Hierarchy of CDP Needs

How To Achieve Brand Self-Actualization

At Gartner Marketing Xpo, ActionIQ and Merkle provide a hierarchy of CDP needs, a concrete framework to enable any organization determine whether a CDP provider is the right provider for them.  Watch the video now or read the transcript below:

Justin Debrabant, Head of Product, ActionIQ:
First of all, thanks everybody for coming. At a 7:30am session, you never know what to expect, but this turnout is pretty impressive. You know, we kind of went back and forth on what the topic was going to be. What we settled on is a capabilities-agnostic approach to understand the CDPs space. We saw a lot of talks from different CDP vendors focusing on capabilities and use cases. Given that we had the first slot in the morning, we decided to take of a capabilities-agnostic approach instead and give our take on the category overall and the value of the category as a whole.
Justin:
By way of introductions, I am joined by Craig Howard, chief solutions architect for Merkel, and I'm JustineBrant, the head of product at ActionIQ. We have three defining goals for the talk today. First, I am definitely going to opine a little bit. I want to put out our viewpoint on the space as a whole. Next, we're going to try to define some dimensions and boundaries that you guys can think of as you're evaluating potential CDP vendors. Obviously I'm biased. I work at ActionIQ, but I really do believe that this space can only benefit as a whole from more clarity, better definition. It doesn't do ActionIQ any good as an individual vendor to have so much confusion in the space.
Justin:
And finally, one of the big takeaways that I want you to get from today is that there's no one-size-fits-all, perfect vendor for everybody out there. And that's true of ActionIQ, too. As much as I would love for you to all be potential customers of ActionIQ, we might not be the best fit for you. And we're going to talk through some of the ways to figure that out. So hopefully you can walk away a little more informed, with some key question, and key dimensions as you go to, different talks from different vendors in the CDP space throughout the expo—and I know there are a lot. I know that Craig and I can sit up here and talk for 30 minutes, but we're happy to also take questions, especially as we get into the conversation.
Justin:
First of all, no, there aren't really 87 CDPs. I actually don't know how many there are. I haven't gone through the exercise of categorizing which are CDPs and which are not . But I'm very confident that there are not 87. 87 is a number that we got from a well-published a institute in the CDP space. But I'm up here saying that they are not 87 CDPs—especially as this number keeps growing, which week over week it has been. The other point I want to anchor on throughout the course of this talk is that I really feel like a CDP should ultimately turn first-party data from a cost center to a revenue generator.
Justin:
We've all been around for the hype of big data. It's been 10, 15 years.There's been a ton of investment at different enterprises, both big and small, Hadoop and data lakes, plus a lot of buzz, a lot of effort. In my previous career, I was actually consulting on the CIO side of the world, and I saw a lot of these projects. And I think what was always missing from those projects was the question: Now what? We've invested in the state of infrastructure, we have this big shiny data lake. It's so awesome. There's petabytes of data, so what do we do with it? And really, I think that the kind of defining characteristic of any CDP, independent of individual capabilities, is that it should turn that first-party data from a cost center into a revenue generator. And so that's really main theme of what I want to dig into today.
Justin:
I'm sure all of you have seen Maslow's hierarchy of needs in some form. The idea is that you have to meet more basic needs at the bottom of the period pyramid before you can actually get up to self actualization. If the ultmate goal is to drive personalized experiences, create one-to-one marketing experiences, or whatever kind of marketing terminology you want to use, the base of the pyramid is to unify your data. But it's a means to an end, right? You definitely have to have your data unified across silos and resolve identities. But it's a hierarchy as you go up. So you kind of start with these, and it focus projects. We don't need to go into the details, but a lot of data collection involves a lot of ETL and MDM and identity resolution.
Justin:
And then you start to move up into the analytics world. You're doing customer analytics. You're doing behavioral personas and journey analytics. Then finally you start to have the data in one place. You know a little bit about your customers. And finally you start to get to the more marketing-driven use cases. It starts with audience management, then omni-channel campaigns, channel testing, journey orchestration. Again, these are just some examples of the types of use cases that exist in each of these categories. And the end result is personalized experiences. So the question is, how do vendors map to these different needs? I'm not up here to call out specific vendors or go deep into these capabilities, but I do think that there are buckets that are worth mentioning.
