Why is ad testing important?
There are two primary goals for ad testing. The first, and most obvious, is to increase the amount of money you make from your website. The second, and often forgotten, is ensure that your ads aren't having a negative effect on your users.
What does Ezoic test?
We test five things in aggregate
- The position of ads on each page
- The size of each ad
- The colors (background, text color, etc) of each ad
- How many ads to show
- When to show the user ads in their session (first page, later pages, etc)
Why is testing ads so hard on your own?
At first glance, testing ads may appear to be quite simple, however, it's actually really complex if you want to do a good job. Consider the following:
- The revenue from your site will often vary greatly by time of year, time of week, time of day as well as traffic source and geographics. To effectively compare different ad combinations, you need to use split testing and/or account for outside variables.
- Each ad placement affects the performance of other ads on the same page, both for CPC (cost per click) and CTR (click-through rate). So, you must test combinations of ad placements rather than each one separately because even though, for example, an ad at the top of the page, performs well, it may be costing you more money from other ads.
- Effective ad optimization is done per user session, not per ad placement or per page. When you have a fixed number of visitors coming to your site, the goal is to make the most money from them over their lifetime -- not just on a single page. It is for this reason that you must test and measure user experience as well because even though an ad placement earns alot, it may be driving your users away or causing less pageviews which is actually costing you more than it's making you.
What are the best ad positions and sizes?
Interestingly, there is no answer to this question as the optimal placements and sizes is drastically different on each site. For example, on a "quick answer" website such as an dictionary, the pageviews per visit tend to be low. In this scenario, the best ad strategy to place a larger square ad directly under or beside the information that the visitor is looking for (the definition) so the ad is seen immediately when the user is done with their primary activity. On another, high pageview per visit site, the best strategy might be to place a small number of ads on the initial page views in each session, increasing with each pageview.
How many ad combinations do you test?
We test many thousands of combinations, depending on the size of our site. We used advanced statistical methods to predict winners and cluster various tests together to speed up the optimization process.