Some Dell laptops have an Intel processor inside, and some Betty Crocker brownie mixes use Hershey’s chocolate. The idea behind such co-branding is to generate synergies and marketing efficiencies. Does the strategy really work?
Yes, suggests research by Chicago Booth’s Sanjog Misra and Bradley Shapiro and Booth PhD candidate Yewon Kim—although it works better for some parties than others. And a difficult reality for managers and researchers is that predicting the magnitude of such a collaboration effect prospectively is nearly impossible.
The researchers studied brand collaboration in an unusual setting. Rather than analyze data involving commercial products, Kim, Misra, and Shapiro looked at three major museums all located in the same US city. While arts institutions aren’t typical commercial products, the fast-growing arts industry represented $704 billion in spending in 2013, compared with $619 billion for construction and $270 billion for utilities, write the researchers, citing data from the National Endowment for the Arts and the US Bureau of Economic Analysis.
Memberships rose overall after two museums separately collaborated on joint, co-branded exhibitions with the staff of a third local museum that had closed for building renovations. But research finds that a significant portion of the new memberships came from first-timers.
Kim et al., 2019
The researchers don’t identify the museums involved, citing a nondisclosure agreement, but write that during the time period they studied, “one major museum with a highly recognized brand” closed for a three-year renovation. While the work was being done, this museum collaborated separately with two other museums and held exhibitions in their buildings, with both the primary museum’s and the partners’ branding. The participating institutions shared their collections as well as their curatorial staffs. Exhibitions were displayed cohesively, mixing collections from both the primary institution and its partners. Marketing campaigns emphasized the joint nature of the exhibitions, and the collaborating institutions used the same descriptions on their websites and in other promotional materials. They jointly hosted membership events.
To gauge the effects of co-branding, Kim, Misra, and Shapiro tapped SMU DataArts, a collection of information compiled by the National Center for Arts Research, for four years’ worth of the museums’ membership sales. They find that collaborating with the major museum led to an increase in memberships at both partner museums. During the collaboration year, people who hadn’t previously been members of the partner museums joined them. Meanwhile, demand dropped among people who had previously been members.
The results highlight how hard it can be for marketers to predict the effects, much less plan on the profit impacts of a collaboration. The museum partnerships increased demand among a new group of people, and “this makes for an interesting and difficult problem for managers,” Shapiro says, because “it is particularly difficult to predict the behavior of customers that you have never before seen in your data.” Standard models of consumer demand consider only data from before a collaboration and don’t have the benefit of later information. “You’re counting on behavior changes among a set of customers that has never before shown up in your data. As such, it is really difficult to know what their preferences are.”
Modeling data from both before and during the collaborations to try to predict the effects of the co-branding experience told a similar story. Collaboration can benefit a lesser-known brand—and for better-known brands, it may carry the risk of brand dilution. But, crucially, the research indicates, it’s hard to know the magnitude of the effects ahead of time.