Marketing and Analytics – Part 2: The Value of Data Quality

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. In Part 1, we discussed the chief data challenges faced by his organization, and the disconnect between knowing that things are “trackable,” and knowing what that actually means.

Q: In defense of marketers who are struggling to keep up: the marketing/tech landscape is moving very quickly. A lot of marketers are feeling like they’ve been caught with their pants down, with all they’re expected to suddenly know.
A: Things have moved really fast, especially in the tracking space. The funny thing is, we take all this tracking for granted now. But if you actually look around at the number of companies that are effectively using tracking, and how long they’ve been doing it, there aren’t many companies that are truly using tracking codes the way they’re intended to work. They’re not truly using tracking code data, they’re using these long strings instead.

Q: Part of the issue with the perception that “everything is trackable” is that there are all these vendors in the market place claiming to be able to do very fine granular attribution, and full optimization for marketing spend. But maybe there’s a bit of the cart-before-the-horse going on… All the best attribution and optimization tools in the market aren’t going to save you if the first instance of data collection is corrupt, right? Do you agree that companies should be focused more on the quality of the data going in?
A: Oh boy. I agree with you. Unfortunately, I think we stand in a small circle of people who actually work with data, the business of delivering data and understanding data insights. Executives often like to work with the new, bright shiny things. They want to see a pretty report that says, “hey, plug in our tool and we’ll tell you where to spend your media dollars.” Execs will just eat that up, as I’m sure you’ve seen. So you have all these fancy tools, starving for good data. And your execs are asking, “why is my data always garbage?”
Right when I started in my current role, I made a bunch of recommendations. We have a decent-sized organization, so I recommended we devote half a head to the purpose of ensuring data quality.  And they said, “well, isn’t that part of what the implementation work should be?” And I said, “yeah, but it needs to be monitored and enhanced and documented…” They said, “No, we don’t want to pay for that.” I agree with you that data quality and governance is foundational. There’s a huge disconnect in the industry, where it’s woefully undervalued.

Q: We actually see this as a bit of an advocacy mission at Claravine. It surprises me how difficult it is to explain the value of data quality and how much confusion exists about what that means. We’re in the business of trying to help explain that problem.
A: I can build on that — I’ve built my career thinking about his problem. One thing that’s helped me in the last five years has been studying the topic of leadership, and looking at leaders across different organizations, both through experience and through reading. What I’ve found is that executives are often good at spending money, but not often good at leading teams or building processes. There was an Aha! moment for me, when I connected that to the symptom, which is when an executive says “Well, I bought Claravine and Observepoint; don’t I have data quality now? I bought a solution for that project, and everyone said that project was done, so I should have quality data.” What they’ve done is try to buy a series of silver-bullet solutions, rather than leading and prioritizing and building an actual business process. I hate coming back to culture, but it’s a culture problem. It’s a big missing link.

Q: So do you think the challenge here lies with People, Process, or Technology?
A: I think it’s primarily a People issue. And when I say people I don’t mean individuals or talent. I don’t want to disparage any one person. It’s culture. How are priorities determined, what do we think is important as a team and an organization?  Process is a paper. It’s a powerpoint that lays out how we do things. But when people don’t follow the process, it’s because we never instilled the culture of discipline.

Q: And People, that could also include having an understanding of what the business’s needs are, a plan for staffing against that, and securing buy-in from all the stakeholders.
A: That’s right.

Q: I’d love to know — do you have a sense of how much of your ad spend goes untracked because of data quality challenges?
A: I’ll limit my answer to Digital ad spend. As we’ve been setting up these new disciplines, we’ve had periods of darkness, due to one technical or process issue or another. When you look at those kinds of one-off implementation process “growing pains,” I’d say about 50% of our spend is untracked. But I wouldn’t categorize it as chronic. When I look just at chronic problems — issues that always come up, like people who fail to append CIDs, or a server that is always failing — it’s not actually that much, probably 10 – 20%.
Our main problem isn’t that things are going untracked, it’s more that things are suboptimally tracked. It’s no longer that people aren’t tagging their ads, it’s that they’re using the same tag for all their files. Or maybe they never fill out a particular field of information. The field might be important, but people don’t fill it out because they don’t know what it’s asking for. Or, people fill it out but in three different ways…one person uses a capital, one lower-case, and two or three have a misspelling. Claravine is actually helping us to overcome those problems, and get to that granularity.

If you relate to this, check out Part 1 of our interview, and stay tuned for Part 3, where we’ll discuss the future of the Analytics organization.