One of the most instructive papers on serial innovation comes out of left field from Griffin, Hoffmann, Price and Vojak – the latter three out of the University of Illinois, Urbana-Champaign. (Which will bring a smile to regular readers of this space.)

Their paper doesn’t reference Roger Martin, but it confirms much of what he’s described about the saliency phase of opposable thinking.

The best piece that I’ve taken away from the paper is the definition of an Interesting Problem. A problem is deemed interesting only if:

  • The firm can actually solve it and management accept the solution. (Feasibility)
  • Customers will pay to have that problem solved. (Marketability)
  • Will it be a big deal for the firm. (Impact)

Prepare to be annoyed if you just don’t think in such terms.

Problem Exploration

A lot of effort goes into problem identification and then really understanding it. Why does the problem exist? What’s driving it? What reinforces it? Is it a wicked problem?

Problem exploration is fraught with inductive-deductive tension. For instance, nobody was walking around in 1860 expressing a need for a mechanical broom. Many generations have since bought vacuum cleaners though! The deductive mode of thinking typically ends with “nobody is saying that they have this problem” and sometimes ends with “if it was a problem worth solving, somebody would have already invented and sold a solution”. It’s akin to the whole ‘It’s impossible to find a loose $20 bill on the street’ economist argument.

The inductive reasoning is that given a set of gaps and opportunities, people should want this problem to be solved.

Arriving at a well defined problem set requires the consideration of both types of thinking. While it may be true that nobody out there is enunciating a problem, is there an opportunity to point out a problem to a market?

After all, what demand would there be for nightlights if children didn’t have the frame of reference of a monster under the bed?


There’s a yawning gap between ‘free’ and the payment of a single dollar for the solution.

So – is the problem or opportunity so great that a group of self-referential people will actually buy it? Will they actually take out their wallet and give you money for solving that problem. It’s a huge threshold. It’s greater than the surrendering of an email address.


Can the problem actually be solved? Frequently, some of the most profitable problems just can’t be technically solved.

The second aspect is social. Sometimes a solution to a problem requires humility. Certain solution sets are simply not acceptable, culturally, to a given group of people. Taboo’s apply.

It’s not uncommon for me to explore feasibility while exploring a problem and the market. You talk to smart people. They often have good insights into the feasibility of certain sets of problem, barriers you may not have known, and opportunities that may exist.


Some problems may be profitable and feasible to solve, a market segment may be served, however, in the grand scale of things – it doesn’t matter to the firm. It’s a much more nuanced. Even a $300,000,000 might not be interesting enough for a large enough firm.

I’m reminded of how Amphicoelias survived. They were massive animals, were vegetarians, and ate huge amounts of food to support themselves. The awesome size of their stomachs produced a fairly large ecosystem of bacteria that would break everything all down. To a human, a bunch of fiddleheads, if they existed back then (and they probably did), look like a pretty good meal. To an Amphicoelias – not so much. They’re not going to lower their head and die like a Giraffe would. They won’t stoop to that level.

Prospective impact really matters in relationship to the size of the firm.

In Sum

It’s fuzzy.

I’ll often get really excited about an interesting problem, only to discover that it isn’t interesting at all. And then I’m no longer excited.

You’re reconsidering the problem set a lot at the front end. It means frustrating a lot of people as you reconsider the set. It might also mean frustrating a lot of prospective consumers – people who are not paying you anything for a freemium version. You’re evolving and iterating. Testing and checking. And I suppose this is a very different mindset than what you might find elsewhere.

The paper provided welcomed clarity and some good common language around this area. I’ve summed it up for you, but it’s well worth the read if you want more detail.