The following was transcribed from a recent interview on The Agile Brand with Greg Kihlström podcast. 

Listen to the Episode

Prefer to listen? Click play on the video below to listen to the episode. 

We’re going to talk about the future of CDPs, and how Composable Customer Data Platforms point a way to how brands can utilize their data in a flexible and sustainable manner. ActionIQ recently put out The Enterprise Guide to Composable CDPs, which helps shed some light on this topic. To help me discuss this in more depth, I’d like to welcome Florian Delval, Director, Technical Product Marketing Manager at ActionIQ.

[Greg Kihlstrom] We are going to start by talking about composable CDPs, and I'm going to include a link to an article you recently wrote. It's actually how we met; I read your article and reached out, and wanted to talk some more. So you wrote an article for the ActionIQ blog. I’ll put that in the show notes for some more background. But, first, for those less familiar with the term, how would you describe what “composable” means?

Yeah, it's a great question to get things started. First, for the audience that might be familiar with CDPs, but just in case, CDP stands for Customer Data Platform. And there are different definitions, but I guess I could use the definition from CDP Institute, which defines CDPs as package software that creates a persistent unified customer database which is accessible to our systems. And now that you know about CDPs, we can talk about the notion of composability. And here Gartner is defining that idea of composable technologies as digital assets package that can deliver independent care on complete business value. It's quite a mouthful. So maybe I'll try to translate that a little bit. 

I like to use a food analogy. I don't know if, Greg, you're a foodie?

A little bit, a little bit – not as much as others, but some.

OK, sounds good. What do you think about fixed menus, like tasting menus?

You know, so I'm a vegetarian, so it's sometimes a bit challenging. 

Yeah, for sure, so, you see, instead of paying for a fixed menu where you get exactly what the restaurant wants to serve you, the idea of composable means that you can pick and choose your meal yourself from that menu. So you're not forced into a tasting menu. And so Gartner estimates that, within the next three years, 60 percent of organizations will invest in composable technology. So it's not a small shift; it's actually a pretty big shift and change. And the other thing I would add is pre-packaged CDPs, they were created at a time where the majority of data warehouses were still on premise and were meant to support analytics workloads, not customer experience workloads. But that part is not true anymore. So with composable CDPs, you can complement your existing investments without being forced in duplication within your stack.

I've seen similar things about the growth of composable and just the investments in it. And this certainly presents an interesting alternative to those, whether it's an all-in-one cloud platform or just some of these very large systems that try to do a little bit of everything, sometimes to varying successes, and I think there's over 160 CDPs, or at least platforms calling themselves CDPs, out there on the market. What are some of the challenges that non-technical marketers face that a composable CDP might help them solve?

On your first comment on the alternatives, yeah, it's definitely an alternative to, like, all-in-one marketing clouds such as Adobe, Salesforce, on the traditional pre-packaged CDPs on the market. Vendors usually are requiring you to buy everything instead of what you need. So imagine, for example, you have to pay for a three-course meal when you already had an appetizer earlier, or maybe you have a different place in mind for dessert. Another challenge is being forced into copying data from your source of truth into these business applications. We can chat about this later, but I really see a future where there would be less cookies and a lot more control access on that data. And going back to your question on non-technical marketers, I find it fascinating because I see composability as an opportunity to give choice for IT. But I don't think you should take a step back from helping business. I've seen some vendors who, you know, are suggesting to unbundle the CDP, but what it means is providing a secure interface back, on not allowing business users to leverage that – like, we know that business users are not SQL experts. So I guess my point here is you want to give optionality, but you don't want to do that at the cost of the experience on operations for business users.

How about, in addition to the non-technical, what about those technical, so those engineering and IT professionals? How does composable CDP help them and the work that they're often asked to do to support a marketing team or a CX team?

