Editor’s note: Shari Morwood is an innovation process facilitator at Minneapolis-based research firm Ideas To Go. This is an edited version of a post that originally appeared here under the same title.
How important is it to start with a good idea? Companies invest a lot in developing new products and finding ways to talk about them. So coming up with good ideas early in the process really is critical. But can you tell good ideas from bad all the way back at the fuzzy front-end of innovation?
I recently had a conversation near the campus of the University of Colorado at Boulder with Dr. Laura Kornish on this very subject. The value of a good idea is a topic she is very interested in – she actively researches innovation and new products, credibility of information and search problems.
Morwood: How did you get interested in this topic?
Kornish: It started when I met my co-author Karl Ulrich at a conference. Karl is vice dean of innovation and CIBC Professor of Entrepreneurship and e-Commerce at the Wharton School. He has authored several books, including Innovation Tournaments as well as his most recent [free] e-book Design: Creation of Artifacts in Society. He and I – and many other professors – teach product development courses where students generate and develop ideas for new products. I told him that I was hearing a lot of the same ideas in class over and over again: the high-heel shoe with the removable heel; the hands-free toilet; an electronic gadget to help find lost keys. In our first paper together, we measured how much repetition there was in sets of ideas generated by our students. We found that there was some strict repetition but not much. But there was a lot of similarity in clusters of ideas – nuanced or not-so-nuanced variations on a theme.
We wondered what it meant when an idea was generated more often. Did that mean the idea was obvious and maybe not such a good idea? Or did that mean the idea was popular and had a lot of market potential? We used the Darwinator system [an online idea management tool] to gather people’s ratings on the ideas. We found support for the hypothesis that ideas with more similar ideas were also rated higher: those ideas were more popular in both generation and in rating.
The next step was to ask whether those higher ratings really meant anything, in terms of dollars-and-cents in market outcomes. For our current study, we are working with a set of ideas that were actually commercialized, so we can see if there is a connection between how good the initial idea seems and whether or not the product based on that idea sells. A logical place to start with that was to ask product development and marketing professionals to rate the ideas, so we would have an informed opinion about which ideas were good and which were not.
That seems like a great way to go since experts and consultants are hired every day to help companies pick winning ideas. How did things turn out?
Worse than we expected. The experts did not agree and when we don’t have agreement, we don’t expect the results to be predictive. And they were not.
So what did you do next?
We decided to ask consumers what was a good idea. And that turned out to be a very good thing to do. We were able to ask a larger panel of consumers about the ideas – about 100 for each idea – compared to the panel of experts – only seven, which is actually a lot of experts. We see some crowd wisdom: the consumers’ ratings are predictive. We find that the ideas rated more highly by consumers are more likely to be commercially successful in the marketplace.
Our philosophy at Ideas To Go is that it’s important to start with the very best ideas you can to create something that wins in the marketplace. What have you found in your work?
I notice that it’s hard for companies to think about “investment” in idea generation. Ideas seem cheap and easy to come by. In a sense, people have no respect for ideas because they always think “there’s more where those came from.” But we find that the ideas do matter. On average, better ideas will lead to more successful products. So you need discipline in the idea-generation process to come up with really good ideas. The more ideas you have, the better the chance that you will find outstanding ones. Think about it in terms of height: the three tallest people out of a group of 1,000 will probably be taller than the three tallest people out of a group of 100. It’s statistical logic – the bigger the set of ideas you draw from, the better the bests ones will be.
We see people that are very wedded to the ideas they generate and perhaps a little afraid to push out too far. What have you seen in this regard?
Yes, people have a rebuttal to the above stated statistical logic and it’s that the ideas they generate are not random, therefore they believe their first ideas will be the best. But that’s not the case. A big mistake companies make in new product invention is that they don’t explore broadly enough at the beginning. It’s cheap to try out ideas and eliminate them early. It’s costly to do it later. They don’t take the time to develop a range of ideas. Getting good ideas is a process that requires discipline. Falling in love with one of your ideas or pushing through something the boss likes means there isn’t enough exploration around the possibilities.
When you get a lot of ideas, what’s the best next step?
We’ve found the best thing to do is screen them through consumers early. Consumers can help sort the dogs from the winners early. We see that better ideas create better products and that asking consumers is how to tell which ideas are better.