Definition
A three-way classification of problems proposed by Brenda Zimmerman (York University) and Sholom Glouberman (University of Toronto), used by Atul Gawande as the diagnostic that decides which intervention fits.
- Simple — a problem like baking a cake from a mix. There is a recipe; once a few basic techniques are mastered, following the recipe brings reliable success.
- Complicated — a problem like sending a rocket to the moon. Multiple people, multiple teams, specialist expertise, no straightforward recipe — but solvable, and the solution generalises: once you can do it, the next rocket is roughly the same problem.
- Complex — a problem like raising a child. Every instance is unique. Experience helps but is not sufficient. The next child may require an entirely different approach. Outcomes remain uncertain. Yet we know it can be done well; it is complex, that's all.
Why it matters
How it works
The diagnostic is a quick triage. Ask: Can the work be reduced to a sequence of well-defined steps where doing them in order produces the expected result? If yes, the problem is simple. If no but the work generalises — solving it once teaches you to solve the next one — it is complicated. If no and every instance is unique, it is complex.
Gawande applies the classification at the level of parts of work rather than whole jobs. Surgery as a whole is complex (patients vary in irreducible ways), but within surgery, not skipping the antibiotic before incision is a simple problem — well-defined, the same answer every time, easily checklisted. The construction industry's two-checklist system uses exactly this decomposition: simple parts get the task schedule, complex parts get the submittal schedule.
The deepest move the classification makes is to treat complex as a real category — not as "complicated but harder," not as the failure of analysis to reduce a problem fully, but as a distinct kind of problem with its own appropriate intervention.