Daniel Starch stopped people on the street to ask if they had read the ads.
In the early twentieth century, companies spent money on advertising with no systematic way to measure whether it worked. Daniel Starch, a psychologist at Harvard, developed one of the first standardized methods for measuring advertising readership in the 1920s. His "recognition method" involved showing magazine advertisements to readers and asking whether they had seen, read, or remembered each one.1
Around the same time, Arthur Nielsen Sr. founded the A.C. Nielsen Company in 1923 in Chicago, initially offering performance surveys for industrial equipment before shifting to consumer research.2 George Gallup, at Northwestern University, developed survey sampling techniques in the 1930s that would become foundational to both political polling and consumer research.
These early practitioners formalized the idea that consumer behavior could be measured, categorized, and predicted. Before them, a company's understanding of its customers relied on the intuition of salesmen and owners who dealt with buyers directly. The market researcher introduced a new premise: that opinions, preferences, and habits could be extracted from a sample of people and generalized to an entire population.3
Nielsen's consumer index, launched in 1933, tracked retail purchases across a panel of stores, giving manufacturers their first view of what was actually selling where.4 His company later extended the same panel methodology to radio and then television audiences, producing the Nielsen ratings that determined which programs survived and which were canceled.
The profession created a permanent intermediary between companies and the people they sold to. The buyer's preferences, once communicated face to face, were now data points in a report. The market researcher's job was to translate human behavior into numbers that could justify corporate decisions.5