Opt-In Design Do’s & Don’ts for Apple’s App Tracking Transparency (ATT) on iOS 14
New guidelines on iOS 14 have emerged since this post was published, meaning some of the information in this article may be outdated. Please reach out to your dedicated Adjust rep, or firstname.lastname@example.org, for the latest guidance on iOS 14.
Since Apple’s announcement of the App Tracking Transparency (ATT) framework in 2020, we at Adjust have worked diligently with clients worldwide to prepare for the change that lies ahead. From understanding how to best embrace the ATT framework to aiding in user testing, we have worked closely with our partners to chart a path forward together. We have learned a lot about what steps to take to improve opt-in rates — and what steps to avoid.
In this post, we have compiled critical learnings from our partners that offer you a crash course into the do’s and don’ts of using design to improve your opt-in rates. And if you and your app are keen on being a part of this testing phase, we are still speaking with clients and welcome you to join — simply reach out to your Adjust representative.
With that said, we start with a cautionary tale.
Beware of Dark Patterns
London-based UX Designer, Harry Brignull, coined the term “Dark Patterns'' to define instances where designers use their understanding of human behavior to influence end-user decisions. More specifically, they use design psychology to subconsciously coerce these users into doing things that may not be in their best interest. Types of dark patterns to influence user behavior include:
It’s important to note that the utilization of Dark Patterns is not always intentional, and many companies are making a conscious effort to avoid them. When experimenting with your ATT implementation, you should do your best to identify and avoid the use of these patterns. If you would like to dive deeper into Dark Patterns, we encourage you to check out this research paper on the topic.
In some conversations we’ve had with app developers, a few have reported that their app was rejected from the App Store before they could begin testing their ATT implementation. Based on their feedback, we’ve collected three patterns that have resulted in App Store rejection thus far:
In the context of Dark Patterns, nagging is a repeated intrusion where the end-user is interrupted one or more times by another task not directly related to the one they are focusing on.
One developer made use of a pre-permission prompt before triggering the ATT consent popup. If a user declined to consent to the ATT popup, the app would trigger a custom prompt again, either later in the same session or a subsequent session, with a deep link that leads into the end user’s privacy settings. Once there, users would be able to change their previous response to the IDFA pop-up. Another developer informed us that their app was similarly rejected for repeating the popup after the end user’s decision was made.
In response to their rejections, Apple viewed these implementations as “nagging” and responded:
"The permission request process on iOS is designed to give users control of their personal information. It is important to respect the user's decisions about how they want their data used. If they decide to not grant permission to your app, they should not be prompted to change their mind or be forced to reject the request multiple times."
Sneaking is defined as an attempt to hide, disguise, or delay divulging information relevant to the user. Sneaking often occurs in order to make the user perform an action they may have objected to if they knew of it.
Another app saw itself rejected after triggering the ATT pop-up right after the user agreed to the main consent GDPR banner. Within the context of this proximity and timing, Apple believed that this specific positioning would influence the user’s decision to opt-in to the ATT.
Interface interference is defined as any manipulation of the user interface that privileges speciﬁc actions over others, thereby confusing the user or limiting discoverability of important action possibilities.
In this instance, the app developer triggered their pre-permission prompt and displayed visual queues alongside it that may subconsciously influence the end user. By positioning a thumbs up emoji near the ‘allow’ button, it was seen as interference when it came to the end-user making their own, unbiased, choice to consent to tracking.
Additionally, the pre-permission prompt should not act as a consent prompt on its own. While this was not the actual IDFA pop-up, Apple maintains that pre-prompts should serve only to educate the end user about their decision and not influence the action they take.
In the above instances, the app publishers with whom we were in contact were all able to correct the highlighted issues and get back into the store. Now that we know what steps to consciously avoid, let’s dive into the designs that have yielded more favorable results.
Common trends and patterns
In the course of working with clients from a range of verticals, we noticed there are common trends that influence users’ response to Apple’s ATT pop-up. Here, we’ll look more closely at the variables that have the biggest impact.
Pinpointing the precise moment in the user journey to serve your opt-in request is arguably the most important factor to consider. From the testing results we’ve seen, the most successful prompts are displayed during the onboarding flow.
Include in the onboarding flow
Serving a pre-permission prompt as part of your onboarding flow helps to prepare users. It can act as a buffer, so they are not overwhelmed when the Apple pop-up appears. As they move through the onboarding screens, data privacy can be addressed as a topic on its own screen. Here, you can address the topic at a broad level, also including your GDPR messaging. This way, the user is not taken by surprise and instead understands the consent request in the same way as asking for the personal information; a non-intrusive formality.
Using a pre-permission prompt within your onboarding flow lets you serve the Apple pop-up in a way that feels natural. We have seen some very promising responses to this approach, with opt-in rates reaching 65%. However, remember that this approach can only be fully realised with new users.
Pre-permission prompts that cue the ATT framework and emphasize the value users can gain from opting-in have typically performed well. However, you need to toe the line here between delivering value without creating a content overload. We recommend staying within the range of 2 - 3 short and succinct sentences. As always, follow best practices for UI copy: opt to use the fewest words to say the most, and choose the easiest words to read and understand.
What was apparent from the tests was that users are particularly looking for direct, personal value. This means demonstrating what content your app offers to them specifically. In this respect, it is important to note that social comparison did not perform as well.
Fairly consistently across all tests, we have seen that users respond better to pre-permission prompts that are full-screen rather than modals. The reasoning for this is that the full-screen prompt has a more seamless user experience, especially when it is designed to align with the specific screen it is served after. By contrast, modal screens feel like an interruption and are reminiscent of unsolicited pop-ups and ads. This may be a reason for them seeing far lower opt-in rates across the board.
The placement and design of your call-to-action buttons can be treated as low-hanging fruit. Although this does not contribute as strongly to opt-in rates as the other trends discussed here, both academic research into GDPR notices and anecdotal evidence from iOS 14 testing has offered largely aligned results.
Using simple terms such as “Later” and “Next” horizontally next to one another, and in that order, prompts users to take an affirmative action. Using the positive acceptance on the right hand sign aligns with how we understand moving on to a next step.
In contrast, vertically stacking your buttons forces readers to slow down and more carefully pick their choice. This does not drive user engagement, and it can frustrate users, as it is more difficult to reach this part of the screen.
The view ahead
Whether you have already conducted multiple tests or are just getting started now, there’s still time to devise an opt-in strategy. We encourage all app developers to perform tests to help benchmark their own user performance. However, if you don’t have the time or resources for this, you can still utilize the trends and learnings we’ve summarised here. Although there is no way to guarantee attaining a certain opt-in rate, we will continue to support you with these continued insights and best practices.
To stay updated with all of the latest news and learnings for all things iOS14, be sure to check our blog regularly. If you’re interested in having a 1:1 discussion with our UX experts or participating in a roundtable with other businesses looking to discuss their preparation strategy, reach out to your Adjust rep or send an email to email@example.com.