There is a wrong way to implement a customer data platform… please don’t do it.
- February 2, 2021
- By Kelsey Abegg
Nobody implements a customer data platform (CDP) on a whim. It’s a calculated, strategic decision to replace marketing-as-usual with marketing-by-individual. So, you would think that companies put a lot of time and thought into how they can make their CDP implementation a success from the get-go. And you would be right… about twenty percent of the time. The other eighty percent, well, not so much.
It’s not that companies don’t recognize the value of preparation. They’re just not sure what they should be preparing for, in large part because CDPs are still a relatively new technology. Even the term “CDP” has become a kind of catch-all description for anything that doesn’t fit into the traditional data management platform category. In fact, some “CDPs” are nothing more than glorified ETL (extraction, transformation, loading) products.
But if you’re talking about the actual customer-behavioral-insights-with-the-whole-data-science-thing CDP, keep reading. You’ll thank me for it later.
“I want this CDP implementation to fail so I can be fired!”
Admit it: You probably read that line even before you finished reading the first two paragraphs. Because no one ever says that. And yet, not preparing for the success of your CDP in the early stages of implementation is a recipe for failure. What are the ingredients that make up a successful CDP implementation project? The right people, the right data and the right proof points that show your CDP is working (and by “working,” I mean boosting your bottom line).
Make sure you have the right people in the room, right from the beginning
This piece of common advice applies to any big project, but it’s especially important with a CDP implementation. Teamwork is critical to its success and there is no such thing as a CDP team until you build one.
A CDP team requires representation from each marketing channel, IT, privacy, and any group that is responsible for collecting and using customer data to drive decision making. The reality is that marketing teams have become fractured over time, splitting into their own channels with their own sets of goals and metrics for success. CDPs are not only about bringing the right data together (which I’ll talk about in a moment) but also bringing the right people together to share that data, ensuring that the customer interactions flow smoothly from channel to channel and from experience to experience.
Take small bites of your data
Your Mom probably wasn’t a data scientist (and if she was, you were sooo lucky), but the motherly advice of taking small bites totally applies to CDPs. We often see customers who want to take too big a bite from their data and bring in everything while they’re still learning how to use a CDP.
That’s a bad idea, because most of the data you have doesn’t even need to be in a CDP. In our experience, less than ten percent of all data provides valuable insights that help predict customer behavior. Instead, we recommend starting with a list of 25-30 data fields and then adding fields very selectively to the CDP. If you’re thinking about bringing in 100 different data fields right from the start, that’s a big red flag. CDPs are all about having the right data, not all the data.
Start simple before you go big
The most successful CDP implementations start with a few simple use cases. This allows marketing to demonstrate the value of a CDP early on and build up the executive support needed for bigger projects. Converting unknown users into known users is a good early use case. Suppressing engaged users from your social media ads to reduce ad spend is another way that marketers can quickly attach a monetary value to their CDP.
Once you’ve got a few early wins under your belt, you can start to expand the reach of your behavioral-based decisioning. Over time, you’ll find that CDP use cases follow a cycle: develop a use case, gather behavioral data to support the use case, enrich that data with insights and analysis, build another use case based on those behavioral insights, etc. It begins to feel very intuitive with time, but at first it can feel a little uncomfortable, kind of like driving a car.
Go forth and prospect
Change is scary. Different is scary. And CDPs represent both a change and a different way of doing things for marketers. The fear of the unknown, however, doesn’t have to come into play with a CDP implementation. Technical consultants (like me) have been through it all. We know exactly what works and what doesn’t work as companies roll out CDP technology in their business. Follow the advice above, and you’ll end up exactly where you hoped to be: closer to your customers on a meaningful journey that feels authentic, because it is authentically their own journey.
Are you ready to implement a CDP? Find out with our CDP Readiness Guide.