The Rise and Fall of the UDID
The UDID was never supposed to be such a prominent player in mobile marketing. But as the iPhone’s popularity soared, advertising dollars began to flow into mobile ads, and with them came demand for accurate, “closed-loop” attribution: figuring out who clicked which ads, and what they did after they clicked.
Unfortunately, tracking options were limited. Familiar tracking methods from the desktop – tracking pixels and cookies – were unavailable. Mobile alternatives like HTML5 and digital fingerprinting have their uses, but they’re simply not as accurate when it comes to connecting specific ad clicks with later actions.
The UDID was designed to let developers reliably and permanently identify specific devices, primarily to store data remotely, such as high scores, settings and preferences, or account information. But more importantly to advertisers, it provided a surefire way of attributing ad clicks and targeting ads. Advertisers flocked to it and made it the de facto standard for iOS advertising tracking.
Until, that is, the reliable and permanent nature of the UDID became a liability. As far back as 2009, privacy advocates began to raise concerns about the UDID due to users’ inability to opt out of its use or delete it.
In 2011, Apple announced that it would be deprecating the UDID for advertising; in September 2012, it released its replacement as part of iOS 6: the Advertising Identifier, which provides the same functionality with better privacy options.
And then, inevitably, came the announcement:
It’s been a long, well-publicized transition, enabling most app marketers to begin to implement other tracking solutions. Even the May 1 cutoff date doesn’t mean advertisers have to stop using UDIDs: apps already approved in the app store can continue to use the UDID for now. But the Advertising Identifier – often abbreviated IDFA after its original name, the Identifier for Advertisers – promises comparable utility with much less potential for abuse.
IDFA gets it right from the start
In some ways, the IDFA is very similar to the UDID. A long, indecipherable string of characters, it’s tied to a specific phone and can be used to serve ads that marketers think will be relevant to a user. Like the UDID, it can be used by developers and advertisers to keep track of specific phones for all the same uses.
However, that’s where the similarities end. In contrast to UDIDs, the IDFA was purpose-built for ad tracking. It doesn’t carry any personally identifiable information (PII), and offers users the chance to opt out from its use for ad tracking entirely, addressing the privacy concern most commonly associated with the UDID.
Like any transition in a technology standard, the shift from UDIDs to IDFAs may cause some confusion. However, Apple’s significant advance warning has given the industry considerable time to consider the situation and make their plans.
The key next steps are clear: all iOS app developers should remove UDID tracking code from their next update, and add IDFA support at the same time. Ad networks and publishers are rapidly updating their infrastructure to support the IDFA, and while it took a while to gain traction, the transition is well underway.
Dropping UDID support does mean that advertisers won’t be able to target the roughly 11 percent of iOS traffic running iOS 5 or earlier, but there’s nothing to be done about that other than bite the bullet and adjust your expectations.
Fiksu’s long-time recommendation has been that app marketers should support multiple tracking options, which offers several benefits:
- In the rapidly-evolving mobile marketplace, it makes sense to distribute your risk.
- The various tracking technologies have different strengths and weaknesses.
- Different advertising networks and traffic sources require different tracking technologies, so accessing with widest range of media means using multiple methods for tracking.
Getting noticed among 800,000 other apps in the App Store isn’t easy. Don’t limit your chances by choosing the wrong tracking technology.
No comments yet