Most organizations intend to improve their products and services. If you are a manager, would you welcome one superb idea, or are you hungry for many more so-so ideas from which to choose? If you have to choose, would you go for quality or quantity? Given how much I detest dichotomous, either-or, framing, the answer to the above questions is – of course – “It depends.”
It depends on whether we are dealing with incremental improvement in a process, such as manufacturing operations, or with innovative breakthroughs that might lead to increased brand recognition or dramatic market share increases. For the former, the organization aims for improved performance based on tweaking here and modifying there. For the latter, the organization desires the “extremely” good idea that could dramatically change the organization’s market position.
And who better generates innovative ideas than a group of bright individuals, right? After all, more minds spawn more ideas than individuals working alone.
Not so fast.
According to Girotra, K., Terwiesch, C, and Ulrich, K, in their recently published paper in Management Science, “Idea Generation and the Quality of the Best Idea,” a conventionally-formatted group, working together in the same space and time, generates less impressive outcomes than a “hybrid” team does. Individuals in a hybrid team work on ideas alone first before they collaborate on improving the collection of ideas: assessing, selecting, and modifying. These authors also confirm what others have pointed out over the decades, that brainstorming is not nearly as effective as it was first advocated in the late 50s. Why?
In a typical group, people deal with not only free riding (see footnote below), but more importantly, the inhibitions they implicitly place on each other (“evaluation apprehension”). There is always a tendency for groups to move toward a norm, and thereby discourage “wild” ideas. More often than not, group members “build onto” each other’s ideas…a useful technique for incremental improvement, but not so much for innovation. What’s more, when one person is delivering her idea(s) in a meeting, others have to wait for their turns (“production blocking”). It is not the most efficient way of spending one’s time. And I am not even going to discuss the issue of introvert-extrovert, even though this dimension is bound to have impacts on group performance.
In their paper on idea quality, the authors highlight three flaws of previous studies on groups and innovation:
- Previous literature on innovative ideas has been focusing on the number of ideas generated by groups, assuming more ideas are likely to lead to better ideas.
- The traditional literature defines “quality” of innovative ideas by using average rating. As the authors point out in their theoretic foundation, innovation is about the best ideas, not the average of ideas.
- In addition, often only a couple of research assistants rated the quality of ideas. This practice makes both the validity and the reliability questionable.
So, the authors of the current paper provided a much more rigorous challenge and realistic scenario for participants on idea generation. As for rating ideas, they used a “web-based” rating system so that each idea received on average 20 ratings. They also introduced a “purchase-intent” survey for more realistic assessment.
The authors found that the hybrid team generated three times more ideas than the conventional team and that hybrid team’s best ideas were much better than the conventional team’s.
Granted, this study’s findings need to be replicated and confirmed before we should fully embrace the hybrid team concept. However, the logic of the paper is sound, and the findings are strong.
Yes, I have biases; I especially like it whenever my suspicion is proven right or my preference is vindicated, as in this case. Yet, studies like these also frustrate me because they remind me again of the disconnect between what we have known and what we have practiced in management. For instance, tying CEO’s pay to company performance has not proven effective, yet we still have skyrocketing rises in CEO’s pay. Brainstorming is not especially effective as conventionally done, yet we keep doing it. We know that the relationship between teamwork and performance is tenuous at best, yet we still impose teaming as our first course of action.
Dan Pink’s (link below) talk on motivation, highlighting “autonomy, mastery, & purpose” as the incentives to bring about better productivity and results, might be too dramatic for most managers to embrace immediately. However, forgoing brainstorming shouldn’t be that difficult. As my M&M piece (link below) indicates, canceling a few meetings will win gratitude and likely bring about higher productivity. Seriously, is there any danger to experimenting with the “hybrid team” model? Why not try it and see for yourself?
Till next time,
Staying Sane and Charging Ahead.
Direct Contact: firstname.lastname@example.org
“Free ridership” is a problem that will probably persist in perpetuity. Free riders occupy one end of the “bell curve” in the normal distribution. A more pertinent question is, how much resources do we want to devote to detecting and monitoring them? Granted if “free riders” increase over time, it is a problem. However, we need to assess how many is too many? Personally, I would much rather see organizations devoting resources to improving the work environment, thereby making the “free riders” problem almost insignificant.
Editor’s note: Dr. Yang has a PhD in Management from the Wharton Business School of the University of Pennsylvania. She taught at Wharton for a number of years, and consulted for small groups and small organizations and on cross-cultural issues. Her professional worldview comprises three pillars: 1. All organizations are social systems in which elements are inter-related. 2. To improve organizations, the focus should be on the positive dimensions on which to build. This philosophical foundation is Appreciative Inquiry. 3. Yang subscribes to the methodological perspective that she is part of the instrument from which to gain quality data from respondents, and with which to compare and contrast with others’ realities.