People Want Groups to be Just the Right Size
When WeldWerks Brewing Company introduced Raspberry Lemonade Milkshake IPA, twenty-one other breweries pitched in. The launch drummed up a good deal of interest in the new offering but also raised a few eyebrows on the beer-reviewing platform, Untappd. “How many breweries were a part of this beer?” asked one, while another wondered “how much all the collaborators actually put into this beer.”
The sheer number of contributors led consumers to doubt the final product. Still, mega-collabs such as this were inevitable as people tested the boundaries of how big might be considered too big in the rush to reap the rewards of collaborating.
Decades earlier, collaboration might as well have been termed bi-laboration, as few dared to add even a third party to their carefully curated duos. Air Jordans took flight in 1985, attesting to the acceptability among sneakerheads of a shoe company working with a professional athlete. When Karl Lagerfeld developed a line of clothing with H&M in 2004, he characterized the crossover as too successful. As Lagerfeld lamented at the time, the availability of his items “was supposed to last two weeks and it’s over in 25 minutes.” There appeared to be no restrictions on who could benefit by pairing up.
While most firms were riffing on the public’s belief that two heads are better than one, others started to ask whether mere dyadic collaborations might be supplanted by triads, quartets, or baker’s dozens. As with musical supergroups and blockbuster film franchises, when you hear that somewhere in the neighborhood of five to twelve A-listers joined forces, you just ask where and when to get your hands on their work.
But, as the size of a collaboration waxes, interest in each specific member of the collaboration wanes. So when Raspberry Lemonade Milkshake came along, consumers cared less about the name of every brewery on the can’s label and more about the list’s length. The beer’s existence presumed that, if people believe that a handful of collaborators is good, then twenty-something heads must somehow be seen as better still.
My colleagues and I—regular participants in pub trivia nights where team diversity is good, but increasing the size of the group for its own sake admits idiots who drag down our winnings—had our doubts. And we suspected that other people, having personal experiences with school projects and work deliverables bolstered and burdened by variation in group size, might also prefer collaboration but still recoil at the broth-spoiling tendencies of too many cooks.
Our question was not whether group size makes things better or worse but rather what intuitions outsiders (like beer drinkers and moviegoers) have about joint efforts. In an exploratory study, we brainstormed different domains of collaborations, then asked people what they thought was the best number of collaborators needed to produce something in each domain: What’s the best number of programmers to create an app, the best number of brands to design a jacket, or the best number of planners to route a fire escape? The responses were rarely less than two collaborators and rarely over thirty.
To us, this suggested that people have a sweet spot in mind when they think about the best team size, endorsing the Goldilocks goal of neither too small nor too big. When later studies asked research participants to evaluate products that we claimed had been made by groups of various sizes, participants were more enthusiastic when they thought the products, which were otherwise the same, were crafted by moderately-sized teams (such as three shoe designers) rather than smaller teams (1 designer) or bigger teams (9 designers).
This enthusiasm for teams that are neither too small nor too large makes good experiences better, like when we had research participants eat a cookie as part of a taste-test. Some were told that one baker made the cookie, others that eight bakers made the cookie, and both groups were lukewarm toward it. Those in a third group, after learning that it was supposedly baked by four people, raved after eating the exact same cookie.
Back on Untappd, beer enthusiasts regularly assign lower scores to collaborations that seem too big. Other WeldWerks collaboration beers smaller in size regularly score around 4.5 stars while the rating for Raspberry Lemonade Milkshake hovers around 4. One possible explanation for this drop-off is that people extrapolate from their own firsthand experiences of overly large groups failing to function as well as nimbler ones.
But what of the biggest big number: infinity? If less than a hundred can seem too big, then growing larger still might risk even more precipitous falls in perception. In the realm of collaborations, though, an easy synonym for ‘infinite’ is ‘open-source’, and prominent modern examples point to this all-for-one framing as a way to revel in collaborative riches.
In response to COVID-19, Other Half Brewing Company publicly released a beer recipe. Cast as a worldwide collaboration, the company called for others to tweak the recipe, brew it, sell it, and donate the proceeds to hard-hit hospitality workers. To date, hundreds of breweries have put their own spin on the original recipe and raised thousands of dollars in the process. Meanwhile, Wikipedia and Reddit, who rank among the most trafficked sites online, open their doors for anyone to edit their content.
Although our results showcase the benefits of collaborations of a certain just-right size, perhaps another, universal boost might result from an unspecified and unlimited number of like-minded people working together—or at least being seeing that way.
For Further Reading
Maglio, S. J., Wong, O., Rabaglia, C. D., Polman, E., Reich, T., Huang, J. Y., Hershfield, H. E., & Lane, S. P. (2020). Perceptions of collaborations: How many cooks seem to spoil the broth? Social Psychological and Personality Science, 11 (2), 236-43.
Rao, A. R., & Ruekert, R. W. (1994). Brand alliances as signals of product quality. Sloan Management Review, 36, 87–97.
Sam Maglio is an associate professor of marketing and psychology at the University of Toronto Scarborough. His faith in this research led him, after one experiment found that people put the collaborative ideal at 8 for a behavioral science project, to invite seven of his closest friends to join as named coauthors on the paper.