Alma Ochoa

Data Analytics Consultant

Customer Data Platform: How to Handle First-Party Data

6 November 2021
8 min

What is a Customer Data Platform (CDP)?

Before we jump in, we first need to understand its current context, the issues we currently face in terms of capturing data, and the diverse solutions for moving forward. Let’s first discuss the current context:

  • Regulatory changes: the General Data Protection Regulation (GDPR) in Europe and the California Consumer Privacy Act (CCPA) have impacted how data can be stored, collected, and used.
  • Updates: companies are no longer able to harness traditional data collection methods (for example: the third-party cookies and device identifiers) due to the increase in control.
  • Privacy control: Today, users demand more control and transparency over the data collected about them, which is then used for ad personalization.

The main and direct consequence in today’s digital ecosystem is clear:

  • Cookieless: we are no longer able to track our users nor their activity, behaviors, nor interests
  • Cookieless world third-party: limitations like Intelligent Tracking Prevention (or ITP, a Safari feature) – or Enhanced Tracking Protection (or ETP, a Mozilla feature) mean that our marketing strategies based on third party audiences and cookies will be the most affected.

There are diverse solutions for solving the cookieless puzzle. Users now have the option to choose how their data is collected. However, with Consent Mode and GA4, we can mitigate the loss of data from our users through data modeling. For example, we can use ServerToServer measurement (which means more first party data). Most importantly, this grants us ownership of how this data is stored.

Here, as we confront the concept of first party data, is where the concept of CDP is reasonable, because it is a technological solution for the management of our data as well as its subsequent activation. CDP should be understood as a solution – not just as a tool. 

Why use a CDP?

There are many different tools in the digital market that store and activate data. However, from a Making Science perspective, we recommend CDP. This is how it can be approached from a client’s perspective.

This is a simplified diagram of how a CDP works:

Let’s first recognize that we presently have a lot of information (much more than we can usually analyze and activate). Additionally, this information is stored and created by multiple data sources, creating silos which rarely interact with each other. This causes data duplication, because there could be different sources with the same user information. As it’s processed, the data can become blurred, generating a lack of information about attribution and purchase data. This makes it difficult to administer the marketing mix budget and nearly impossible to optimize the investment.

A CDP will allow us to have one repository for all our data, whose initial purpose is to standardize and unify customers’ information according to all the user touch points generated. This will allow us to activate the data on a per-customer basis, so we can meet their needs according to the time of purchase or the point of maturity of the customer with the brand.

The significant advantage of a CDP is that it allows us to work with the data as if it were First Party Data (as if we were the owners of the data). It is important to have a measurement strategy related to solutions such as the aforementioned Consent Mode or ServerToServer. Since a CDP requires several development iterations we’ll have to question ourselves the following in this second stage:

  1. What goals do we hope to achieve working along with a CDP? What kind of problems could we face with the platform? At the end of the day, it is just a technology solution, however we must first identify what problems we want to solve.
  2. Which CDP should I use? How can I get it?

Factors to consider

For companies with a high level of technological development, it may be feasible to generate a proprietary CDP from their stack of tools. Others may prefer to go to the market in search of solutions that have already been developed. In any case, we suggest the following four factors to take into account when making this decision:

  • Timing: What’s the fastest route? When will my CDP be functional? For marketplace solutions, it will depend on the connection options with the tools of your digital ecosystem. For an ad hoc option, one must consider what kind of connection options exist as well as the effort needed to create them. This leads us to the second factor:
  • Connections: which option will allow us to link easily with our current stack of tools? If your company already works with a market solution, this embedded connection should be easier; otherwise you should evaluate the effort to generate those connections out of the box.
  • Scalability: Which connection option will grant the most flexibility to adapt to changes in your company? For market solutions, it will depend on the evolution of their adaptations. An ad hoc one should be easier for you in this respect.
  • Cost: What is the most cost-effective solution? Consider not only the initial monetary cost, but also the cost of CDP evolution according to your main objectives and stack of tools.

In conclusion, these are just a few hints of what CDP implies as another solution in the current digital landscape. It is not a simple decision to make for companies. At Making Science, we recommend CDPs because we understand how this solution should be approached: always focus on generating solutions according to the current market and the company’s particular needs.