Announcements, analysis and opinions on industry trends around the mobile programmatic world.
Disclaimer: The data shared on this blog post comes from our forecasting tools and reflects the traffic we get from the SSP we are connected to and the 150+ countries where there are active campaigns. Therefore, it is not intended to reflect overall trends in the market, but only what the mediasmart platform sees.
It’s been three months since iOS14.5 dropped on the advertising world, changing how marketers would track and process user data. While the jury is still out, our team has been tracking the progress of the adoption of Apple’s App Tracking Transparency (ATT), present in iOS 14.5 and subsequent versions of iOS.
Here are our main findings.
The adoption of iOS14.5 has been overall slower than the other iOS14 versions, but the real adoption came with iOS14.6, which has surpassed it already and is close to 70% of all iPhone traffic. What is clear is that iOS14.4 and older versions are quickly declining, and the amount of inventory available with IDFA is decreasing as well. Moreover, we are also seeing a steady although slow decline of inventory with consent (the Do Not Track (DNT) flag off or positive GDPR signal).
We have not seen a major change in the amount of Bid Offers, Bid %, Won % or Fill % we get for iOS devices. In terms of pricing, we have seen a slight increase in CPMs, but not only for iOS inventory with IDFAs but also for Android inventory with GAIDs, as some demand has switched from iOS to Android.
Since its launch on April 26th, we can see that the percentage of iOS 14.5 traffic grew at a much slower pace than the other versions of iOS14, and was easily surpassed by the adoption of iOS14.6.
As you can see on the graph above, the overall percentage of iOS14+ inventory we have been receiving has picked up at 100% by the end of May. On the same line, iOS14+ inventory with IDFA has slowly declined from around 70% to 47% in these last months. The good news, however, is that only considering iOS14.6, the amount of inventory with IDFA went up from 19% to 40%. Find below a graph with the evolution of the percentage of bid offers with IDFA that we are receiving since April.
Even if the percentage of iOS inventory that comes with IDFA keeps decreasing, mainly due to the higher penetration of iOS 14.6. It is still a higher percentage than what we expected at this point. This is for sure influenced by the fact that many app users have not yet been exposed to the ATT signal.
The App Tracking Transparency (ATT) work-frame consists of requesting user authorization to access app-related data for tracking the said user. In the last 3 months, we have not recorded big differences in the volumes we were getting with the different ATT signals, which are: not available, unsupported, not shown yet, restricted, denied, authorized.
Probably, the most interesting fact is that the percentage of users who have denied access to IDFA for the 14.5+ versions went down from 17% to 15%, while the “not available option” grew from 61% to 65%. If we look at the percentage of users who have authorized the publisher to access the IDFA in the iOS 14.5+ version, the bid offers we are receiving with the ‘Authorized’ ATT Signal is 10% of all iPhone inventory. Find below the evolution of the trend divided by geography.
Even if we are seeing positive evolution, as bid offers with the ‘Authorized' ATT signal for iOS14.6+ went from 4.9% to 10.8% (more than double), we feel necessary to call attention to the fact that 70% of the supply we are receiving still does not support the ATT signal from Apple or comes without it.
There is no question that, not only Apple’s ATT but also other privacy regulations (the TCF 2.0 standard for GDPR, for example) have impacted the amount of inventory available with high-precision location signals coming from the GPS. But maybe this is not a bad thing if it finally drives the ecosystem to properly appreciate the need to measure incrementality: attributed visits measured is not what really matters, but what visits (measured or not) have happened because of the campaign, and therefore would not have happened otherwise.
On the Drive-to-store front, less traffic with IDFA mainly means measuring fewer visits attributed to the campaigns in the short term: we see lower inventory volume we can identify, therefore we measure fewer visits. But should the opt-in rates that have been observed so far be sustained, mediasmart’s methodology for calculating incremental metrics still stands. Campaigns will simply need to gather a few more impressions to reach a statistically relevant amount of data to calculate incremental metrics in real-time, and, therefore calculate the real impact of your campaign.
On top of our usual safeguards to ensure we follow regulations and the validation of geolocated information, here are some strategies we are following to maintain the reach and performance of your campaigns in this new environment:
And, if your Proximity Campaigns also target audiences:
Our outlook remains positive as we observe the opt-in rates are better than we initially expected them to be. We have also developed our technology to work with the changes and comply with the regulations. So, don’t fret! All our technology and knowledge is at your disposal to navigate this post-iOS14.5 world, with limited ad tracking, more limited access to geolocation information and different privacy regulations at play.
In the particular case of Proximity campaigns, you are covered! To measure success in Drive-to-store campaigns, mediasmart will use real-time, Incremental Metrics. We cannot stress enough that the performance focus should shift from ‘simple’ Attributed visits to Incremental KPIs since those continue to hold. You will still be able to get the percentage of increase in visits any of your campaigns prompts to physical or digital locations. The changes we are seeing now will lead to the next big shift in our ecosystem, finally bringing into focus the real value an advertising campaign brings: appreciating the importance of measuring incrementality.