TURF stands for Total Unduplicated Reach Frequency. It is a statistical technique to help in making certain decisions about assortments, or selecting best combinations of options for maximum reach (for example, while choosing TV channels during media planning). It is especially insightful when we have overlapping data between multiple options and can’t decide with the straightforward answers.
An example to illustrate
For example, you want to open T-shirt stall during your school function. You are given a limited area, and therefore, you can only store limited inventory and display limited colors. What do you do?
You need to select the best selling colors so that your inventory is well managed and you make the most amount of sales. This essentially means, selecting colors liked by the maximum number of people (or lets say, colors with maximum reach). Besides, you have capital only enough to buy upto 3 colors, based on the minimum order quantities.
In order to be better prepared, you float a survey with a Likert scale.
You decide to ask 40 of your classmates in school to rate the 5 color choices on a scale of:
I love it – I like it – Indifferent – I don’t like it – I hate it
So, here are the first look at the responses:
Color | “I love it“ |
Yellow | 50% |
Pink | 45% |
Red | 40% |
Black | 20% |
Blue | 10% |
Yellow seems to have won overwhelmingly, while, Pink and Red seem to be the next choices. On an ordinary day, you would gone with these choices. But not today. Today, you must make a better informed choice.
So, let us delve deeper. Lets dig!
Once you analyse the respondents, you might find that some people who actually chose “I love it” for one of the Yellow, Pink and Red. Lets say, for the sake of this example, this number is 70%.
Moreover, digging deeper might reveal that some of the colors like Black and Blue have very fierce following. You might find that most (lets say 90%) of those who have chosen Black and Blue do not go for any other choices, which translates to 18% people dedicated to Black and 9% to Blue.
So, now you have 2 options:
- Go for the 70% reach for people who like any of Yellow, Pink and Red.
- Go for the 77% reach for people who like any of Yellow, Black and Blue.
As you can see, the reach for the second option is much higher than that of the first option.
Good. This was a simple case where we just looked at the reach. But to make the matters more complicated, there is another factor called frequency, which essentially translates to how many T-shirts one could buy.
So, one would have to again look at another set of data to be able to determine the best possible combination that would get one to the desired maximization.
This kind of analysis was originally used to make decisions on media planning. But it can be branched and used in many other ways. Nowadays, there are quite a few tools that can be used to apply TURF analysis to data.