Total Unduplicated Reach and Frequency (TURF): Optimizing Product Line-ups
The Business Objective
You want to optimize a set of items to reach the maximum number of people possible.
Imagine stocking flavors of ice cream. The grocer decides there is limited space – he can only stock up to 8 flavors of ice cream (out of 30 possible). He wants to maximize the chance that shoppers will find their favorite flavor (or a flavor that meets a certain preference level) in the freezer. When a respondent has his needs met by a flavor in the freezer, the respondent is counted as “reached.” The problem isn’t as simple as including the 8 most preferred flavors. Flavors appealing to population niches (that can increase total reach) would be overlooked.
Total Unduplicated Reach And Frequency (TURF)
The TURF procedure examines a variety of possible subsets of items.
For example, one might examine subsets of 7 or 8 items out of a total of perhaps 20 or 30
For each set, TURF counts how many respondents are “reached.”
The top sets of items that maximize “reach” are listed in the output with the percent of respondents reached shown next to each one.
What Kind Of Data Is Used In TURF?
One could argue that the best data to pull into a TURF analysis is MaxDiff data, however most any kind of data can be used.
Likert scales
100-point ratings
Ranking data
Consideration set data (0/1 per brand)
Magazine readership (0/1 per magazine)
Special Notes For MaxDiff Data
Using MaxDiff data as the input for TURF introduces the ability to conduct simulations similar to conjoint-style market simulations. You select which items are to be made available to respondents (as if they were in competition with one another within a marketplace). The percent of respondents projected to “choose” each item as “best in market” is computed.
Challenges & Advances
One challenge with TURF is that many solutions yield near equal reach. However, this should be viewed as an opportunity rather than a problem. One can bring other information to bear on the decision (such as expert opinion) to help decide which set is best to solve the business problem. In the ice cream example, if one knew that a particular flavor (appearing in many of the top sets) tends to spoil very quickly, such solutions would be avoided in favor of other similar-reach solutions.
Advances in TURF methodology allow for the examination of very large subsets from very large total sets. For more information about whether TURF is right for your needs, contact us at contactus@elucidatenow.com.