Abstract:

 

TURF (Total Unduplicated Reach and Frequency) is a commonly used market research tool based on the media concepts of reach and frequency. In the original application of TURF, media schedulers wanted to maximize the number of people reached by and/or the frequency of individual exposures in a media campaign. In this application, TURF was used to select the optimal set of media elements. When used for market research applications, however, TURF was typically used to find optimal combinations of items within a portfolio, with the goal of reaching as many consumers as possible with at least one item in the portfolio. While TURF was very popular in the late 1990s and early 2000s, its popularity faded as it became apparent that the original TURF algorithms had difficulty running on large datasets in reasonable time. In this report, we review how these computational limitations have been removed with recent advances in discrete mathematics, and demonstrate how these advances help to determine the optimal size and composition of a portfolio.

​

​

​