During the early days of Ad Refresh, several publishers were not taking viewability into account. Instead, they were blindly refreshing their inventory. This resulted in a short-term massive revenue uplift for them. However, this way of running ad refresh is in reality a disaster for their inventory in the long run. As they were refreshing ads that were not necessarily being seen, the overall viewability level of their placements was plummetting. Thus making their overall inventory quality crash.
From a viewability killer to a viewability booster
If we take the example of a placement at the very bottom of a page that is being endlessly refreshed, it is easy to imagine that its viewability will be close to none. This will inevitably lead to issues with advertisers and SSPs. Publishers can implement an ad refresh solution that only refreshes ads that are being viewed to avoid this issue. While this may seem like somewhat of a big constraint that will limit the revenue uplift, it is mandatory to be compliant with industry standards. This will not only protect publishers’ inventories from decreasing, but it will also contribute to improving their viewability while generating additional revenue at the same time.
Since Ad Refresh leads directly to an uplift in terms of impressions, having them based on viewability will generate a pool of highly viewable impressions. These impressions are usually above 90%. They will play an important role in the improvement of the overall quality of the inventory. This is how it is possible to transform Ad Refresh’s initial viewability killer effect into a viewability booster effect. The impact of the improvement very much depends on the ad placements, some will see much bigger uplifts than others. Below is the data from six different ad placements, before and after the activation of the Pubstack ad refresh solution based on viewability:
Ad refresh also works on low viewability placements. The bottom curve of the graph shows an example where the placement is at the very bottom of the page. Before ad refresh, it showed less than 10% viewability. Activating the solution helped generate more viewable impressions, with viewability roughly doubling on this ad placement. Ad refresh will not help to increase the number of impressions. But it will significantly contribute to increasing the quality of the inventory for advertisers and monetization partners.
Again, looking at the in-content ad placement here shows that there is not much of an uplift. This is mostly due to the format: it doesn’t stay long on the screen, so there is not really an opportunity for ad refresh. However, there is a little uplift percentage, increasing from just under 60% to just over 60%. We recommend publishers fine-tune their ad refresh rules based on the nature of the placement. You should not be refreshing a sticky desktop placement the same way as an in-content mobile placement.
When implemented correctly, ad refresh can have great virtues in terms of viewability. Setting up an ad refresh solution that only refreshes in-view ads is the fundamental best practice that will allow the viewability booster effect to occur. It is also important for publishers to be able to closely monitor the effect on each of their ad units. Each of them will react differently to ad refresh. Having access to this type of granular data is a must-have for publishers who want to set up ad refresh without harming their inventory.