Craig Howard, Chief Solution Architect, Merkle:
The first is, if you think of the x axis here as the strength of offering of a group of vendors— vendors that have more of a marketing focus—I would say they are kind of the channel vendors. There are your traditional ESPs and mobile push vendors. They're really focused on a specific channel. Obviously they're leveraged primarily by marketing. They actually run the marketing campaigns, and they probably have a little bit of analytics built in, but they're very, very light on it. They really don't have the capabilities or infrastructure needed. The other main category is data infrastructure, right? This is IT-owned data infrastructure. There's probably some analytics capabilities, but there's a pretty big disconnect between them and marketing. And I would argue that there's also kind of a category in between that's called operationalized insights.
Justin:
So it's the insights that have been developed by a one-off thing, or custom models. And maybe it connects a little bit to a marketing campaign, but it's probably not really at scale and across all your kind of marketing touch points. And so these are the three main categories of different capabilities, different vendors, different value props. The question is, is a CDP the best of all these? It's just kind of marketing nirvana. I would argue no, because I don't think there is really a best of these things, and I don't think that a CDP should be approached as the solution to everything. And the hype around this is driving a lot of the confusion.
Justin:
Even being a head of product at a CDP vendor, I will admit that no vendor, no product is good at everything. And it doesn't matter if you have 20,000 employees and you're a marketing cloud, or you have 100 employees and you're a startup. You built a product for a specific set of use cases. And this is a very complex space. The needs around data and the needs around analytics and the needs around marketing are all very complex, very nuanced, and there's no way to be perfect at everything for everybody. That's the reality.
Justin:
Good is relative, right? What is good is going to be largely driven by the organization and what your organization needs—the size of the organization, the maturity of investments across those different buckets. So a vendor might be good for somebody over here, but they're not a good fit for somebody over there, because the needs are different. And this is really the meat of the conversation today. Now I'll hand it over to Craig to talk about the questions to ask yourself or your organization as you go through this evaluation process.
Justin:
The sweet spot is in the middle. You want to try to figure out what the vendor's strengths are. You want to figure out what your organizational needs are. And this is tough, right? Because you know, it probably feels like you know no vendor is perfect and your organization just has so many needs, and you don't know where to start. Any salesperson is telling you that their strengths are just in perfect alignment with your organization. And sometimes this is true. But certainly not all the time. So, how do you delve into this and figure out what that sweet spot is for your company among the vendors that you're looking at —or even whether a CDP vendor is right for you at this time? So the first question that we'll dive into the first category of questions is really why are you buying? Why is the organization that you're in buying? Right?
Craig:
Again, please feel free to interrupt us in and throw out some more questions. We have a few thought starters here. At Merkle, we do a lot of deploying of technology solutions and analytic solutions, but we also spend a lot of our time just helping organizations rationalize their technology investments and put together technology roadmaps to help them execute on their business strategies. And one of the things that often comes up in this conversation, because it is a very confusing space right now, is, if you're in fact to Justin's point trying to really drive towards value and become a revenue generator, we often tell organization, if you can't figure out a use case that you can execute on in two to three months, then a CDP might not be right for you right now. It might just be the wrong timing. It might be right for you at a certain point in time.
Craig:
There's gotta be something that you can really move fast on. Because CDPs are more than anything about really accelerating new capabilities, giving your organization the ability to execute on a a new use case. Maybe leverage an audience in a new channel. There's a whole variety of different approaches that you can take there, but it's gotta be about moving fast and speed-to-value. So, part of what you have to understand is, what are your current capabilities? And at Merkel, we look thoroughly through your marketing tech stack. We're looking at your data, how you are managing data, what your insight capabilities are, how much you're orchestrating your programs, and how you're creating those experiences ultimately.
