Index measures how likely your audience is to have an attribute compared to the baseline, which by default is the total adult online population. It answers the question, "How likely is your audience to have these attributes compared to the average U.S. online population (or to your selected baseline)?"
An index of 100 represents average, meaning your audience is not more or less likely to have that attribute compared to the baseline. If the index number is above 100, people in your audience are more likely to have the attribute compared to the baseline. If the index number is below 100, the attribute is less present among people in your audience compared to the baseline.
For example, you would interpret the insights above as follows:
- Millennial Moms are 62% more likely than the average online adult population to belong to or access/consume information from Pinterest.
- Millennial Moms are 20% less likely than the average online adult population to belong to or access/consume information from Reddit.
As a guideline, we look for attributes with an index of 120 or above to identify attributes that are unique and meaningful to the audience and attributes with an index of 80 or below to identify attributes that are not important or distinctive about the audience (20% more or less likely).
Low-indexing attributes are also worth examining, for they tell you what your audience doesn’t care about. Don’t waste your budget on a creative that won’t click with your audience!
Changing Your Baseline
Index is calculated by dividing your audience composition with your baseline composition. Therefore, changing the baseline means you’re changing the calculation that determines the index.
By default, the baseline is the online adult population. You can view the baseline when you’re building an analysis in the Intelligence Center underneath the audience’s name in the left pane.
In our previous example, Millennial Moms were 62% more likely than the average online adult population to belong to or access/consume information from Pinterest.
Let’s say that you’d like to understand how more or less likely Millennial Moms are to use Pinterest compared to other Moms, instead of compared to the online adult population. You would change your baseline to Moms, and then you would interpret your insight as:
- Millennial Moms are only 3% more likely than all Moms to belong to or access/consume information from Pinterest.