Pity the poor digital marketing analyst. Just when we think we’ve solved one problem – user-level attribution – check! cross-channel optimization – nailed! – another one appears like the nemesis gopher in Caddyshack, chirping “NYYYUUUCK!

The latest is multi-device attribution. Before you doze off, let me assure you it’s a significant issue that threatens to undermine progress that’s been made in recent years to improve our understanding of digital marketing’s impact on success. And there’s no obvious solution.

Imagine the following mini-drama:

  • Scene 1: You wake up, check Facebook, see a “sponsored story” for a car brand
  • Scene 2: On the bus, you see a rich media ad for the same car brand on the USA Today site, on your tablet
  • Scene 3: At work, you check Edmunds.com on your desktop and see banner ads for the car; later, watching Family Guy on your lunch hour, you see a pre-roll ad for the same car
  • Scene 4: That night you’re watching TV and see a spot for the car, cry “uncle!” fire up your home laptop, go to the car brand’s website and schedule a test drive

Today, it’s likely none of that flood of advertising gets credit for your test drive. Why? Because campaign attribution currently sees you as five different cookies (or events), rather than what you are: one hyper-connected dude with five different devices.

So while attribution vendors are getting very good at tracking exposure to banner ads and search – and doling out credit – they are actually following cookies, not people. In a world where more than 60% of us own a smartphone, almost half own a tablet, and 82% of global consumers in a recent Microsoft-led survey said they like “multi-screening,” the damage in terms of inefficient marketing is major.

Any solution would require a common ID and probably a platform that users engage with across devices. Obvious candidates include social networks and media platforms, so it was no surprise when Google announced a potential solution at its I/O developers’ conference last May.

Google wants to extend its Universal Analytics platform to track users across devices, and even off-line. However, it only works for users who are signed in to the website or app and can be authenticated. Facebook itself is a strong contender, with high cross-device penetration and a persistent login. But so far, it’s not in the attribution business. eBay owns an attribution platform, Clearsaleing, but it has lower penetration. Savvy retailers like Amazon have potential. Another approach could be through data hubs, which pool cookie and third-party data from multiple sources. Vendors and agencies such as Collective and OMD are pursuing this angle. For now, our advice is to be aware of the issue, especially as it impacts your attribution models.

Still, you have to admire consumers. They’re always a few steps ahead of the marketers, who are left to sputter, like Bill Murray’s Carl Spakler character at the end of Caddyshack: “In the immortal words of Jean Paul Sartre, ‘Au revoir, gopher!’”

2 Comments
  1. August 13, 2013 at 7:15 am
    Yehoshua Coren says:

    If I understand this article correctly, the main point is that people use multiple devices and marketers should be aware of that.

    Is that it?

    As an aside, being logged into a Google Account (what you referred to as Google+) does not have anything to do with the Universal Analytics platform. UA does visitor stitching based upon “opt-in” authentication. In other words, you need to log into the website / app in order for UA to authenticate you’re the same user. A website could do that with a Facebook login to their site as well.

    • August 13, 2013 at 2:22 pm
      Martin Kihn says:

      Thanks for the clarification Yehoshua. As for the article’s point, I was trying to raise awareness of a problem for readers less expert than yourself. (As a ninja, you are among the elite.) You’d be surprised how many companies believe they’ve “solved” the attribution problem already. I don’t have a solution to the multi-device problem — but it doesn’t seem that anybody does yet. I’d be happy to be proved wrong.

Comments are closed.