 |
|
|
|
||
|
|
The hunt for a cookieless targeting solution marches on. One promising candidate is targeting based on first-party mobile app data, which users willingly enter when setting up an app. User-generated data has the advantage of being more reliable than third-party data, and avoid concerns surrounding consumer privacy, since users consent when they submit it.
Seattle-based startup nFluence is all about user-generated mobile data. It’s developed a solution for app developers called Autograph that builds consumer profiles based on preferences they enter.
It works like this: a user is shown a scrollable row of 40 tiles, each with a consumer brand logo. The user swipes up to say they like the brand, or down to say they don’t. After they’ve “swoted” (swipe-voted) all of the tiles, the app cross-references their choices against a list of 5,500 attributes – things like “you like high fashion” and “you like technology” – and offers about 20 that it determines describe the user. The user gets to see their attributes and un-check ones that Autograph got wrong. The whole process takes under a minute.
The tech can be integrated into any mobile app. It can be used by brands to survey their audience or by retailers to offer automated personalized promotions. So far it’s been used in trials by Westfield Malls UK, which gets users to swote store brands that they can find at the mall. After they’re done, they see a “wall of content” with ads for movies or sales they’d be interested in.
Autograph generates a detailed profile surprisingly quickly from only a few interactions. At the speed that users are swoting, it’s not even really a survey – which is the point. nFluence wants users to swipe without thinking.
“We make our decisions not with our minds, but with our gut,” says nFluence director of marketing Arnel Levya. “Most people think they know what they want, but they don’t really know what they need. So we’re getting past that whole thought process to where we’re able to take your emotional responses and do some really cool math with it.”
The app’s visual design draws on modern research in behavioural psychology that shows most human thinking happens in the automatic, intuitive part of our brains. Our gut reactions towards brands tell us how we really feel about them, and which ones we’ll lean towards even when we’re not aware we have a preference. To provoke a gut reaction, rather than a considered response, Autograph presents only the brand image, without text or other information.
Using Westfield UK's Autograph-enabled app, users "swote" brands up or down.
Once they've swoted their way through, they see a wall of personalized content.
The company’s psychology-driven approach has drawn the interest of Ogilvy Labs UK, which featured Autograph at a recent conference and has developed a strategic partnership to introduce the technology to its clients. It’s starting to get attention from UK advertisers, including a large telecom company that the company is keeping anonymous for the time being.
The larger question that Autograph raises is whether the technology could be applied on a broader scale – not just to individual mobile apps, but to publisher networks and programmatically purchased ad impressions. The incredible amount of detailed, accurate, first party targeting data Autograph generates is a digital advertiser’s dream. And it comes without any of the usual caveats about privacy, since the user offers it freely.
There are barriers preventing Autograph’s current incarnation from being applied to publisher networks. For one, Autograph is built on an anonymous user ID. The way nFluence sees it, the data has to be anonymous, because it wants to get away from personal tracking and other creepy data collection practices that would discourage users from participating.
“The mobile device is the most personal device of all modern devices,” says Levya. “Any information that advertisers or anyone else pushes, you really have to do it a way where people feel safe, and you’re upholding the sanctity of the relationship they have with their mobile device.”
An anonymous ID is certainly fine for a native app, but it’s hard to see it working for a big open mobile exchange, delivering ads across apps and pages all over the mobile web. On the other hand, it may be a promising technology for large individual publishers that already use active sign-ins, like Microsoft or AOL. Think of an Autograph preference pane on the Toronto Star or TSN mobile app, helping deliver targeted content and advertising. Many of these publishers already use content recommendation engines like Taboola or Outbrain, but by adding an easy-to-use, privacy-protected personalization feature, they could certainly grow their value to consumers and advertisers.
Levya presents Autograph as an alternative to buying user data from Facebook or Google. “If we could talk with Rogers, we’d do it tomorrow,” he says. “We’ll go wherever a large network that has been disintermediated by Google and Facebook really wants to do business.”
