iOS 14.5 & Facebook State of the Union
It's happening! iOS 14.5 has officially launched. We're all in the same boat, but here's what we know now and what we're testing to learn more:
Here’s what we know:
Apple’s App Tracking Transparency (ATT) prompt hasn’t been pushed to Facebook for most people yet.
It doesn’t show up immediately when you update to iOS 14.5, and will likely be rolled out over the next three to four weeks in batches as iOS applications start to comply and enable the prompt with updates.
Here’s what you’re going to see:
- Facebook is auto opting people OUT when they update their iOS (not auto opt-in) – This is the official stance, however, we have seen and heard of examples from users who haven’t seen the prompt but have updated to iOS 14.5 and are still opted in for ads.
- Attribution has changed to 7-day click – 28-day click is gone and 1-day view is going to be “modelled.” Until Apple enforces the ATT prompt, advertisers may still be able to access data for all existing attribution windows (28-day click & 1-day view, 7-day click & 1-day view, 1-day click) with the Comparing Windows feature. Note that the Comparing Windows feature will be removed once Apple enforces their ATT prompt and the only way to access old attribution models will be through the API. Currently, many advertisers are only able to access 7-day click attribution data.
- Offsite conversion events will be reported based on the time the conversions occur and not the time of ad impressions – Translation: you’ll no longer have delayed attribution. All purchases are now attributed to the day of purchase, instead of at the time of the first engagement/click. All of your reporting in Ads Manager will be reported on the day of conversion (this will also affect historical data). This will help your data line up with Shopify more closely, but it requires a shift in mindset. Now you need to make sure you’re hitting your ROAS targets on the day. All conversions are showing up when the conversion actually happens and will not be credited to the first touch.
- Custom audiences will start to shrink over time – Exclusions will be less accurate, so look out for past purchasers in your prospecting campaigns.
Our team has seen that Android is outperforming iOS (almost) across the board. This is only the case in U.S. campaigns – worldwide, iOS still performs better. This is noteworthy, but considering the massive 60% market share that iOS enjoys in the U.S., shifting focus away from Apple towards Android is not a viable solution. Our experience has shown and continues to show at scale that targeting all will still outperform at scale but we will be monitoring closely.
Check Google Analytics to see what ratio of your audience has updated to 14.5 so far – this will help in projecting the iOS 14.5 adoption rate of your particular audience.
At Pilothouse we’re in full testing mode. Here are a couple of things we’re doing to monitor iOS 14.5 impacts:
Trio Split Testing by Device
- Three campaigns all in an A/B grouping to ensure no overlap.
- An iOS-only campaign, Android-only campaign, and a both-devices campaign.
- All TOF fully open, CBO with one ad set per campaign. All the same top five ads that have been scaling in our main campaigns over the last few weeks. Social proof wiped from all.
Monitoring within iOS
- A/B test with the same structure as above, but targeting iOS versions 14.4 and under (versus 14.5 and up). Building a campaign for each, including a control of all iOS targeting, same as above.
- This way we can track performance changes over time. CPMs can signal competition rates. It could point to opportunities – building out and targeting specific segmented pockets – or show that open consolidation is still best.
- Additionally, we can monitor audience sizes to determine iOS adoption rates over time.
- Worst case, it doesn’t convert – but we can track CPM changes day over day and/or CPC changes.
Different Attribution Targeting
- We're testing campaign optimizing towards 1-day click against 7-day click
iOS 14.5 Checklist
- Your domain has to be verified, or you won’t be able to run ads.
- You should confirm your eight events – it will default to a standard prioritization but you should review to confirm it matches your business objectives. You’ll be able to define these in Events Manager using Aggregated Event Management. If a user opts out of tracking, we will only see purchase (highest priority event) reported in ads manager. But we’ll only see these events if optimizing for 1-day click (with or without 1-day view). For example, if I am a user who opts out and I make a purchase, you will still see the purchase event fire, but you will not see ATC and I/C events. Potentially, this makes your retargeting pools smaller.
- Double-check your rules. If you didn’t update your rules (previous to the change-over) to this new attribution window, they may not work as intended.
- Execute CAPI integration (recommended, not required) – if you are a Shopify customer this can be done directly in Shopify. CAPI tracks server to server and will help ensure as much data as possible to feed the algorithm.
Other strategies & suggestions:
- Record historical 7-day metrics and Google Analytics last-click revenue.
- Acquire synced email list of purchases.
- Ensure attribution is set up for your accounts.
- Keep an eye on retargeting audiences to see if they’re shrinking.
- Split test 1-day VS. 7-day attribution window.
- Split campaigns by device/os.
We’re all in the same boat. Right now we’re testing like crazy and will get back to you when we have more data to share!