Data Speaks Volumes, If You Know How to Listen

The streetlight effect is often used to describe a bias whereby humans look for things in the easiest place to find them, and it is often mentioned as a critique of social science research that relies on traditional—and flawed—datasets.

In the early 1990s, Peter Golder suspected the streetlight effect was behind the conventional wisdom that market pioneers often become long-term market leaders. Up to that point, the academic literature had used two databases that only sampled survivor firms and, in the process, relied on self-reports of single informants to classify pioneers. Moreover, one of the databases was not exacting in its definition of “pioneer,” as it grouped “early entrants” in the same category as “first movers.” All these weaknesses could easily lead one to overstate the market pioneer advantage.

Golder didn’t want to look for new insights under the same old streetlamp. So, he donned the cap of a historian and did a painstaking search for archival evidence in 450 periodicals and 250 books—very untraditional sources in marketing research. As he and his coauthor, Gerard J. Tellis, reported in their paper “Pioneer Advantage: Marketing Legend or Marketing Logic,” which was published in the Journal of Marketing Research in 1993, an examination of 500 brands in 50 product categories revealed that almost half of market pioneers fail, and their mean market share is much lower than previously found in other studies. They also found that early market leaders have “much greater long-term success and enter an average of 13 years after pioneers.”

Seven years later, having won the William F. Odell Award for long-term contribution to the marketing discipline for his Pioneer Advantage paper, Golder wrote the first peer-reviewed journal article on the methodological engine he had advanced. In “Historical Method in Marketing Research with New Evidence on Long-Term Market Share Stability,” Golder not only presented a complete description of the historical method, but also used it to debunk the prevailing literature that said market shares were stable over time.

Fast forward to 2020. Golder had branched out and conducted award-winning research on quality and new product development, while maintaining his stature as a thought leader on the historical method in marketing research. The U.S. Supreme Court was hearing a case on whether the travel site Booking.com had a generic name and was therefore ineligible for trademark protection. The main test of genericness, according to Supreme Court jurisprudence, is whether consumers perceive a term to be generic. Golder, along with Tuck professor Kevin Lane Keller and other marketing scholars, wrote a friend-of-the-court brief to argue that it was “scientifically possible” to collect empirical evidence from print and digital media sources to answer that question—basically using Golder’s historical method to solve a current problem. The court, in an 8-1 decision authored by Associate Justice Ruth Bader Ginsburg, agreed. In Patent and Trademark Office v. Booking.com B.V., 591 U.S.549 (2020), the court affirmed the lower court’s finding that consumers did not perceive Booking.com as a generic term, so therefore it was not generic, while specifically encouraging the use of empirical evidence of the sort Golder mentioned in his brief. “The opinion was a validation that we can use data to find the right answer to hard questions,” Golder says.

More recently, Golder used his historical method to answer a different question: do movie critics lead or follow consumers in evaluating movies? In “Critics’ Conformity to Consumers in Movie Evaluation,” which was published in the Journal of the Academy of Marketing Science in 2022, Golder and his coauthors, Jun Pang and Angela Xia Liu, studied the effect of online consumer ratings on critic ratings, using a dataset of 408 movies released in the U.S. between 2011 and 2014. They found that consumers influenced critics, independent of movie production budget or release strategy. They also found that the “influence increased with the volume of consumer reviews a movie has received and decreased when critics published their reviews in specialized (vs. general) media.” These findings turned the traditional theory of movie criticism on its head, questioning the status of critics as elite tastemakers. “This sets up an interesting tension about what it means to be a thought leader in this environment,” Golder says. “You stop being a leader when people stop following you.”

Last year, three decades after his groundbreaking paper on pioneer advantage, Golder published what he called a “challenging-the-boundaries” article—specifically advocating for a data-driven, rather than theory-driven, paradigm for marketing research. The article is “Learning from Data: An Empirics-First Approach to Relevant Knowledge Generation.” In 2024, Golder and coauthors were honored with the 2023 Shelby D. Hunt/Harold H. Maynard Award. The annual award recognizes the article published in the AMA’s Journal of Marketing that has made the most significant contribution to marketing theory in the previous year. In it, Golder and his coauthors—Marnik Dekimpe, Jake An, Harald van Heerde, Darren Kim, and Joseph Alba—describe the “empirics-first” approach as one that comes to a problem without the preconceptions and strictures of prior literature, letting the data be the guide. Once the data is analyzed, theory can follow (but it doesn’t have to).

It’s not easy to chart a new path on an unobserved phenomenon, compiling unique datasets, rather than extend a path cleared by others. However, for Golder, the labor-intensive data collection process satisfies an urge to unearth new knowledge. “It’s a lot more work,” he says, “but the insights are worth it.”

This story originally appeared in print in the winter 2025 issue of Tuck Today magazine.