Craig:
And you've got to understand what your capabilities are today. Those that you think you can leverage that you've invested in, to figure out how can map a new capability that CDP can support to the gaps you currently have in your ecosystem. So you have to understand your ecosystem. As you think about those gaps that you're trying to fill, you want also know where the particular vendors fit in, including into your long-term journey and where you're going. And we see organizations adopt a CDP and it doesn't go much further than a glorified email trigger mechanism, or it solves a simple use case around abandoned cart triggers or something like that, and if that's the extent you're going to take it, then you're not going to realize the return on investment that you've made in that particular technology. You've got to have that longer-term view, as well as a shorter-term view in mind when you're thinking about what you're trying to buy.
Justin:
And why you're buying given the existing tech stack and an existing ecosystem within your organization. Right? And that's part of the challenge of CDPs. It's much easier from a budget standpoint, from a marketecture diagram standpoint, to say, okay, this box I have here, I'm going to take it out and replace it with a CDP. I mean, that would be great. That's very easy. The reality is, that's probably not true. A lot of what a CDP does is new, because this is a new technology, right? It doesn't fit neatly into an existing box. And really the way to approach this is, how does it raise the water for all the existing technology that you have? It doesn't mean that you won't eventually replace some of your existing tech stack with the capabilities of a CDP, but it's just not always neatly defined like that.
Justin:
And so this is part of the challenge and part of what you have to assess for within your organization. Is the CDP right for me, and what are the use cases that I'm driving towards, the how is it going to fit in my stock? And you know, at ActionIQ, we have a lot of conversations with enterprises big and small around just helping them understand how this integrates, how this fits, what's this replace? And the reality is, there is never a perfectly straightforward answer. And it takes a lot of organizational understanding to really get to a solution that everybody's comfortable with. So, again, it would be easier if there was a box and we could replace that with the CDP, but that's probably not the reality. And you need to ask yourself in your organization some of these questions to really help figure out, okay, what value am I trying to drive, and ultimately how is this going to work within my existing tech stack and ecosystem?
Craig:
The organization's important too. We often see where now that you've centralize your data, now you've centralized your decision-making around who your audiences are and how you're going to be communicating with them, this changes the dynamic of the way that your teams might work if you've got very channel-focused teams. Now, you still have experts in the channels and will continue to have them, but they will do a lot less orchestration within the channel, because it will be done in a much more centralized way. So your organization changes a bit and, and often with the adoption of the CDP will actually force that change within the organization and how the teams work and, and quite frankly, what people's roles and responsibilities are. However, you still leverage the same people. It's just that their day-to-day activities may change a little.
Justin:
That’s a good point. Organizational change management is an important part of this. We always say, technology is necessary but it's not sufficient, right? It's not going to get you to this self actualization, right? You need to really adopt the mindset and, and work your organization around the way that you want to move your marketing team and marketing stack forward. So, it's important to realize that you're not just going to drop this in, find the good spot for it in your tech stack and everybody goes on, business as usual. To get to personalized experiences and ultimately more revenue, you really have to adopt the mindset of driving change, both from a technology standpoint and an organizational standpoint as well. Any thoughts, questions?
Audience Member:
Can you tell us about successful companies, and what they did well as well as about those that may have failed at it on their first try? What kinds of tips could you give us as we go into this journey?
Craig:
I think a good principal to have is, going back to Justin's point, sometimes good enough is good. Do not try to chase perfection. Get something up and running very quickly to demonstrate value, either impacting your program or just demonstrating to the organization how quickly you can execute on something new, which is often a big win. I think that's an important thing to keep in mind. On the other hand, I think you can't lose sight of the fact that no piece of software is going to solve all of your problems. And specifically what I mean is, data. It's still dirty, and managing data quality is still required. So you still have to keep that in mind and understand where heavy lifting has to occur. And make sure that if that's part of what you're trying to execute on, that you solve for that.
