A Problem Framing Kernel

What you need before you have a problem worth solving

A few weeks ago, I caught up with my good friend Lucas Mara, a thoughtful design leader. We were deep in a conversation about the value of design in the age of AI when he mentioned Richard Rumelt’s classic, Good Strategy Bad Strategy. It had been years since I’d read it, so I went back.

Revisiting a book after fifteen more years of business and academic leadership is a strange experience. The book hasn’t changed, but you have. And the line that stopped me this time was one I’d underlined years ago without fully appreciating: “bad strategy is not simply the absence of good strategy.” Rumelt argues that most organizations mistake ambition for strategy. They list goals, dress them in inspiring language, and call it a plan.

“A real strategy has a kernel with three components: a diagnosis of the challenge, a guiding policy for dealing with it, and a set of coherent actions. Without all three, you don’t have a strategy. You have a wish.” — Rumelt

That idea sat with me. Because the same problem exists in design.

The missing kernel of problem framing

Designers and organizations talk about “problem framing” constantly. It appears in every design thinking workshop, every strategy deck, every innovation brief. But what is actually at the heart of a problem frame? What are the bare minimum elements without which you simply don’t have one?

It is easy to make an extensive list of everything that goes into framing a problem. It is much harder to reduce it to its essentials. As Blaise Pascal once wrote, “I would have written a shorter letter, but I did not have the time.” So I set myself the challenge to spend the time and find the problem-framing kernel.

A good problem frame has three elements: collect, connect, commit.

1. Collect

The first act of problem framing is to collect everything that might be relevant. Concepts, functions, needs, requirements, precedents, stakeholders, constraints, assumptions, and even the unknowns you suspect are lurking. Design problems are inherently ill-defined. You never start with a complete picture. So the work begins by deliberately expanding what you can see, increasing both the size and the resolution of the canvas on which the frame will eventually be drawn.

This is not brainstorming. It is disciplined divergence. You are assembling the raw material of the problem space, not waiting for it to appear. Without a rich collection, the following components have nothing to work with. A narrow canvas produces a narrow frame, no matter how sharp the thinking that follows.

2. Connect

Once the elements are on the table, the real thinking begins. Connections are the relationships between collected elements, and they are where insight lives. Some connections are obvious and expected. Others are surprising, weak, or counterintuitive. Some elements support and strengthen each other. Others oppose and contradict.

Both kinds matter. In fact, contradiction and tension tend to produce far better problem frames than tidy agreement. A frame built only from elements that sit comfortably together is usually a frame that has avoided the hard parts. Good problems have interconnected elements. They rarely include anything that floats free of the rest.

Without connections, a designer has no basis for deciding what to include in a frame and what to leave out. It is the tensions and dynamics between elements that reveal where the real problem lives.

3. Commit

Here is where most problem framing falls apart. You have collected your elements. You have mapped the connections between them. But until you commit, deciding which elements and connections to foreground and which to deliberately set aside, you do not yet have a frame. You have a landscape.

Committing is an act of editorial judgment. It means saying, “We are going to work on this and not on that.” It is closer to a film director choosing the final cut than a researcher cataloguing data. There is always more you could include.

What you leave out of the frame is as important as what you keep in, and the discipline of framing lives precisely in that choice. This is where the designer’s perspective becomes visible. Two people can collect the same elements, map the same connections, and commit to entirely different frames, not because one is wrong but because they’ve made different judgments about what matters most. That is not a flaw in the process. That is the process.

Why a kernel matters

Rumelt’s insight was that without a kernel, strategy degenerates into wishful thinking. The same is true for problem framing. Without collecting broadly, you frame from incomplete information. Without connecting, you frame from isolated assumptions. Without committing, you never actually frame at all. You just describe.

The three elements also work as a diagnostic. If a problem frame feels weak or unactionable, the failure almost always traces to one of the three. Did you collect too narrowly? Are the connections between your elements unexplored? Or have you avoided the hard part, refusing to commit to a point of view?

A kernel is not a full methodology. It is the minimum that must be present before anything else can work. Everything beyond it, the tools, the workshops, the canvases, is scaffolding. Useful scaffolding, sometimes essential scaffolding, but scaffolding nonetheless.

If you’re framing problems without this kernel, you might be doing something useful. But you’re not framing yet.

The article originally appeared on LinkedIn.
Featured image courtesy: Yasin Arıbuğa.