Mobile Targeting: Buying Audiences Without UDIDs

Written on
Sep 21, 2011 
Krishna Subramanian  |

ADOTAS – Apple has quickly and undoubtedly become the ultimate showman in today’s fast-paced technology theme park. Product launches leave captivated crowds in awestruck silence and amazement. Magical technology advances change the way we listen to music, talk on the phone and interact with the world.

Of course, along the way there are victims of success that Steve Jobs and his crew have stepped on along the way to the top. Sometimes these victims are forced to innovate themselves, adapt or shut down.

A few weeks ago Apple sent the advertising world in a frenzy when it announced the deprecation of UDIDs in iOS5. However, confusion remains as to who this decision would really affect.

UDID stands for Unique Device Identifier. Every iPhone, iPod and iPad has a 40 character UDID used by developers to provide personalized preferences for individual users within games, provision apps for users outside of the app store, and is also used across the board for testing and optimization.

By deprecating UDIDs, Apple has given developers a warning that they may or may not remove UDIDs from future releases — as of iOS5 Beta 7, UDIDs still persist and no major changes are expected between now and the official release at the end of next week. For the near future, ad networks and publishers alike will continue to operate in the same fashion and leverage UDIDs for their respective purposes around advertising, scoreboards, etc.

1. Building Rich Profiles and Serving Targeted Ads: In the realm of online advertising, advertisers serve targeted ads by installing cookies on web browsers that measure audience behavior. Cookies installed on an online web browser are not limited to a single website but can anonymously follow a user across multiple websites, thereby building a rich user profile. Users benefit from cookies because they are served more relevant ads, while advertisers benefit by being able to target an audience that is more receptive to their ads.

Nonetheless, the user has the upper hand in this relationship as they can delete cookies at any point. Furthermore, the cookies are not shared by networks or publishers; they are only accessible to the platform that installs the cookies.

There are essentially two ways to serve ads on a mobile device — a) through the mobile web (aka web apps) and b) through mobile apps (aka native apps). If a user utilizes a web app, the website can generally utilize first-party cookies on the mobile phone.

Bump Equity’s Bob Walczak has written about how the uses for  first-party cookies are limited, as it can only be used across a single website. This means that when a user utilizes a mobile browser such as Apple’s Safari to search a website, the information transmitted by the cookie is “sandboxed” or limited to the user visiting the single website and cannot be linked with multiple websites. From an advertising perspective, this creates a very limited profile for serving relevant ads.

On the other hand, when a user downloads an app onto their phone, developers can install cookies in an app or they can transmit the device ID. Developers prefer utilizing the device ID because it is not limited to the specific app. According to Walczak, the device ID differs from a cookie on mobile in two significant ways: a) the device ID can share information with advertisers about all apps that a user has downloaded; and b) this information is shared by all parties who have access to the device ID, enabling advertisers to build a comprehensive user profile for serving relevant ads.

2. Measuring Conversions and Downloads: UDIDs are also utilized as a way to measure the effectiveness of an ad campaign, including conversions and downloads.

Alternatives to the Device ID

Ad networks, exchanges and developers are collaborating to identify other ways to build user profiles for serving relevant ads. Some of the alternatives being discussed in the industry include:

Unique ID’s Intrinsic to Apps: Apple has asked developers to create unique identifiers that work specifically with apps to replace the phone UDID. This allows advertisers to target ads and build profiles within individual apps rather than across apps.

Utilizing a MAC Address: Another alternative is to utilize the MAC address. The MAC address alone could create the same concerns as the UDID but could be a solid short term solution.

Enabling Third-Party Cookies on Mobile: At present, iOS devices default to blocking third-party cookies. One possible solution would be for Apple to set the default on iOS 5 to enable third-party cookies. However, iOS 5 should give users the option to change default settings and delete cookies at any given time.

ODIN: The open device identification number (ODIN) is a number designed for mobile app developers and service providers to uniquely identify their users’ devices in a convenient and interoperable manner that is anonymous, consistent, and secure. ODIN leverages device identifiers in addition to the MAC address.

OFUDID: Open Feint’s single sign system for the game developer platform enables developers to analyze users across multiple games without UDID. The catch here is it requires the app to enable Open Feint’s platform which will be a solution only for mobile gaming.

Device fingerprinting: Blue Cava and Phluant mobile are a new breed of mobile analytics companies that are leveraging device idnetifiers to create anonymous audiences across devices and platforms. This is still early but definitely showing promise.

Targeting is reality. Educating consumers to the benefits of targeting is as important as targeting itself.As audience buying continues to grow for mobile advertisers, an emergence of RTB will become more and more important. RTB allows advertisers to break out of silos on specific apps and reach their specific audiences that will have the highest engagement rates.

Of course, to do that successfully the industry as a whole will need to work together and jump behind a solution that solves both short-term and long-term issues around audience identification. Apple has given the exploding mobile ecosystem a heads up to start looking and the industry has done a solid job of proactively responding.

Krishna Subramanian is a cofounder of the Velti Mobclix Exchange and drives the advertising and marketing strategy for the company. He was also a cofounder and General Manager of New Business for BlueLithium (acquired by Yahoo! for $300 million), the world’s largest behavioral targeted ad network. Subramanian led the development of the company's affiliate network, lead generation platform, contextual and behavioral targeting platforms.

Before than, he launched the first offline/online social network, MingleNow, which focused on mobile content and communication such as group text messaging and mobile video sharing to bridge the gap between the offline and online worlds. In addition, Subramanian was a cofounder of Burrp! (acquired by media conglomerate TV18), a localized review and recommendation internet portal for metropolitan India, where he ran the product team and focused on ad supported content delivery (text, audio, video, WAP decks) to mobile phones. Prior to that, Subramanian founded and ran a successful internet design services and web content development firm for mobile devices, ConceptK, growing the firm to service many Fortune 500 clients.

Reader Comments.

Great post on UDID alternatives. I was looking for something like this. I’ll add a link to your post on my blog (judewa.com).

Posted by Judewa | 11:17 am on September 21, 2011.

Leave a Comment

Add a comment

Tags: , , , , , , , , and
Article Sponsor

More Features