Taking a step back, when we look at the situation, IT spends millions to centralize data, in modern cloud data, so Snowflake, Databricks, BigQuery, just to name a few. And the last thing that IT wants is to be forced to duplicate that data into every business application. That's pretty much the way it works today. You don't want that investment, that centralization to be thrown away by immediately bringing back the silos, by duplicating data into some singularly original tools like an email provider or an SMS provider, et cetera. A way, again, trying to translate this is imagine you’re creating a perfect meal with a balance of flavors. But somehow, when you have visitors at your restaurants, they have a small plate, so they can only put some of that meal you prepared, some of those ingredients, on their plate. They certainly won't benefit from what you created initially from that mix of flavors. And what it means is they won't be satisfied with their meal. 

So, you know, if I could bring up that food analogy for another minute, let's imagine you're at a restaurant and IT’s providing the ingredients. So we’ll say it's the equivalent of your customer data. And you have appliances, which represent your business applications here. For marketing to prepare meals and serve them to customers, if you need a new gas range, you shouldn't be forced to also buy a fridge, buy a microwave. You should have the choice to just take what you need. And in the same way, IT needs to have control over those ingredients. If it's duplicated everywhere, they might forget about fresh produce which would be in a corner of a space; they would perish or eventually be used to prepare a meal by the marketing team by accident. And I have one more for you. You want to be able to add items on your menu to be able to expand revenue opportunities. So, really, at the end of the day, if I try to summarize, I see composable CDPs as four benefits. It's choice, control, agility and innovation.

I think the metaphor definitely helps paint that picture, to use a different metaphor. So for those that haven't quite moved in this direction yet, I mean, as you mentioned earlier, there's a lot of companies that are moving in this direction. But for those that are learning more, that are trying to make a move in this direction, what are some of the things that are required in order to make a move to a more composable customer data stack?

Yeah, that's another great question, and I'm going to move away from the food analogy for a minute, but I'm going to talk about my home, actually. Mortgage rates are crazy, but I still went ahead with my home purchase project. And so, if you want to improve your home, unless you have unlimited time or budget, which, by the way, is not my case, so we have to split the project into pieces. And so the flooring could be one example of a piece of the project. The kitchen could be another one, the bathroom, et cetera. You get the idea. And so the same goes with starting to compose your data stack. You need first to identify what are the components which go into that stack. And so an example could be to have an identity component, and maybe for some people it would be even a couple of identity components, one for anonymous profiles, one for non-profiles. So you can define best components.

And the next step I would recommend is look at which things you have in your stack. On start, look at the custom operations. And start with whatever would be the lowest effort with the highest impact. And on our side, if you want to be even more concrete, I would say what we're seeing people start with in general is this notion of audience segmentation. And the reason for that is because it's really the first step into getting the data to turn it to action. You know, everything before that, when you collect the data and so on, it's a cost center until you can put that data into action to generate revenue. 

So you touched on a few things related to this, but to, kind of, go a little bit further here and to talk a little more generally about innovation within organizations, innovation certainly takes speed; it takes many other things. How can a composable approach help organizations to innovate more quickly and effectively?

Yeah, so I would say, until recently, making a future-proof decision was all about making a decision that would be definitive. If you think about pyramids, they were built to stay, you know, like it was really meant to be steady. But now we’re in a very different world. I would say we're more in a Lego world, right, where you build from blocks on the expectation that you should be able to switch things. And change will continue to happen whether you like it or not. So what you need is to find technology which allows you to apply that innovation, to answer to change.

And I can give you an example. I'm sure you've heard about third-party cookies going away at this point. I think everybody is familiar with that. We've been talking about it for quite some time. But the use cases which are used, using that third-party cookie, they're not going away. And so a lot of others came up with alternative ideas which they called “cookie-less ID.” And so if I was asking you, Greg, which cookie-less ID would you adopt, or how would you go with choosing one?

Yeah, I mean, I don’t know enough about that to make a decision, to be honest. I mean, certainly there's privacy considerations. There's ease of use of data. There's all kinds of ways to decide, I guess.

