Try searching for
Identifier for Advertisers, or IDFA for short, is a unique, random identifier (device ID) that Apple assigns to every iOS device. An IDFA would be the equivalent of a web cookie, in the sense that it enables advertisers to monitor users’ engagement with their ads, and keep track of their post-install activity.
Much like Google’s cookies, Apple’s IDFA was originally designed to allow advertisers to measure their campaigns’ performance and user journeys.
Unlike browser cookies, however, which have a relatively short lifespan, an IDFA sticks around until a user decides to manually reset or disable it, i.e. turn on Limit Ad Tracking (LAT).
LAT is a privacy feature on Apple devices that has been around since 2016, which allows users to choose whether or not advertisers can receive data about activity generated by their devices.
When turned on, LAT zeros users’ IDFA, which means that instead of a string of numbers — measurement providers receive a string of zeros for a given Device ID.
Post Apple’s ecosystem-shaking move to make iOS more privacy-centric, the App Tracking Transparency Framework, or ATT, puts the option of device ID-based measurement in users’ hands, offering them the choice to opt-in (instead of the traditional opt-out).
Apple’s move towards user privacy dramatically limited advertisers’ ability to rely on IDFA for mobile campaign measurement.
That said, it’s worth mentioning that the IDFA isn’t dead. In fact, the latest statistics put ATT opt-in rates at around 46%.
While Apple offers their own attribution solution called SKAdNetwork (or SKAN, for short), which is a way for advertisers to measure ad impressions and clicks-to-app installs on iOS without infringing on user-level or device-level privacy — it’s far from perfect.
Not having the ability to measure the following poses major challenges for advertisers:
When IDFA is available for advertisers, it’s highly useful for campaign measurement.
For example, when users click a banner ad, play a video, or install an app, ad networks pass their IDFA with the relevant data. And although not all ad networks pass IDFAs to advertisers, but allow them to target specific IDs within their properties — most ad networks do.
IDFAs were also used as a way to detect ad fraud. By monitoring their stream of ad actions and their associated IDFAs, advertisers were able to get important signals that flagged fraudulent activities. Nowadays, the majority of anti-fraud logic relies on statistical, large scale thresholds that rarely rely on individual device IDs.
Simply put, not being able to access IDFAs as freely as before, impacts advertisers’ ability to engage their audiences, create personalized experiences, and measure their campaign effectiveness.
Some of the ways in which advertisers and publishers are impacted by the IDFA scarcity include:
First introduced by Apple in 2018, the SKAdNetwork framework is a type of direct install attribution. It was developed to provide privacy-preserving, accurate attribution for iOS campaigns to advertisers — without revealing any user-level or device-specific data.
There are two types of engagements that SKAdNetwork registers:
Since we’re living in a highly dynamic ecosystem that never stops evolving, let’s explore a few of the key changes brought about by iOS 15, SKAN 4.0, and how you can make the most of both as an advertiser.
Before iOS 15, SKAN postbacks were sent only to ad networks, and then forwarded to the advertiser or MMP on the advertiser’s behalf.
Starting with iOS 15 and moving forward, however, advertisers are given direct access to SKAdNetwork postbacks, and a much-deserved ownership of their data.
By configuring their MMP as an endpoint, the MMP:
In WWCD 2022, Apple announced the next version of SKAN, which introduces significant changes that will allow advertisers and ad networks to measure more while maintaining user privacy.
Here’s a recap of the four most fundamental updates and what each one actually means:
Hierarchical source identifiers also adhere to Apple’s privacy threshold — meaning the higher the level of crowd anonymity, the higher the level of granularity would be.
The aftershocks created by Apple’s ATT and the subsequent decline of the IDFA, continue to generate major challenges for advertisers and ad networks alike.
When comparing pre-ATT to post-ATT market share (H1 2020 & H1 2022 respectively) using AppsFlyer’s Single Source of Truth (SSOT) logic, which combines and dedupes data from SKAN and AppsFlyer’s attribution — we can see that Meta lost a significant share of its iOS install market share.
