Oftentimes, when publishers develop their own ad refresh tool, they struggle to keep track of what is going on precisely. Getting all the granular data that is needed to measure ad refresh performance is difficult. In this case, the publisher was not refreshing some of his ad units, and he had no way of realizing that himself with the data he had access to. Furthermore, the ad refresh timer was set at 40 seconds, meaning that there were not many opportunities for ad refresh to take place. And because the in-house tool was tough to manage, any changes he wanted to bring to the ad refresh rules were time-consuming.
For the test, Pubstack’s solution was refreshing 10% of the inventory and the rest was managed by the publisher himself. Since Pubstack’s product offers full control over ad refresh rules it was easy to adjust the refresh timer to have a slightly more aggressive strategy.
By setting the timer to 20 seconds, we were able to generate a massive revenue uplift that had no impact on the quality of the inventory. Furthermore, with the level of granularity of Pubstack’s data, we could make sure that all the ad units were being refreshed. This allowed the publisher to avoid leaving any revenue on the table.
By adjusting the timer and making sure that the entire inventory was being refreshed, Pubstack’s solution was generating 120% more revenue than the existing solution. Tracking the performances of ad refresh had also been greatly simplified.
Finally, the quality of the refreshes has also improved. Across all ad units, the viewability level for the refreshes with the in-house solution was between 50% and 80%. With Pubstack’s solution, the viewability had risen to above 90% for all ad units.