becomes clear.
It also helps to be clear about the boundary with neighboring tools. 5 Whys digs into the cause of a problem that has already been defined, and an issue tree breaks down a problem that has already been defined. Framing comes before both: the step where you choose which hole to dig. That is why teams write several candidate statements in the 'How Might We' format. Not because any one sentence holds the right answer, but to compare side by side the solutions that follow each time the wording changes.
The most common mistake in the field is disguising a solution as a problem. 'We need more training' and 'We need to replace the system' are already answers, so a proposal written this way produces not a single alternative at the review stage. At the other extreme, some teams keep reframing for its own sake and delay kickoff by two months, so it is better to cap the exercise at 3 to 5 candidate statements in one session and put the chosen definition on the first line of the document. Simply copying down the problem as management handed it over is risky too. At least once, ask 'What gets left out if we solve it with this definition?'
John KimInstructor John Kim, author of B2B Marketing Design and CEO of Nanum Consulting
Ara LimSpecialist in global leadership, AI, and business communicationIt developed out of problem-structuring research and the 'problem reframing' stage of design thinking. Theodore Levitt's insight that 'customers don't want a drill, they want a hole' is the classic example cited.
Consider a hypothetical case. Company A makes electronic components, with 420 employees and KRW 95 billion (≈ US$71,075,900) in annual revenue. When client complaints over the last 3 months rose to 18 a month, the quality team asked for 4 more inspectors. Just before sign-off, the plant manager pushed back, and the team had to start over by redefining the problem.
- When the plant manager asked the quality head, "Of the 18 claims, how many could inspection have caught?" no answer came, and approval for the added headcount was put on hold for two weeks.
- When a production engineering manager sorted 18 cases from three months, 14 were concentrated in two clients, and 12 had surfaced within two weeks of a design-change notice.
- The planning director rewrote the problem in three sentences: Is it a lack of inspection capability, design-change information reaching the floor too late, or a client-approved spec that is too vague?
- After comparing the solutions that came out of the three statements, they adopted a gap-style definition: 'It takes an average of 6 days for a design change to reach the work standard; the target is 1 day.'
- After four months of running automatic change notifications and an approval procedure for updating standard documents, the lead time for reflecting changes dropped from 6 days to 1.5 days, and claims fell from 18 to 7 per month.
- The request for 4 additional hires was withdrawn, so KRW 190 million (≈ US$142,200) a year in labor costs was never spent. The quality head later recalled, "Even with more inspection stations, those 12 cases would have gotten through anyway."