UNIVERSITY PARK, Pa. — Brand collaborations have been growing in popularity as a powerful marketing tool for businesses looking to generate hype and expand their customer base, but brand managers often rely on costly and time-consuming surveys and screenings to identify potential matchups. A recent study by a Penn State researcher and collaborators proposes a new method: an artificial intelligence (AI)-powered framework that can uncover promising brand collaborations by leveraging the social media chatter of consumers.
Combining traditional brand characteristics with information gleaned from social media posts that co-mention brands together, the proposed model could provide brand managers with a suite of potential collaboration opportunities tailored to their specific businesses, according to the researchers, who published their framework in the journal Information Systems Research .
In a survey of brand managers, brand pairs identified as favorable by the framework had nearly a 90% chance of receiving a higher selection rate than those the model eliminated as unlikely pairings, the researchers wrote. In addition, the framework categorizes its recommendations and provides practical guidance on the best way to pursue a collaboration, giving companies clearer insights into which brand alliances they should prioritize.
"Finding brand collaborations is a needle-in-the-haystack problem," said co-author Keran Zhao , assistant professor of supply chain and information systems at Penn State's Smeal College of Business. "Brand A might consider brands B, C and D as options, but they will never comprehend the whole brand network. The advantage of this framework is that it provides a systematic evaluation of hundreds of brands, as well as a detailed explanation about the how and why behind an opportunity."
The proposed framework is described by the researchers as a four-step system built around two key concepts: signal fusion — the ability to synthesize disparate bits of information into a cohesive insight — and actionability. The first phase focuses on the former, combining real-world brand info with co-mentions on social media to create a comprehensive network of brand connections that can identify promising potential partnerships.
For the study, the researchers collected brand personality survey information on characteristics like sophistication and familiarity for 450 top business-to-consumer brands, as well as 477,933 posts mentioning these brands together on the social media site Twitter, now X, in 2020. Using AI-based text classifiers, the researchers filtered out complaint-related posts and analyzed whether consumers viewed brands as complementary or competing. They then combined these insights with brand characteristics and co-mention frequencies to develop a broader network of brand relationships that could identify promising alliance opportunities.
"The four of us could sit together for eight to 10 hours reading Twitter posts, and we would still only see a small fraction of the conversations happening around these brands," Zhao said. "AI allows us to analyze this information at a much larger scale."
The researchers also tested the model's ability to predict future real-world alliances that hadn't yet developed, finding that the model was able to accurately predict 213 of 251 alliances that occurred after the data sample's cut-off date.
Phase two of the framework builds upon this validation test by surveying stakeholders about the model's proposed brand alliances. In addition to a consumer survey with 350 participants that found a strong preference for the model's recommendations over pairings the model eliminated, the researchers also surveyed 57 brand managers to gauge their interest in the model's proposals. Each brand manager evaluated 24 brand pairs — 12 predicted, 12 eliminated — and selected up to eight of the most promising pairs. On average, the managers selected the model's proposed alliances roughly 34% of the time compared to only around 7% for the eliminated pairs, a nearly fivefold increase in interest. The researchers noted that phase two of the framework was solely conducted for validation purposes, meaning users would not be required to conduct such surveys during real-world applications of the model.
Phases three and four were both designed to provide actionable insights into the brand alliance opportunities generated in phase one. For the third phase, the researchers used machine learning systems, a subset of AI that learns from patterns in data, to categorize the proposed collaborations based on how frequently the two brands appeared in posts together, the sentiment of the posts and the number of likes, replies and reposts. The model then plotted the proposals on a matrix with four distinct quadrants ranging from "premier opportunities," which represented the strongest and most high-visibility alliances, to "hidden niches," which highlighted more exploratory collaborations in emerging or specialized markets.
"One brand manager might run the model, and it might suggest 100 brands they should collaborate with," Zhao said. "That's practically impossible to do, so this stage gives a more systematic view for explaining to brands how important the opportunities are and which ones they should prioritize."
This matrix can be tweaked to tailor its insights to different stakeholder roles, like giving a portfolio-level perspective that could help identify internal brand competition or organizing by industry to highlight cross-sector opportunities
Finally, phase four used natural language processing and generative AI to generate two pieces of practical guidance for individual brand collaborations using context from social media posts. The first piece of generated guidance detailed which specific marketing element consumer discourse is dominated by, such as discussions about products or promotions. The second gave deeper insights into what topics the discourse is thematically about, like style or sustainability efforts.
"AI is very powerful at summarizing information," Zhao said. "If we can locate what customers are talking about and feed that information to AI, it can give us some direct interpretations about which direction to go when pursuing a brand alliance opportunity."
Finding a promising brand alliance, whether that's a jointly launched product or a collaborative ad campaign, can be a difficult task on its own, the researchers explained, but committing to one also presents a significant risk for a business, especially given the time, money and brand reputation at stake once such a collaboration is announced.
"A scalable system that can identify promising opportunities and provide specific recommendations for how to proceed could be an invaluable tool for brand managers — and an intriguing well of crowd-sourced creativity," Zhao said, noting that he and his collaborators would test the system with unexpected brand combinations for fun. "A lot of surprising things can happen."
Co-authors on the study include Pankhuri Malhotra, University of Oklahoma; Daniel Ringel, Vienna University of Economics and Business; and Yaxin Cui, who earned a doctorate degree at Northwestern University and is now at Wells Fargo.
