8 min.
Three good reasons not to invest in a CDP (right now)
1L’art de la gestion de projet2Un projet à succès commence par une bonne gouvernance3Cascade, agilité, demandes de changement?

Three good reasons not to invest in a CDP (right now)


With the disappearance of third-party cookies and mobile identifiers, marketers and their respective agencies are starting to realize just how important primary data is. Faced with the need to organize their clients’ data, many marketing specialists (particularly in digital) are turning to the most used solution in the industry in recent years and investing in the current favourite MarTech platform.

And by all accounts, everyone’s favourite right now is the CDP (customer data platform), which clearly appears to be the natural successor to the DMP (data management platform). After all, CDPs were created for the sole purpose of managing your client’s primary data. What’s more, they come with a bunch of connectors (upstream and downstream) that enable you to process and send your clients’ audience data to all your other marketing platforms (and often there are very many of them, as I’m sure you know!).

But is a CDP what you really need to prepare yourself for this new era in marketing? Having road-tested many implementations of CDPs, DMPs and other cloud-based analytics or MarTech platforms of all kinds, I believe in many cases there are at least three good reasons to answer “no” to this question.

By the way, I should warn you: Many readers will likely find this article to be a bit harsh and even brutal in some respects. But don’t be too fearful—I wrote this text from a place of love (tough love, for sure, but love nonetheless) and you should realize that these words come from battle scars earned over the last ten years of combat. 😉

Reason 1: You’re underutilizing your current platforms

According to certain studies, marketing specialists have on average up to 16 MarTech platforms in their technology stack (and up to 20 for B2B marketing specialists). However, in many cases, the capacities of these platforms are not fully exploited.

Take for example Google Analytics. Having worked on many mandates related to data platforms, I’ve been surprised to see to what extent marketers ignore all of the possibilities offered by GA. Basically, a good number of CDP functionalities aren’t necessary if you’re using Google Analytics properly. This is even more true if it’s combined with Google Ads and Google Optimize (not to mention Google Cloud), particularly with regard to audience segmentation, activation and personalization. All that’s needed is a bit of imagination to develop a few good use cases for testing.

And yet, for reasons I don’t understand, very few marketing specialists (even those who consider themselves to be more “mature” and advanced in the field) make good use of these tools.

The result: Instead of realizing that it’s possible to more fully exploit the potential of their existing platforms, these specialists instead think they need to add a new platform that’s even more powerful to reach their objectives and improve their results. As I mentioned previously in an article called How many millions ($) are hiding under your marketing data?, I call this phenomenon “platform hoarding syndrome.”

Unfortunately, those suffering from this malady wind up with a new gadget that is too big and complex to manage, in addition to not suiting their immediate needs. As a result, after driving hopes to the maximum possible (to the very summit of the Gartner hype cycle), unrealistic expectations quickly crash, tumbling into the trough of disillusionment. Eventually, the new tech will wind up in the trash, alongside innumerable other costly platforms wasted by companies over the last decade.

Reason 2: You don’t have time to wait for a CDP to become operational

What we’re witnessing now is the end of third-party digital identifiers. Yes, officially the end is planned for 2024, but in fact it’s already happening now and the official end date is merely symbolic. You therefore need to ensure that your primary data solutions are operational as soon as possible. That’s why you must rigorously analyze your current situation and draw up the most efficient road map to fulfill your needs, for both the medium and short term.

In my experience, the thing about CDPs is that it takes 3 to 4 years on average for a company to get up to speed with this tool. You generally need about one or two years to make the platform relatively operational, then another one or two years to acquire the level of mastery needed to use it to its full potential. In some cases, this requires even more time, and in other cases the maximum potential of the tool is never fully reached. You of course need to also consider that the bigger the company, the more stakeholders or departments are involved, and the longer the optimization process will be.

You must be asking yourself why it takes so much time to make a CDP operational. Ever heard of the three Ps of IT? If not, here they are:

  1. People
  2. Process
  3. Platform

These haven’t been presented in a random order—rather, they must be approached in this order of priority.

The problem is that too many marketers are only familiar with the last P. As a result, they launch CDP implementation projects without first evaluating their needs in human resources as well as the organizational and cultural changes involved in such an implementation. Too often, it’s this basic error that explains the failure of these projects.

