Customer Data Platform Evolution

The Customer Data Platform (CDP) space is on the verge of its next evolutionary upheaval. New capabilities are emerging. Users are looking towards new ways of working. And, as always, new buzzwords are appearing to confuse the unwary.

In this first post in a short series, I’m going to set out the key trends in the space. Then later posts will look at them in more depth, explain the language surrounding them, and explore what they mean for CMOs and CIOs.

So the first trend on the list is the rise of composable architectures. Until recently, CDPs have been big systems integration projects. They involved huge investment and you didn’t start to realise value from them for 12 to 18 months. The rise of composable means two things change. The first is that CDPs become modularised. Rather than thinking about all the things a CDP can do, you can focus instead on what your business needs it to do, your specific use-cases. As a result, the overall investment in technology and services is smaller, you realise value sooner, and you reduce your risk.

This also means CDP services can be deployed on your own cloud infrastructure, allowing you to benefit from efficiencies in scale of your cloud commitment for usage and data storage. You can also procure services via your cloud partner’s marketplace, streamlining procurement cycles. Crucially, DPOs love this approach, as your customer data never leaves your own cloud environment.

The next trend is CDPs becoming tools for business users, rather than having to be operated by the IT and Data teams. This also democratises access to the brand’s most valuable asset, its customer data, putting it in the hands of the people who can use it to deliver real business outcomes. The first CDPs had very technical and complex interfaces, but today those interfaces have to be simple and intuitive, making it very clear how to realise true business value.

Next ID stitching and the single customer view. This is a foundational element. Customers engage across multiple channels over a long period of time, creating huge pools of data. You need to be able to stitch all that data together accurately to create individual profiles for each of them. You also need the ability to expand this view beyond your own data, to give you better reach across your customer journey. Stitching is a complex and time-consuming process, but now AI-based predictive modelling can automate a lot of it. It can work out which identifiers get stitched to which customer profile. Even an initial implementation of the technology can get you 80% of the way there, saving huge amounts of time and effort.

ID stitching forms the core foundation, now next generation CDPs must support  true real-time events and sub one-second activations across the customer journey. Historically, CDPs have always operated on a batch basis. Something would happen and, after a few days or a week, you’d be able to respond with an action. Now, many high value use cases require data to be processed in less than a second, so you can optimise and personalise that specific customer journey.

This unlocks some really interesting use-cases, particularly around ecommerce. For example, a customer puts some items in their basket, but they don’t go through to complete the purchase. Now you can take that signal in real time and trigger a targeted email that brings them back immediately, ‘striking while the iron is hot’.

ID stitching is also vital to an omnichannel approach to journey management. Marketers have access to a very complex ecosystem of channels through which they can communicate with customers. But not only are the channels themselves quite siloed; the teams responsible for them are siloed as well. You’ll have your direct-to-consumer team, your email marketing team, your CRM team. You’ll also have your digital media team, your social marketing team, and whatever else. And in the worst case scenario, none of them will have ever even spoken to each other. A centralised omnichannel journey management strategy allows you to bring these teams together. You can facilitate and incentivise cooperation between them while creating sophisticated and engaging customer journeys

The crucial tool in all this is, of course, Artificial Intelligence. AI is the newest buzzword, but you need to unpack it and understand the use-cases it’s powering within your business. That could be improving operational efficiency by automating processes such as ID stitching. Or it could be enabling real-time decisions to improve the customer journey and the resulting business outcomes.

Further evolution will see the widespread adoption of agentic capabilities. At Zeotap, our agent will be able to answer natural language questions about building your audience strategy, for example. Once again, it’s all about moving CDP capabilities out of IT and into the hands of the users.

Then the final trend is more sector-specific; the CDP as an enabler of retail media. Retail media used to simply mean advertising that appears within a retailer’s own properties, such as Amazon’s Sponsored Products. Now it also involves using retailer-owned data to target their customers outside of those properties. Once a brand has organised its data in a CDP, it can enable that data by connecting it into the media ecosystem or by leveraging cloud native clean rooms to collaborate with supplier partners. And it can use AI to drive personalised retail media journeys in real-time.

So that’s the highlights reel. Next time we’ll start looking at each of these trends in more detail, and at how marketers can ride them for maximum advantage. And we’ll begin by pulling back the curtain and exploring composable architectures.