Commentary

Channels Or Devices? Across The Attribution Divide

Marketers are hung up on cross-channel attribution. But they should be thinking about cross-device attribution. That’s the view at Zaius, a B2C solutions provider.

For insights into getting one view of the customer, and on how to align metrics with business goals, we spoke with Eric Keating, vice president of marketing for Zaius.

Please explain the difference between a channel and a device.
The device would be a desktop, a tablet, a smartphone. We expect devices will continue to be brought to market, like the Apple Watch. It’s a challenge because whereas there may be a cookie associated with a browser or a desktop, you’ll get a unique cookie if they visited on a mobile device. Marketers need to know that when I browse on a desktop, I’m actually the same person.

How do you overcome that challenge?
This is based on auto integrations with email service providers and e-commerce platforms, and data from mobile apps. We’re also browsing from the Web — the tag can grab anything from the product ID. Do they come from Facebook, via campaign X?

advertisement

advertisement

What kind of metrics are you looking at?
By unifying cross-channel interactions and cross-device identifiers down to individual customers, marketers can understand Web site interactions and attributions far beyond opens and clicks. When it comes to stitching profiles across channels, there is a basic identifier — a customer ID. An email address is the most obvious. The other way to stitch it together based on the probable behavior of this person — it’s the same behavior as this other identifier.

What’s driving this need?
I hear all the time that customers are demanding a connected experience across every channel they engage with. And if they don’t get it, you risk losing them. Companies I shop with learn about me. They should learn that I do open emails, and that I buy men’s outerwear. I don’t want emails about women’s clothes.

Who’s successful at this?
Primarily retailers, any business with an e-commerce play. It’s better with a business that needs to prompt more reactionary buying, and does less well with subscription-type businesses. We work with Polyvore, a recommendation engine that drives clicks to fashion brands. They’re a division of Yahoo. They’re focused on curating fashion recommendations for visitors from multiple retailers. If I’m a stylist, I can put a trunk show with this shirt from L.L. Bean and these pants from Banana Republic.

Then here’s Hard Rock, which is using offline behavior to drive digital marketing. They ask, ‘Who’s this person, how are they behaving in shops and restaurants,’ and they will send emails.  

Despite that, are most marketers there yet?
They’re catching up as new channels and devices are introduced. The ownership of customer data is pretty interesting. You start to have these walled gardens like Facebook. They do want you to be successful spending money with them. For example, Facebook and Google both have ways you can use your own customer data and how to understand to better target customers. It allows you to upload a list of your customers, and target them directly with ads. You can segment based on behavior, and you can look at lookalike features.

I wish Facebook wasn’t serving me ads for magazines I’ve already subscribed to.
That may be the magazine’s fault. It’s a horrible thing for the magazine — you’re getting annoyed, you’ve already subscribed. They should know that you were the same person who subscribes.

So what do you see in the future?
It’s about uniting your customer data. Companies have these customers in their e-commerce system, and in an email service provider, and people on a Web site. All the behavior is disconnected. It’s hard to really understand a single customer. Uniting data is the key to unlocking any kind of cross-channel attribution.  

Next story loading loading..