Yeah, you know, it's really not simple. It's not a simple question. The reality is brands will choose the ID which allows them to advertise where they need. On the other side, to be sure, we want to adopt the sort that brands will choose. So it's a little bit of “chicken and the egg” thing, but you can't be in that choice paralysis. You really need to pick one and start testing. And so the challenge here is to make sure that you can do that, but without being locked in. You want to be able to change your decision in the future if you need to. And so, again, here the takeaway is adopting a composable stack, for me, is giving your organization the ability to accelerate innovation, but without having to worry about making a decision that you can't change. 

Well, let's switch gears here a little bit, and let's talk a little bit about ActionIQ, where you work. So, you know, as we talked about earlier, there's a lot of players that are not only in the marketing technology field; I mean, that's way over 10,000 players in that MarTech infographic that's so famous in our circles. But there's also a lot of platforms, again, that either call themselves CDPs or act as full CDPs. What would you say makes ActionIQ stand out from the others?

Yeah, for sure. So I think, first of all, I would maintain that ActionIQ was really built for enterprise customers. That should be a first step to narrow the options on the market. And then I would say that, the same way we were talking during this show about providing technology for change, ActionIQ evolved over time to answer organizational needs. So rather than looking for the status quo, we're embracing change ourselves. So without going into too many details here, we talked about composable CDPs, and I think it's a perfect example. ActionIQ was built and developed at the time, as we said, where data warehouses couldn't handle the load required by business teams, and so ActionIQ developed a purpose-built infrastructure for customer data. Forvase for consumer data. But now that we're able to see the evolution of the space, we're offering new deployment options for IT where you can choose where data lives, where data is queried. ANd doing that, we're still providing a business for any UI for marketers. And so we truly believe in a world where IT and business are successful without concession. On one side, you will have IT, which can increase the value on the existing investments, drive architectural decisions, keep control on their data. And on the other side, the business can get the tools they need to be successful with their CX objectives. They can reduce the customer acquisition costs, increase lifetime value, and more. So, in other words, I would say we believe that IT has the ingredients, great ingredients, and we provide appliances for business teams to make recipes which will convert people reading your menu in restaurants into customers and best first-time customers into loyal customers coming back over and over again.

That's great. And so, besides composable, and you mentioned a few things just now, even, but besides composable, where do you see the CDP market headed? What should we be expecting in the months ahead?

Yeah, I think there are certainly a few things we could talk about, and some are predictions to be confirmed in the future. But the first one I have in mind that I’ve loved tracking, in the past couple of years, is about the unfolding of advertising world with tech regulation changes. We chatted about third-party cookies, but as we said, best use cases are not going away. Organizations are still looking to prospect, to acquire new customers. So I think there is an opportunity here for solutions like ours, to help brands across the entire life cycle, from the very first touchpoints all the way to best, more typical CDP use cases, loyalty and retention. And I have a few more predictions I’m happy to chat about separately, not going too deep here, but all of them definitely are on this shift to first-party data strategies. I really think this is what drives a lot of the changes on the market today.

About Florian Delval

Director, Technical Product Marketing Manager, ActionIQ

For the last decade, Florian’s mission has been to empower enterprise organizations in their digital transformation process via the definition and deployment of strong Customer Experience Stacks. Florian educates enterprise organizations to increase the value of their technology investments in a complex and ever changing environment.

About the Host, Greg Kihlström

Greg Kihlstrom is a best selling author, speaker, and entrepreneur and host of The Agile Brand podcast. He has worked with some of the world’s leading organizations on customer experience, employee experience, and digital transformation initiatives, both before and after selling his award-winning digital experience agency, Carousel30, in 2017.  Currently, he is Principal and Chief Strategist at GK5A. He has worked with some of the world’s top brands, including AOL, Choice Hotels, Coca-Cola, Dell, FedEx, GEICO, Marriott, MTV, Starbucks, Toyota and VMware. He currently serves on the University of Richmond’s Customer Experience Advisory Board, was the founding Chair of the American Advertising Federation’s National Innovation Committee, and served on the Virginia Tech Pamplin College of Business Marketing Mentorship Advisory Board.  Greg is Lean Six Sigma Black Belt certified, and holds a certification in Business Agility from ICP-BAF.