As an advertiser, it can be challenging to understand when it is time to stop investing in some elements of your campaigns. What if you just need to wait a bit more until you see great results?
How to be sure that the good results you are getting in some elements are not a coincidence but something stable and that you can re-use in your future campaigns?
The Significance functionality helps you overcome those challenges by telling you when the results on a specific KPI are stable enough and have a high probability to repeat so that you can confidently make the best decision for your campaigns.
You can activate or deactivate the Significance analysis with the option highlighted below in your Data table widgets.
Two different values can come up from the Significance analysis:
The KPI is significant
This means the results you are getting at the moment are stable and have a high likelihood of being repeated.
The KPI is not significant
This means the results you are getting are not yet stable and the KPI might change drastically over time.
How can you use this information?
Let's take the example below of the performance analysis by ages in the platform and let's assume my goal is to optimize the cost per click:
For ages 55-64, the cost per click is very high and the result is significant, this means there are few chances that the performance will improve. So you can either pause this segment of age in your campaigns or bring new creatives that resonate better with that audience to improve performance.
For ages 45-54, even though the cost per click is high, the results are not significant yet and the trend can change, the recommendation would be to keep this audience active until there is enough information to determine if this segment is really not working for you.
For ages 25-34 and 35-44, You are reaching your goal + the results are significant, which means, you can include this audience segment confidently in the targeting of your campaigns to achieve your goals.