Justin:
And I would say for our clients that have been successful really from, from day one, the thing that made them very successful was the organizational alignment around what they were doing and what everybody's role in that was. This is not a technology that somebody in marketing can go off and buy and then go to IT and analytics teams and say, hey guys, we bought this CDP thing. I think we need data for it. If that happens after the fact, it's going to stall the whole process, both the rollout and adoption.
Justin:
So I think you need to be upfront that, as this will touch a lot of different parts of the organization. Get everybody on the same page about the value that you're driving to as a company. And importantly, establish everybody's role in that process. For clients of ours that have done that really well, I think that's been the differentiator. They've had really strong alignment within tech, within analytics, within marketing. Everybody knew everybody was on the same page. You know, it's always a good sign when we're doing kickoffs and we have representatives from every part of the organization there. And everybody's, excited and aligned and think the solution is going to make their day-to-day a lot better. And again, we have had clients where they took a more siloed approach to it. And the adoption always has to happen organization-wide. I mean it's just too foundational a technology. Driving that alignment from early on I think is important.
Audience Member:
Are there any specific use cases that a regular marketing automation platform cannot do?
Craig:
Yeah, I think oftentimes it's the ability to mix behaviors that were previously very difficult to integrate or too costly to integrate. Back to acceleration, the CDP should allow you to do that. So for example, understanding intent based on site behavior and then mixing that with actual offline data that you have about, say, purchase behavior—depending on the organization, of course. That's just two different channels, but when you start mixing more than that in there, then you've got a really good view of what this particular customer is doing, what they're interested in, so that you can tailor those communications better. Those are the types of use cases where you start to see something unique that's being enabled by a CDP as opposed to the older way you've been doing things with, for example, a campaign management tool or, or in more manual, time-intensive, ad hoc way.
Justin:
Yeah, mixing behavioral signals. I think that's definitely a key, right? Active customers are not just people who opened an email in the last 90 days. There's a lot of other ways they interact with your brand. And being able to pull all that together for a really high-def view of every individual customer is key, and that enables a lot of different use cases.
Audience Member:
Along similar lines, I'm from our customer experience team, not really marketing. I often have heard a CDP is a marketing tool. But I feel like a CDP is more of an enterprise tool. So what is your observation? And how do you see a CDP working after a year or two? Is it going to be one tool which going to do everything, or it's going to be one of four or five or seven tools?
Justin:
I can take a first pass. The first question about whether it is a marketing-only tool—the focus on it being marketing owned is really that you're democratizing the data that was typically locked away in IT-owned infrastructure.. So, a CDP makes marketing more self service. Now to your other point, I actually don't think that first-party data is just a thing that's useful for marketing. It's useful for customer success. It's useful for clienteling use cases. It's useful for a wide array of different kinds of organizations within the larger enterprise. I think that makes a CDP an even more of a strategic investment. And I know the focus is on marketing right now, but with our clients, we actually, we do integrate with a lot of customer success and clienteling and analytics applications. It becomes an enabler of first-party data throughout the entire organization. I think the emphasis is on marketing right now and that's okay. Where I see the CDP in DMP space moving forward really depends on the organization and how much you're really leveraging the parts of a DMP that a CDP will not provide. If you're leveraging a DMP to target first-party audiences around the big, walled-garden paid media channels and 80% to 90% of your spend is in that, it's probably not worth it to have a DMP if you're buying a lot of third party data. If you're optimizing across a lot of different smaller channels, then yeah, a DMP probably makes sense as a complimentary piece of third party data to a CDPs first-party data. But we have had clients that have scaled down their involvement in activities with their DMP because through ActionIQ they started orchestrating across the big paid media channels. You know, Google, Facebook, Snapchat. More and more of the spend is going to those walled gardens, and our customers have direct integrations with them that they leverage. So it really depends on what the media strategy is.
Justin:
So, now let's turn to the questions and guidelines that can help you approach the evaluation process. So now hopefully you have a good sense for where your organization is, but that doesn't mean you know where the different vendors are. So how do you figure out who's a good fit for the needs that you have?
