Television advertising could be more effective if it targets non-brand and light brand users rather than existing heavy buyers, according to a study published in the Journal of Advertising Research (JAR).

Henry Assael, Masakazu Ishihara (both from the Stern School of Business, New York University) and Baek Jung Kim (Sauder School of Business, University of British Columbia) were the authors of this research.

Their study assessed the sales impact of a TV campaign for a chocolate-candy brand that ran from March 28 to June 16, 2016. The dataset covered a total of 453,327 households.

And their main overall finding was that, for this brand, “a broader-based targeting to light brand users and nonusers is more effective than a more concentrated targeting to the heavy quarter or heavy tenth of brand users”.

Put another way, the results suggested “the causal effect of television advertisements for heavier brand purchasers is less effective than for nonbrand and lighter brand purchasers”.

The study was entitled Accounting for causality when measuring sales lift from television advertising: Television campaigns are shown to be more effective for lighter brand users.

Among the factors considered by the analysis were if a household was exposed to an ad during the campaign, if they purchased the product in question, and pricing factors.

It also addressed different buyer levels: the top 50%, who bought the brand at least twice in the year before the campaign, the top 25%, who did so at least three times, and the top 10%, who did so six times or more.

Breaking out such categories helped assess variations in consumer characteristics that affect advertising responses (or “heterogeneity”).

Alongside that consideration, the study incorporated the impact of contributors other than advertising relating to sales (known as “endogeneity”) to achieve greater precision.

The process of matching the results of the campaign with pre-campaign sales data at the granular level, the authors proposed, was fundamental in determining the impact of advertising.

Without “matching for pre-campaign purchases, advertising effects in the campaign period become insignificant,” the study asserted.

“This demonstrates the importance of matching by purchase predispositions to account for advertising effects in evaluating the advertising-to-sales relationship.”

Another granular learning from the study was that identifying the impact of factors other than ads while looking at consumer characteristics helped avoid the “incorrect insight” that ads to heavier brand users are more impactful.

Sourced from Journal of Advertising Research; additional content by WARC staff

Sourced from Journal of Advertising Research; additional content by WARC staff