Product design approach to the IDFA change
—
What is IDFA?
The Identifier For Advertisers is a unique identifier that Apple assign to a user’s device. It allows third parties (us in this case) to identify a user, without revealing their personal information, we can then deliver customised advertising to the user and measure the effectiveness of our marketing.
What’s the problem?
Apple are changing the way users interact with IDFA. Previously a user had to opt-out, but this is set to change, as Apple introduce a system notification that allows a user to prevent a third party from being able to access their IDFA. Meaning the user now has to opt-in.
The stats; so far the global average for users opting in to IDFA is around 20%, meaning 80% of users can’t be tracked and marketed to.
How are we approaching this?
The biggest limitation and concern we have is that we can only show the IDFA notification to the user once! So picking the right time is crucial. We need to ensure that all of our users will see the notification whilst considering the location and timing of it’s display, so that we can be confident our users have seen enough value in our product in order to opt in.
Given that the whole point of the IDFA notification is that it gives the user control and is a more ethical approach to how we as a third party use their information, we decided that we should also be transparent and honest with our users by letting them know what the benefits are for them and what the outcome of opting in to IDFA will be.
And so the exploration began…
The above flows and designs followed on from an initial kick off and ideation with the team responsible for building IDFA in to the product, deciding what are our most common journeys that generate high amounts of traffic, this would provide us with the ideal moments for when to display the notification.
After that, the designs were taken to the wider product design team, this is so that we could vote on which flow we had the most confidence in producing the best results (in this case, which we thought would produce the most opt ins) and so that I could gather any feedback with regards to the UI or UX.
The flows and confidence score were then presented back to the build team, this is so that the engineers could provide and effort score. This would be used to generate an overall I.C.E. score that would determine the order in which we build the different approaches, as we want to test all of the flows to see which will actually produce the most opt ins, this will be shown when we analyse the data during the feedback stage of our process.
Why did we pick this flow first?
This flow resulted in the highest I.C.E score for the following reasons:
- Impact: Nearly every user (if not all) interacts with this flow at least once a session, this means we can show it to a large amount of traffic.
- Impact: It provided an ‘escape’ for the user without them viewing the IDFA notification, this meant that we would be able to show the IDFA notification to them on their next session. As we can only show it once to a user and we want to ensure that they have all the information they need and they are ready to make an informed decision.
- Confidence: The notification is relevant to this flow, as we use the discount information to help us tailor our marketing to the users.
- Confidence: It provides enough time to surpass if the user is newly signed up before presenting them with a pop up, something that may cause aggravation with the product and reduce user returns.
- Effort: It was relatively cheap to build as we already had a flow we could build on top of and the additions were minor.
Results
We have since released the build shown above and we are monitoring the data; dismissals, opt in rate and user retention. We are pleased to say that currently our approach is performing at a 53% opt in rate, which when compared to the global average of 20% is a huge success!
I think this flow has worked well for a few reasons; the high traffic journey that it is displayed in means that we can surface this to all of our users without being overly disruptive, but also because we sympathised with our audience, students of today have their finger on the pulse when it comes to their data, so by being honest with them and relaying exactly how opting in will impact them, we were able to instill trust and confidence. I also think that providing an “escape” before displaying the notification has helped as we can adhere to the one time display restriction whilst allowing the user to interact with it on their terms and when they are ready.
What’s next?
We’re excited to continue monitoring this first release to make sure it continues to perform just as well, but there is always room for improvement so we will be reflecting on the data feedback to see where we can improve on our current approach.
We will also be looking to test our other ideas, to see whether these have a better outcome than the one currently running.