In the first year, things don’t seem so bad, since everyone is laser-focused on the technical implementation and all the challenges that brings. However, it’s often after one or two years that stakeholders wake up and become increasingly impatient to reap the benefits of all their work. At this stage, they want convincing results and an interesting return on their investment in proportion to the significant amounts of time and money put into the project. The hitch is that, in many cases, it’s only after the implementation is complete that they realize no one (or almost no one) at the company really knows how to actually operate a CDP.

It’s possible that at that moment you might remember that the consultants or firm hired for the implementation had warned you that you would need support and a specialist to get your CDP off and running (at least if they did their job right). They may have even provided you with a complete manual (as we do at Adviso) for supporting and guiding your audience manager in their daily tasks, including a detailed description of the role and tasks of specialists to be hired.

Unfortunately, the advice of these consultants or specialized firms is all-too-often ignored or their importance minimized by companies that are probably hoping this new piece of “all-powerful tech” will take care of everything by itself, handling the task as if by magic.

Sadly, there’s nothing magical about a CDP or any other more advanced MarTech platform. They inevitably require investments in time and resources as well as, above all, a firm commitment by the company to develop an organizational work culture oriented towards data intelligence.

Reason 3: You lack analytics experience

I know you probably find that subtitle to be a bit harsh, but it’s usually true for many companies in Quebec and Canada. According to a study conducted by Deloitte, only one organization in 20 consider themselves focused on data insights, a much lower proportion of companies than you would find in the U.S. (17%). This tallies with my personal experience over the past ten years, during which time I’ve observed this shortcoming while working in the field.

That being said, a company’s analytics experience is intimately linked with its maturity in management and governance of primary data, factors that determine the chances of success with a CDP implementation.

If your digital marketing team is not already doing audience segmentation on your current platforms, the introduction of a CDP won’t change this situation. The advantage of starting to create and, especially, analyze your audience segments now, no matter how basic they are, is that you’ll gain more analytics maturity in the process. Test the limits of your current platforms in segmentation and primary data. This will then enable you, through a process of trial and error, to discover concrete use cases that may require the implementation of a CDP, without having to spend a fortune in the process.

As a result, it might be preferable for your company to invest in a resource that will help you better understand and exploit your primary data and current platforms, instead of investing significant sums in a new platform that no one on your team understands how to use.

Invest first in acquiring data and analytics experience

According to another IDC study sponsored by Collibra (a company that was talking about data governance well before the subject became popular), the companies that are among the most experienced in data intelligence generate three times more benefits than those without that experience. This study also highlights the main goals of these more experienced firms. Here are the particular areas of application that they operationalize within their company culture:

  • improvement of data security (50%);
  • rigorous processes to ensure data quality (44%);
  • efficiency and standardization measures in the management and analysis of data (45%);
  • frames of reference for the way in which these data are used for decision making (42%).

Taking all these factors into account, before just having a good strategy (which is essential), the success of a CDP implementation depends above all on the culture a company has developed regarding data and analytics. Think about the kinds of roles (or people) and the types of strategic governance models (or process) that you need before investing tens or even hundreds of thousands of dollars in a new technology (or platform).

Does all this necessarily mean that a CDP is a bad idea?

Not at all!

It’s just that because a CDP is an extremely powerful tool, you need to think about it very carefully first. All too often I’ve seen such platforms vilified and thrown in the garbage less than two years after their implementation. In general, this was because the companies underestimated the real cost of a CDP, despite all the warnings from their technology partners or external consultants. In IT, it’s called the total cost of ownership (TCO), which is the cost of purchasing an asset plus the cost of using it for the duration of its service life, a data point you absolutely must consider. In other words, there should be no CDP without TCO.

In closing, yes, it’s possible that your company is ready for a CDP. However, you must ensure you rigorously evaluate and verify your level of preparedness and ability to overcome the challenges of having a CDP before getting involved in a long and costly implementation process. This is a simple tip that could help you save a lot of time, energy and money.

If you need any help thinking through the issue, don’t hesitate to get in touch with us.

Best of luck with all of your future digital challenges!