A Customer Data Platform (CDP) is a martech solution that takes in, stores and analyses personal user data from all touchpoints along a customer’s buying journey, then may suggest and implement actionable marketing insights based on this.
They are able to track individual customers, and thus provide a 360° customer view and a more customised Omnichannel shopping experience.
But companies looking to benefit from a CDP may get overwhelmed by the multitude of options out there and confused about how each CDP works, so…
Well, how long is a piece of string?
Because the CDP market is young and has few overall governing bodies, there is a lack of agreement and consistency between Customer Data Platform companies and analysts about what different categories CDP architectures fall under according to their functions.
All the same, it’s important to be clear about the different types of Customer Data Platform software available on the market to choose the right CDP for your business. So here’s a breakdown of the main schools of thought, based on what CDPs can do:
The first and most basic function of a CDP is to ingest source data and store it in a secure location. Many other types of customer management software are able to do this, of course, from data warehouses to data lakes and many other kinds of data storage systems.
Just because a piece of software can store data, doesn’t mean it’s a CDP. But all CDPs need data storage capabilities. Many use data lakes, but a CDP can in theory be connected to any type of data storage platform.
CDPs have the advantage of being able to use data from many sources, including third-party data from cookies, your company’s own first-party data, plus any second- or zero-party data you may have. This makes CDPs uniquely placed to replace third-party cookies when these die off very soon.
If your company already has data storage facilities (and it almost certainly does), look for a CDP that is compatible with it.
The first type of CDP, then, the one that meets the minimum requirements to be called a “Customer Data Platform”, will be able to take data from these sources and link it to each identifiable customer.
This customer tracking is often achieved via tagging, and indeed many companies that offer this kind of CDP began life as tag management specialists, according to the authoritative Customer Data Platform Institute.
The CDP will store the results of this customer identification in a separate database that can be shared with external systems. Only by having a database that can be accessed by other systems is it possible to have persistent, consistent data across all systems for the same customer, creating a unified, holistic view of that customer.
David Raab, Founder of the CDP Institute, calls this type “Data CDPs”, while MarketsandMarkets refers to them as “Access CDPs”.
Moving up, we come to the next role that a CDP can perform, analysing data. CDPs that can analyse data will, in general, be able to store data and create a customer database, too.
Other data analysis tools exist, of course, but not all of them are CDPs if they can’t create a unified, 360° view of the customer.
Almost everyone agrees that we can broadly label Customer Data Platforms that are capable of doing this “Analytics CDPs”.
More specifically, this type of Customer Data Platform architecture can:
Segment customers into groups or for each individual
Create a customer journey map to follow the route that consumer took before and after buying
Perform revenue attribution to match each particular sale to the specific marketing campaign that lead to it
Some of these Analytics CDPs can use machine learning to apply predictive modelling and make predictions about how customers will behave next based on past behaviour
Finally, the sharing of this data to other systems, an integral part of any Customer Data Platform, can often be automated by CDPs at this stage of development.
The fourth function of a CDP takes it one step further and suggests specific, personalised marketing or sales actions to deliver to each individual customer across different channels.
These actions or treatments can include customised emails, notifications or other messages, as well as recommended products or multimedia content which that person may like.
In fact, the sky’s the limit when it comes to possibilities for marketing and sales activities adapted for each person.
These are variously called “Campaign CDPs”, “Engagement CDPs” or “Personalization CDPs”.
The sky’s the limit so it’s time for some blue sky thinking
As if all that weren’t enough, some CDPs are even able to perform the role of a delivery system. They are what could be termed “Delivery CDPs”.
Instead of relying on an external system to put the suggested customer interactions into practice, this final type of CDP can integrate this ability directly into the platform.
This means they can send emails, app notifications, messages on websites, implement advertising campaigns on websites and search engines. Again, the array of possibilities is almost unlimited as to what these types of CDPs could theoretically deliver, if connected to the right system.
However, Customer Data Platform products that do this often started life as delivery systems, and then expanded their remit to include customer database and analytics capabilities, so their main focus and strongest point is on delivery.
If done right, though, with a focus on building a persistent customer database that can be shared, this final type of CDP could be the best, most powerful and comprehensive type of Customer Data Platform. An example of such a Delivery CDP is the CDP 365 from Antsomi.
These categories are not wholly distinct from each other, and may overlap. Generally, CDPs that can do the higher-level, more complex stuff can also do the easy, low-level things. However, some CDPs may be missing certain crucial capabilities.
Some systems that claim to be Customer Data Platforms do provide customer data analytics and personalised marketing insights, but they don’t have a unified, shareable database, like many CRMs and eCommerce platforms.
Others, including Salesforce, claim to cover all the bases, including ingesting the raw data in the first place, but actually the Salesforce CDP relies on real-time data that can’t be shared with other systems, so it’s not a real CDP.
If you’re thinking of getting a CDP for your business, identify what resources and assets you already have at your disposable, and try to find a CDP that will fill these holes, complementing and integrating with your existing structure.
Some companies choose to have 2 or 3 CDPs which focus on and specialise in different areas of their business, and communicate between them.
Some have 1 CDP that can integrate with or even replace their current data storage, management systems, CRMs, DMPs, analytics models and/or delivery systems.
At the end of the day, you will have to be sure what you want your CDP to be able to do, and start your search from there.
In summary, it’s not possible to categorise all CDPs into clear-cut groups (although many industry wags try!). Instead, it makes more sense to have a flexible classification system for CDPs based on their 5 functions: data storage, customer database, data analysis, personal engagement, and campaign delivery.
SmartOSC offers advice, information and software services for CDPs. Contact us now to learn more and get a CDP.