An IDFA Recap
Apple’s deprecation of the IDFA is the consequence of increasing consumer data concerns and awareness. As consumer expectations gravitate towards heightened data privacy, Apple is strategically pivoting to these growing expectations. Recently, Apple allowed users to opt out of data sharing (though many users have not been made aware of this update – approximately 25% of users have opted out), and last year it strengthened its ITP (Intelligent Tracking Prevention) module. However, the deprecation of the IDFA is potentially the most momentous change so far, and could redefine the app market in 2021. Furthermore, if implemented, Google’s Android platform is likely to follow suit as Android’s app policy has tended to evolve parallel to Apple’s iOS.
Apple has made their SKAdNetwork available as an alternative solution. This is an attribution solution for iOS14 that does not allow for user-level campaigns or access to channel attribution data. Instead, they will provide reports on anonymous campaign level installs, source-app and some down-funnel events. Revenue and in-app events will also be significantly limited. This is intended to ensure no back-engineering of the SKAdNetwork solution that could allow for advertisers to directly attribute a user to a campaign.
In response to an outcry from developers, marketers, and advertisers alike, Apple has delayed the implementation of IDFA deprecation until sometime in early 2021. This gives firms time to adjust their advertising strategy and adapt to the newer SKAdNetwork framework, as well as prepare for the various potential impacts of this change to their business.
The following are the most significant IDFA-related changes that will arrive in 2021:
- The need for users to opt in to the IDFA
- Anonymisation of user data by SKAdNetwork
- Apps must have a tailored breakdown of their privacy practices, and feature a link to their full privacy policy on their app store page
How Can you Prepare for the IDFA Opt-In?
In such a radically evolving app ecosystem, consultation and holistic data-enabled insights are more critical than ever for sustained success.
There are a number of important ways that advertisers can adapt to the new app market with decreased IDFA reliance.
- Convincing consumers to opt in with user-friendly consent mechanisms
- Increasing dependence on a probabilistic model of attribution using historical data instead of deterministic models due to the SKAdNetwork only providing advertisers with one user parameter to personalise ads (keep reading for a full break down on what this means)
- Curating in-app experiences with predictive power – generating data through alternative means will be crucial to gaining an edge in the market
There is some concern that many firms will simply delay adjusting fully to the SKAdNetwork until early 2021, rushing to prepare just before the IDFA is deprecated in order to maximise their benefit from the current attribution system. This may result in a number of firms being unprepared for the final deadline for complete adjustment, losing valuable time to optimise and effectively build their new systems for attribution.
Probabilistic Attribution & Limited Personalisation
What is Deterministic Attribution?
Deterministic attribution is currently used by many Mobile Measurement Platforms to exactly match an app install to an advertising campaign at the user level. This means that there is a 100% probability that the install can be attributed to a specific advertising campaign. This is because the IDFA of a user viewing an ad is the same IDFA when a user opens the app.
Opt-In or Opt-Out: Meta-Analysis
So how many users will opt-in or opt-out? This question is crucial to determining the right strategic mixture to use after IDFA deprecation. Although varying, studies appear to have converged on some useful approximations:
- ~20-30% will consent to the IDFA without asking for more information.
- ~25-35% will consider consenting to the IDFA if given a compelling case.
- This means that 45-70% may consent to the IDFA.
- The remaining 35-55% will not give any consent to the IDFA even if given a compelling case.
These are large ranges compiled from a diverse array of sources (including SEL, AdColony, Adikteev, MobileDev, BidSwitch) yet they paint a general picture of to what extent the deprecation of the IDFA will affect the app industry. Optimistically, up to 65% of users could be persuaded to opt in to the IDFA with suitable incentives or explanations. Nevertheless, firms should prepare for a world in which the IDFA’s power is significantly reduced, as it is possible that few of those considering opting in actually will.
What is Probabilistic Attribution?
Probabilistic attribution on the other hand will be used when a user has not opted-in to IDFA. Due to this it is not possible to assign a 100% probability that a specific campaign drove an install, instead there is a set of probabilities that the install was driven by one or more campaigns. Probabilistic attribution does not use any personally identifiable information about a user.
It is increasingly evident that encouraging consumers to opt-in is critical to the effective implementation of more accurate deterministic attribution processes. Although it is unclear to what extent Apple will allow for alternative attribution processes, they have already signalled that there will be no space for loopholes and thus probabilistic attribution will become significantly more important.
