Apple iOS14 New Data Protection Policy Alters Advertising Ecosystem

I wanted to share an important update regarding Apple’s latest iOS14 policy requirements. We’ll go over the impacts it will have on apps in the iOS 14 Apple Store, as well as its effect on the advertising ecosystem as a whole.

Apple has put app publishers and advertisers on notice of impending user privacy changes, and many apps will need redevelopment in order to comply with the policy. Right now, Apple is giving developers a chance to fix their apps before enforcing the policy, but many presume the leniency period will end in January. In this new environment, Apple has provided two approaches to attribution: ATT or SKAdNetwork.

AppTrackingTransparency (ATT) is Apple’s system of gathering IDFA (user data) and managing third-party access. This requires that app developers use Apple’s prompt to opt in, which, if granted, will grant access the user’s IDFA.

SKAdNetwork was released in 2018, and is the second option developers have for attribution. This platform allows ad networks to attribute app installs directly from the App Store without relying on IDFA data. But there’s a catch: It removes user level data entirely and replaces it with consolidated data. When an ad click results in a conversion of an app download, the App Store is notified, and it sends basic attribution parameters to the ad network. However, that attribution data is stripped of any user-identifying information, including ad network ID, campaign ID, and publisher name. How will you be able to determine the effectiveness of your campaigns independently without this attribution data? This is one of a few reasons why many are shying away from relying on SKAdNetwork.

Paul Müller, Co-founder and CTO of the mobile marketing platform Adjust, has written several blogs on the intricacies of iOS 14 and how to navigate them. You can read those here.

What will this look like?

I’m an iOS14 user

From a user perspective, the app store will now help users understand an app’s privacy practices before they download the app. On each app’s product page, a “data nutrition label” allows users to know what data is being collected and how it is (or may be) used. The specific labels are Data Used to Track You, Data Linked to You, and Data Not Linked to You. Each label has a bulleted list of the different data types, like so:

My business has an app

Once an iOS14 user downloads an app, or has an app update, they’ll receive a prompt asking if they’d like to allow tracking. This is on a per-app basis, so a user could opt in on one app and opt out on another. If the user opts in, their Apple-specific user ID will allow data to be shared.

From an app developer perspective, you should be looking at how your mobile software development kits (SDKs) use IDFA and how the CRMs rely on the data. Developers should be planning and preparing a version release using updated SDKs. They will also need to decide if they’ll use Apple’s SKAdNetwork as an additional source of attribution. There are many guidelines and limitations on advertising, and it’s important to become familiar with all of them so you know how they’ll impact business.

Apple’s prompt doesn’t provide context on what it means to allow or limit ad tracking. This means app developers should take the opportunity to build a pre-prompt pop up that essentially asks users to opt in twice. The mobile marketing platform Adjust provides thorough recommendations for preparing and testing app consent messaging and mechanics.

Developers should work through the consent mechanics and prompt messaging in order to achieve healthy opt-in rates. They can test this now in their apps on current users. If an app uses an internal prompt and the user declines, then the app doesn’t need to serve the Apple prompt. However, if a user opts in to the internal prompt, Apple will then serve them their own. If the user agrees to the apple prompt, their IDFA will be shared. You get this opportunity only once (unless the app is reinstalled), so it’s important that your messaging is right in order to ensure a good opt-in rate.

How does this impact the advertising ecosystem?

Apple’s updates will impact businesses that advertise mobile apps, as well as those that optimize, target, and report on web events.

Ad platforms are still able to report out impressions, clicks, and cost data. However, conversions will only represent those users who’ve opted in. This throws off performance metrics like App Installs, Cost Per Lead, Return on Ad Spend, and Conversion Rate.

Audiences used for targeting will also see a decline in size, and developers will need to rebuild ad personalization tools. Retargeting, exclusion targeting, audience segmentation, lookalike audiences, and much more rely on IDFA data.

Restricted, aggregated, and delayed data will impact:

What do major ad platforms have to say about this?

Facebook has been leading the conversation with their disagreement on the policy change. Facebook execs think Apple should have consulted the industry on the impacts before rolling out iOS14. The move has effectively forced all ad platforms to reframe their data collection, as well as the measurements used to determine benchmarks, best practices, performance and ad product functionality. Regardless of their disagreement, Facebook claims that Apple will kick them off the app store if they don’t comply.

Compared to Facebook, Google has been suspiciously quiet on the matter. In February 2021, they released FLOC, which aims to remove third-party tags from Chrome using interest-based audience clustering. However, they didn’t link this release to Apple’s policy requirements. Many strategists recommend utilizing UTM parameters, CRM data, and Google Analytics to try and backfill user insight.

As ad publishers, Facebook and Google have asked users to be patient. Both say that 2021 will bring major changes to the platforms, and to advertising ecosystem as a whole.

Facebook Advertising Impacts

Under this new policy, advertisers can optimize and report on a maximum of 8 conversion events per domain. Conversion events can include purchase, sign up, add to cart, page view, etc. This will heavily impact domains with multi-market presence.

The domain structure will determine how many conversions are allowed. If domains are formatted as Region (dot) Brand or (dot) Com (slash) Region (see chart below), they are considered a single domain. There are only 8 events allows per domain, not per pixel. That means that if a business has multiple agencies, and thus multiple Facebook pixels, the business will need to prioritize their top 8 events to optimize toward across those agencies.

Campaigns optimizing to more than 8 events per domain will be paused. Facebook does not have a solution, calling on all advertisers to help brainstorm ideas. You can still use nine or more events for targeting (e.g. when you’re building custom event audiences), but you are limited to 8 events to optimize and report on.

Facebook will only receive information on one event completed per user. Advertisers must choose what that top event is — for many it will be purchases. As an example, if a user Adds to Cart and Purchases, the event will be counted under Purchase and not included in Add to Cart. Facebook is currently working on a tool in Business Settings that will use historic data to help businesses determine how they should rank conversion events. If/when events are changed, it will take 3 days to get reporting in order to allow the system to attribute properly.

Two actions each Facebook advertiser should take in early January:

• If you’re using the Facebook SDK, update to the latest SDK to version 8.1 or above. OR if you’re using the Facebook pixel, verify your domain in Business Manager.
• Select and rank your top 8 conversion events per domain that you’d like to optimize to.

Begin preparing:

• Attribution window changes are coming. Update automated rules if necessary.
  • 28-day click and view-through attribution will no longer exist.
  • 7-day view-through will no longer exist.
  • 7-day click attribution will continue.
• Identify campaigns and bidding strategies that may be impacted by the attribution windows or event limitations.
• Identify campaigns using tROAS and tCPL bidding strategies for potential migration post-enforcement.

Wheelhouse Is Here to Help!

If you have any further questions about iOS4 privacy changes, let us know in the comments. We’ll do our best to get back to you ASAP, and may even update the post with an answer. You can also reach out any time if you’re interested in having our Digital Advertising & Analytics teams help with any tracking, privacy or advertising challenges your team is facing.