Justin:
And you never want to buy technology for technology's sake, right? So as you're thinking about what the right fit for you is, you always want to start with your business strategy, which is going to really inform the use cases you're trying to execute on, and which therefore dictates to you the requirements and capabilities that you're looking for in CDP vendors as you're going through this buying process and trying to make the right choice. I think another important thing to think about is, the programs that you're executing, those use cases and how you are you going to ask questions of the data. So you bring together all these different behavioral sources of information that you're able to stitch together and understand in a single profile of a customer.
Craig:
And as you think about what you're trying to understand, what you want to actually do, how you want to define those audiences. Make sure that the data can be stored in such a way that supports those questions. And make sure that the platform itself allows you to ask those questions in that way. So it's a very important thing to think about. Maturity is of course important. How mature are the capabilities that are important to you? I also think there's a lot of noise sometimes around connectors and a platform having a certain number of connectors. That's great, but at the end of the day, you probably only need a handful of connectors to support your organization. So make sure that those connectors are really going to accelerate things, in that they're being used in the way that you use those tools that you're trying to connect to. So specifically, do you think about systems integration? I tell a lot of organizations that a CDP implementation project isn't about implementing a CDP as much as it's about integrating your systems. You've got to understand what those handshakes look like, the nuances and the specifics to those handshakes—and make sure that that's going to work appropriately for the way that you use those channels, specific tools or other orchestration tools.
Justin:
I think the question that's always most telling we get asked as a vendor is what our strengths and weaknesses are. And whether we really have use cases and references to back those up client references. I mean there's a lot of snake oil out there. That's why there's 87 CDPs. And it's really easy to claim a lot of this stuff but not have real concrete results of having done it before. So asking those hard questions around, hey, you talked about this use case—do you have a case study or do you have a customer who is able to reference and who looks like our organization now, or the way we want to be in the future? That's important. You need to buy a CDP with a good organizational fit, and finding CDP providers whose clients look like your organization is actually a really good way to do that.
Justin:
I know we're wrapping up, so I just want anchor on this. I do think the easiest way to think about this is a two by two matrix, with the bottom focused more on unifying data and then the top is focused on driving personalized experiences. All of this has depends on whether it's a mid-market or enterprise organization. I do think that these are very different problems. They're very different solutions. Data is different and teams are different at the mid market; enterprises are just so much more complex. There's so much scale. But it's not just about scale of data. I really think it's more about the complexity of data than the scale. It's great that you have petabytes, but being able to transform that into something usable for marketing is really the hard part about dealing with enterprise data.
Justin:
So, I think that there's actually a divide in terms of the CDPs for mid-market organization that's kind of a one-stop shop for everything, versus an enterprise solution that is going to be able to handle scale and complexity. Maybe it doesn't do everything because it's part of a broader tech ecosystem. And just to anchor on where we are, we feel like we're definitely in the enterprise space, with a focus on personalized experiences. We really put an emphasis on the up-the-stack value prop. Again, we obviously are able to handle the kind of the scale and complexity of enterprise data. But that's a means to an end, and the real value is up-the-stack and orchestrating these personalized experiences. If you really ask other CDP vendors to talk about where their strengths are and where they see themselves, my guess is they're probably in different quadrants, and we have our view of who the best-of-breed in these different quadrants are.
Justin:
And again, most of the vendors in the space are startups. And it's bad for us to focus on opportunities that aren't a good fit. We don't have the resources of these big enterprises to throw around. So, you know, we need to find partners where technology is a good fit on both sides. And so it's important to make sure that we qualify you and you qualify us, to know the questions to ask. So this is why, we wanted to give this talk. And the last thing I'll say is that, the Gartner analysts are pretty busy, but this is starting to look a little bit closer to a magic quadrants. I'll send them this template and they're free to use it whenever they get around to publishing it. In full transparency we are happy to answer questions about where we think other vendors are and where our own strengths and weaknesses are in this space. And it's always a good conversation. And hopefully at the very least, maybe you're not a good fit for ActionIQ, but you leave a little more educated and informed about the broader space.
Justin:
Thank you.