Google, in turn, has also lost considerable ground. And although both self-reporting network giants were negatively impacted, ATT has evidently been much more painful for Meta, based on the company’s earnings reports and numerous articles in the press.
Back in 2020, the social network dominated iOS, but nowadays the company is nowhere near where it used to be prior to Apple’s privacy changes in terms of market share.
Before ATT came into effect, Meta was way ahead of the competition with unrivaled user level data-driven optimization. iOS 14 made the social media titan lose that competitive edge, and with SKAN leveling the playing field in some ways — Meta had the most to lose.
For Google, however, iOS is secondary compared to Android, especially compared to the company’s massive web business. Unlike Meta whose main assets are apps (facebook, Instagram, Whatsapp, etc.), Google’s main asset is search — a mobile web environment where the ATT prompt is irrelevant.
As the ecosystem evolves and alternative solutions mature, IDFA is destined to become less and less relevant in the realms of mobile measurement.
It will give way to solutions that involve cohort analysis, data clean rooms, machine learning, predictive analytics, and top-down measurement such as incrementality and media mix modeling (MMM).
To make the most of this new reality, advertisers should also warm up to SKAN measurement and learn how to make it work for them.
Here are a few steps to ensure your business is SKAN-ready:
IDFA is a unique identifier used on Apple devices, allowing advertisers to track data and measure the performance of their campaigns without accessing personal user information. Users can reset or disable it via the Limit Ad Tracking (LAT) feature.
iOS 14 introduced the App Tracking Transparency (ATT) framework, requiring users to opt in for their IDFA to be accessible to advertisers. This shift from opt-out to opt-in significantly limited advertisers’ ability to use IDFA for campaign measurement, despite an average global opt-in rate of 46%.
Advertisers struggle with measurement accuracy, targeting, personalization, and detecting ad fraud due to restricted access to IDFA. These limitations affect their ability to engage audiences and measure campaign effectiveness comprehensively.
SKAdNetwork (SKAN) is Apple’s alternative attribution framework, which is designed to measure ad campaign performance in a privacy-centric way. While it offers a solution for attribution without user or device-level data, it comes with its own set of challenges, such as delayed postbacks and a lack of granularity.
Post-ATT, Meta (Facebook) and Google have seen a significant decline in their iOS market share, with Meta experiencing a more pronounced impact. The loss of access to granular user-level data has diminished their competitive edge in ad optimization and targeting.
Advertisers are exploring alternatives like cohort analysis, data clean rooms, machine learning, and incrementality testing to adapt to the post-IDFA landscape. Embracing SKAdNetwork and working with mobile measurement partners (MMPs) are key steps for advertisers to optimize their campaigns in this new environment.
An IDFA is a unique device ID that Apple assigns to every iOS device, which enables advertisers to monitor users’ engagement with their ads, and keep track of their post-install activity.
Given Apple’s ATT framework and the severe limitation of IDFA post iOS 14:
Measurement – advertisers are struggling to measure LTV and ROI in SKAN, are lacking granularity, and the ability to track conversion has become more limited.
Targeting & remarketing – advertisers lost the ability to identify individual users, unless they’ve actively opted in, and are unable to plan, measure, and optimize their iOS remarketing activities.
Fraud – advertisers are dealing with increasing levels of ad fraud in SKAdNetwork.
Personalization – advertisers lost some of their ability to personalize, run effective A/B testing, and continuously optimize their creative messaging.
Apple’s announcement of SKAN 4.0, has introduced significant changes that will allow advertisers and ad networks to measure more while maintaining user privacy. These updates will help advertisers measure more despite the loss of IDFA.
As the ecosystem evolves and alternative solutions mature, IDFA is destined to become less and less relevant in the realms of mobile measurement.
It will give way to solutions that involve cohort analysis, data clean rooms, machine learning, predictive analytics, and top-down measurement — giving advertisers the tools they need to plan, measure, and optimize their campaigns, while preserving their users’ privacy in full.
