I am increasingly experiencing that there are major problems in understanding when do I need a CDP or for what and when do I need an analytics tool (regardless of whether it is Google Analytics, Matomo or similar software). Here I try to bring some clarity to this question.
Basically there is a rough division. (Google) Analytics is well suited to anonymized data for use on the web, primarily for advertising purposes. A CDP allows you to tailor and personalize interactions with customers by merging and using data from multiple sources. You can say that very roughly, but as is so often the case, it’s not everything.
A CDP is a great tool to consolidate customer data in one place and use it to run customer-focused, customer-centric, personalized campaigns. This can be used to collect information about customer behavior and thereby optimize company offerings, uncover gaps in the portfolio or identify new product options.
Analytical tools, on the other hand, are intended to understand data traffic on a website. This in turn makes it possible, for example, to identify visits for a certain period of time, page views, exit pages, etc. It also becomes clear who comes from an organic search and who comes from a paid offer or how many visitors come from which device (mobile, desktop, etc.).
Of course, depending on the manufacturer, there are overlaps in functionality; a CDP may also be able to determine which end device is used to control which channel.
There is a very broad spectrum in the CDP area and many exciting topics that can be presented with it, as well as with the various analytical tools. Especially when these are combined with, for example, Big Query and other tools.
I would also like to go into which typical use cases can be represented with which tool.
CDP(s) and (Google) Analytics have many similarities such as:
- Collect information about what the visitor does on the website
- Can enrich their audience reports by obtaining information from other sources.
- Profile each visitor
- Can enrich audience reports by adding information from other sources.
… but what exactly is the difference? And is a CDP worth it when many analytical tools are available very cheaply or even for free?
That depends on what you want to do with the data.
Basically, analytic tools collect data anonymously, while CDPs collect personal data – so-called PII data – such as email addresses.
Analytics shows you “how many” not “who”. So if you also want to aggregate data historically from, for example, returning visitors or specific customers, a CDP is the tool of choice.
For example, if you want to specifically target people who have accessed a specific page or downloaded a document and then write to them, a CDP is right for you.
If we start with Google’s analytics tool, we can depict the following scenario.
Google knows us very well, probably better than almost anyone else, at least when it comes to data. We can use Google, for example, to retrieve demographic data or enrich it with data or the gender of the visitor. This is not so easy to do with a CDP, unless you have this information in a third-party system and can enrich the CDP with this and then create target groups based on this demographic data.
As already described, target groups can usually be imported and exported into both tools. With an analytics tool, however, this usually only makes sense for advertising.
In contrast, in a CDP, this data can be used to create segments and forward them to any system such as email tools or CRM, since customers are usually known here. This allows personalized messages to be created based on customer data. If a potential customer has viewed a particular product, you can then invite them to test the product on site or something similar.
Dealing with protected areas is also becoming increasingly relevant. While so-called paywalls (or login areas) are outside of most analytics tools, a CDP can manage them very well.
The quality of data is also becoming increasingly important. While analytics tools pursue a rather simple collecting frenzy, the goal of a CDP is to form a single point of truth, i.e. a holistic and uniform picture of a customer. How often does it happen that customers, for example, have several email addresses and are logged in twice or write out their first name once and then abbreviate it next time. This is where a CDP intervenes and merges related data sets. This capability is clearly outside of any analytical tool.
In summary you can say:
If it’s just about web data and it is largely anonymized for e.g. statistics or web-based advertising, an analytics tool is a good choice. When it comes to personalization, you’re better off with a CDP.
Analytics
- Page views
- Campaign information in the aggregate
- Fluctuations in traffic
- Referrers and traffic sources
- Time on page
- Entrance and exit pages
- Bounce rates
- Age and sex of all users
CDP
- Lifecycles and customer journeys
- Paywalls
- Personalization
- Activations with other parts of your tech stack
- Campaign information by customer or specific customer group
- Advanced demographics of known users
Conclusion:
Both tools have their right to exist and are very useful and complement each other very well. What is important is what you want and expect the appropriate tool to help you make the right choice based on that.