Creative Problem Solving: A Step-by-Step Framework

Most people approach creative problem solving the same way they approach all problems—by jumping to solutions. The research on this is consistent and humbling: solution fixation, the tendency to lock onto the first plausible answer, is one of the most reliable ways to produce mediocre outcomes. The structured alternative is Creative Problem Solving (CPS), a framework that separates the problem-understanding phase from the idea-generation phase, and both of those from evaluation.

CPS isn't a creativity hack. It's a formal methodology with a 70-year research history, documented applications in business, education, and design, and a measurable effect on solution quality. Here's how it works.

What Is Creative Problem Solving?

Creative Problem Solving is a structured process for generating and evaluating novel solutions to complex, ambiguous problems. The key word is structured: CPS works precisely because it interrupts the normal human tendency to evaluate ideas as they're generated, which suppresses divergent thinking before it can produce genuinely original options.

The framework was originally developed by advertising executive Alex Osborn in the early 1950s and later formalized by Sid Parnes at the Buffalo State University. Osborn's 1953 book Applied Imagination introduced the core distinction between divergent thinking (generating many possibilities) and convergent thinking (selecting among them). Parnes expanded this into a six-stage model that remains the basis for most CPS frameworks used today.

The CPS Framework: Six Stages

Stage 1: Identify the Mess

Before defining a problem, CPS asks you to identify the broader situation—what Parnes called "the mess." This involves scanning the full context: What's producing anxiety or confusion? What goals are blocked? What seems wrong without being precisely diagnosable?

The output of this stage isn't a problem statement—it's an inventory of tensions, constraints, and goals that might indicate where the real problem lies. The mess is useful precisely because it resists premature definition.

Stage 2: Gather Data

Once you've identified the mess, collect relevant information: facts, feelings, observations, hunches. CPS deliberately includes subjective data—gut reactions, aesthetic impressions, emotional responses—alongside objective data. This reflects the finding that creative insights often begin with a vague sense of wrongness or dissonance rather than a clearly articulated problem.

The goal here is divergent: generate as many types of data as possible before narrowing.

Stage 3: Frame the Problem

This is the most important stage and the one most often skipped. Problem framing involves translating the mess into a precise challenge statement, typically written as "How might we...?" or "In what ways might we...?"

The framing isn't just stylistic. Research by creativity scholars including Donald Treffinger shows that the way a problem is framed strongly predicts the quality of solutions generated. "How might we reduce customer complaints?" and "How might we design service interactions that customers want to repeat?" are technically the same operational concern, but they unlock completely different solution spaces.

Generate multiple frames before choosing one. The frame you start with is almost never the most generative one.

Stage 4: Generate Ideas

With a problem frame in place, CPS enters the divergent ideation phase. This is where standard brainstorming principles apply: defer judgment, go for quantity, build on others' ideas, combine and extend. But the structure of CPS means that by this point, you've spent significant time on problem definition—so the ideation is more targeted than freeform brainstorming allows.

The divergent thinking exercises on this site are built around this stage: generating multiple solutions without premature evaluation. The psychological research consistently shows that groups and individuals who defer evaluation generate more creative solutions than those who evaluate in real time.

For more on the mechanics of divergent ideation, see divergent thinking examples.

Stage 5: Develop Solutions

Not every idea generated in Stage 4 is a solution. This stage involves selecting the most promising ideas and developing them—strengthening their appeal, adapting them to real constraints, and preparing them for evaluation.

The key tool here is convergent thinking: using criteria to systematically compare options. Good criteria include novelty (is this genuinely different?), feasibility (can it actually be implemented?), and relevance (does it address the problem frame?). For a deeper look at this complementary process, see convergent thinking.

Stage 6: Formulate a Plan

The final stage of CPS translates a selected solution into an implementation plan: what specific steps will be taken, by whom, by when, and with what resources. This stage also includes identifying potential resistance—what could block implementation, and how those obstacles can be addressed.

CPS solutions that skip this stage tend to die in committee. Implementation planning is where creativity meets organizational reality.

CPS vs. Design Thinking

Design thinking, developed at IDEO and formalized at Stanford's d.school, shares CPS's emphasis on problem framing and divergent ideation. The main differences are emphasis and tooling: design thinking places more weight on user research and prototyping, while CPS places more weight on the ideation process itself and the psychological conditions for creative output.

Both frameworks reject the assumption that good solutions emerge from more analysis of the original problem. Both insist on a dedicated divergent phase separated from evaluation. If you've used design thinking, CPS will feel familiar—more psychologically rigorous in some respects, less focused on physical prototyping.

CPS in Practice: Two Examples

Spencer Silver and the Post-it Note: In 1968, Spencer Silver developed an adhesive that was strong enough to bond surfaces but weak enough to peel away without damage. It wasn't useful for anything obvious. Art Fry, a colleague at 3M, recognized years later that Silver's "failed" product solved a problem he'd been frustrated by—marking his place in a hymnal without damaging the pages. The CPS insight wasn't Silver's chemistry; it was Fry's problem reframe: "What if the weakness of this adhesive is actually its defining feature for a different application?"

Netflix's streaming pivot: Netflix's 2007 decision to launch streaming alongside DVD mail was, at its core, a problem reframe. The original problem ("How do we deliver DVDs faster?") had obvious solutions, all incremental. The reframe ("How might we make physical delivery irrelevant?") opened a solution space that competitors weren't looking at. The creative problem solving examples post covers this and other documented cases in more depth.

What Makes a Problem Right for CPS

CPS works best on problems that are open-ended, ambiguous, and important enough to warrant more than a quick fix. It's less appropriate for problems with clear technical solutions, where additional creativity adds noise rather than signal.

Good candidates for CPS:

• Strategic challenges where the goal is clear but the path isn't
• Design problems with multiple valid constraints
• Organizational problems where the obvious solutions have already failed
• Innovation challenges where incremental improvement isn't sufficient

Poor candidates:

• Well-defined technical problems with known solution spaces
• Urgent operational issues requiring immediate action
• Problems where constraints are so tight that most of the solution space is unavailable

Building CPS Skills

The cognitive abilities underlying CPS—divergent production, convergent evaluation, analogical reasoning—can be trained. Creative thinking activities that target these specific mechanisms are more effective than general "creativity" exercises.

For divergent production specifically, the divergent thinking exercise on this site measures fluency, flexibility, and originality across sessions, so you can track improvement over time rather than working on feel alone.

The most useful thing you can do before your next complex problem is practice problem framing in low-stakes contexts. Take a familiar situation—a recurring team conflict, a product feature you're unsatisfied with—and generate five different "How might we...?" frames. The frame that produces the best solutions usually isn't the first one you write.

Ready to train your creativity? Try science-backed exercises that measure and improve your creative thinking. Start a Free Exercise