Marketing and Analytics – Part 1: Are we there yet?
Recently, I sat down with one of our newer clients, the customer intelligence director for a global hospitality brand, to find out more about the data challenges faced by his organization. Our conversation covered a range of topics. It’s a candid picture of the current marketing/analytics landscape, and we present it here in three parts.
Q: Thanks for taking the time to talk! Let’s get right into it. What are the major challenges you face, with regard to marketing data analytics?
A: The biggest challenge right now is data integration, which has two main aspects. There’s an issue with being able to track data accurately across time — in other words, setting things up so that newer data can be consistently compared to older data — and there’s the chief challenge, being able to integrate the data generated by various platforms: DCM, AdWords, Facebook.
Applying tracking codes and getting them tracked correctly is a huge hurdle — they’re not integrated at all. That forces us to have an exogenous process that’s outside the processes that our teams trafficking those advertisements are accustomed to working within — and they don’t like it. When you give them a tool like Claravine, it makes them happier, and helps them to do their job, but it’s still tough.
The whole idea of adding a tracking code to a clickthrough URL tends to make people’s heads explode. They just don’t get it. It’s a big challenge with tracking codes in general. It’s not just managing a classification file — that’s been solved. It’s not just the integration of process and tools — it’s the fact that this is really external to our team’s workflow, and they just don’t get it.
Q: I think part of the problem is that we’re in a moment when people have a superficial understanding that everything is “trackable,” without any real idea of what that means, in terms of breaking it down and knowing how things get tracked. Is that part of the problem?
A: Yes, I think that is a big hurdle. The disconnect between knowing that things are trackable and how that actually gets done is substantial.
Q: And when you say, “people’s heads explode” in response to the concept of tracking, who exactly is having the freakout?
A: Well, everybody. Every non-analytics team I’ve ever tried to describe CIDs to (we call our tracking codes CIDs), has had some kind of brain blockage. Before we had Claravine, getting our partner — the agency trafficking our ads — to generate their own CIDs and attach them consistently using our old CID generator (which wasn’t great), was a huge struggle. It took forever, and mistakes were constantly creeping in. Setting aside the creating, getting the CIDs attached was tough because the team didn’t understand how the query string in the URL should look. At one point, I tried to show them in the campaign parameter how to make sure the value was getting picked up, but I gave up on that. I said “just make sure it’s in the query string when the page lands.” But then we had external technical issues with re-directs that were moving that query string. It turned out that when they trafficked an ad, the agency’s QA process was so poor that query string parameters were being stripped off of landing page URLs, in a way that was causing pages to break and show a 404 error. So this wasn’t even a CID issue. I found out about it because I was tracking CIDs, but then I found a lot of pages that made me say, “wait a minute, this link is entirely broken.” And their attitude was, “oh, don’t look over here.” We got through that issue with our agency, but that gives you an idea of the workflow issues that we were dealing with.
Q: If I’m understanding you right, you were contracting with an outside agency to offload work that had become a pain point for your team, and then you were essentially having to take your time, after the fact, to re-check all their work to make sure they were doing what you asked?
A: Right. This is our media agency partner that goes out and buys media and impressions for us, and we supply them with creative that comes from our ad agency. But yeah, that’s effectively it. The agency responsible for serving the ads is the one having the trouble.
Q: I assume the analytics team doesn’t own the agency relationship? This is a common challenge for teams we work with: having to ensure the work of people for whom they don’t have oversight. What’s your leverage for getting what you need from the agency, to make sure your priority items are implemented?
A: You’re right, the agency relationship is owned by my marketing counterpart, our media director. She owns the relationship, but she’s my peer and we work really well together. My main leverage is that relationship: she’s told them that anything I say is the same as if it came from her.
We’ve also focused on informal relationship building between my analytics team and the agency partner. Our marketing department isn’t really 100% digitally literate. When I was able to demonstrate to the agency that I understand their world and can speak their language of analytics, it went a long way in building that trust.
Stay tuned for Parts 2 and 3, where we’ll discuss the quickly shifting marketing tech landscape, and the future of the Analytics organization.