IDFA deprecation will certainly result in a decrease in the amount of definitely attributed data to advertisers, as the substitute SKAdNetwork anonymises user data and only provides advertisers with a single parameter to analyse.
Anonymised data necessitates the application of probabilistic attribution methods, which use historical consumer data and algorithmic tools. As these are less accurate (generally 60%-70% accurate), it will be more difficult to reliably match consumers with attributes.
This makes ad personalisation and frequency capping (capping the number of times an ad is served to a single user) significantly more difficult, and will negatively affect acquisition channels. These developments are forecast to adversely affect advertisers’ revenue streams, meaning that creating alternate channels for extracting user data with predictive power are integral to remaining successful in the market.
SKAdNetwork Conversion Value
However, Apple has provided Conversion Value within the SKAdNetwork framework, which will help power probabilistic attribution. To protect user privacy, Apple restricted the amount of post-install data that could be reported against an SKAdNetwork campaign. Conversion Value is a single value that is reported against a campaign ID as opposed to an install ID and therefore remains anonymous. It should be sent by the app to Apple once the app developer has collected enough information about an app install to predict the lifetime value (LTV). Whilst Conversion Value can be used for metrics such as revenue and retention, Apple has recommended using predicted LTV as it encompasses all other possible uses for Conversion Value.
Adapting Ahead of iOS 14 and IDFA Deprecation
Although IDFA deprecation certainly poses a myriad of challenges to those in the iOS app marketing ecosystem, there are a number of ways in which firms can adapt.
Alternative Identifiers
Some advertising partners have access to a wealth of other consumer data that can be used as identifiers, such as email addresses and loyalty card information. Such alternative identifiers have powerful potential in a post-IDFA world.
However, it is possible that Apple may soon severely restrict advertisers’ access to this data. In its AppTrackingTransparency network (an AppTrackingTransparency, or ATT, popup is the push notification that asks for users’ permission to access the device IDFA) Apple appears to show intent to restrict data leakage beyond the IDFA. Furthermore, many of these alternative identifiers seem to fall under an EU court’s definition of data that users must give permission to share. In light of these factors, it is important for firms to also seek other ways of adjusting to 2021’s new app market.
User-Friendly Consent Mechanisms
In a future app market where the user default will be opting out of the IDFA, it is more critical than ever to effectively convince users to opt-in to IDFA and provide justification for its use. This is a vital opportunity to capitalise on the potential IDFA attribution of the estimated 38% of users that would consider opting in provided they are given a compelling case for doing so.
There are a number of user-friendly options that could be used to make users more amenable to opting in to the IDFA. Some potential methods include:
- Emphasizing the importance of consumer privacy to the user through the consent form
- Re-stating that the user has complete control over their data
- Highlighting that opting in results in increased ad relevance
- Underlining the central importance of these synergies in allowing the app to continue to effectively provide its services.
These techniques build trust between the user and app, fostering a closer relationship between the two. This may not only drive the user to opt-in to the IDFA, but also make them more open to permitting other data sharing options or engaging in optional in-app experiences that enable a more personalised user experience.
Using an innovative and holistic marketing strategy is integral to not only effectively communicating the need for users to opt-in but also ensuring you are asking at the right time. If done correctly, this could result in over 60% of users opting in, minimising the negative impacts of IDFA deprecation, and potentially building closer and mutually beneficial consumer-advertiser relationships.
Summary
Despite the uncertainty surrounding the IDFA deprecation’s long-term impact, it is clear that it will seriously affect IDFA opt-in rates and reduce the quantity and quality of consumer data available to advertisers. In light of this, it is critical to consider developing a sophisticated and accurate probabilistic framework for attribution, create user-friendly consent mechanisms, implement effective incentives to compel users to opt in to the IDFA, and seek out alternative identifiers that can be used.
An innovative app marketing strategy enabled by in-depth data analysis is integral to maintaining an edge in the iOS app market. As the existing routes to tracking data become increasingly limited, businesses will have to look to incorporate new measures to ethically collate data in line with regulations and establish creative strategies that build user trust and increase long-term user engagement as a result.
If you have worries or need support regarding the future of attribution and iOS 14, please get in touch with our growth team.
Make sure to sign up to our newsletter to get notified when we release a new blog post. Want to find out more about optimising your app and keeping up with the latest OS capabilities? Make sure to subscribe to our Mastering Mobile Marketing video series. You can also get in touch by visiting the Contact Us page. Follow us on LinkedIn, chat with us on Twitter @yodelmobile, and join our #AppMarketingUK LinkedIn group.
