Data warehouse, CDP or DMP: Which should you use?

(Too long? Email me this article)
28, May 2020

In the previous articles in this series on data, I described how you can get the most out of email marketing. By linking data sources and using enriched data you can achieve more personal communication via segmentation. If ultimately you want to have that relevant one-on-one dialogue through email marketing, aim for the 360-degree customer profile. The personalisation of content on a large scale is done on the basis of various (data) sources, not just email.

In that case it often comes down to choosing between a CDP, DMP or (marketing) data warehouse. In this final article of the data series I examine the differences between these possibilities.

data warehouse

This series of articles about data consists of 4 parts. In this article, we will address the highlighted part of this model. We will discuss data warehouse, CDP and the DMP. In the other articles you will be able to read about the other parts of the model.

Data warehouse, CDP or DMP: using data in a smarter way

Organisations usually spread out customer data across various source systems. Many organisations process their data in their CRM (Customer Relationship Management) system, which is one such source system. This system may be leading when it comes to the name and contact details of business contacts.

A CRM system does have its limitations however; it is more difficult to keep it up-to-date than you might think. And if you compare the CRM system with an external database you will find that a large number of potential leads are actually not known to your organisation. As a result, you cannot use your CRM to target these contacts, and so you miss out on possible conversions.

Therefore, there comes a moment when connecting various tools is no longer sufficient for acquiring the insights necessary to undertake particular actions. You are then usually left with the choice: shall I opt for a CDP, DMP or (marketing) data warehouse? All the solutions require a significant investment, so the decision is usually not one that is made hastily. Below, I will describe the possibilities, which can be summarised as follows:

  • CDP: is your option if you primarily want to set up and carry out marketing campaigns in order to promote the (omnichannel) customer journey of your own business contacts;
  • DMP: is your option if you mainly want to set up short-term advertising campaigns and have access to cookie data, but not to data on your business contacts;
  • Data warehouse: you would need this if you want to collect data for strategic and tactical analyses, reports and insights.

A CDP: first party data for an optimal customer journey

In a Customer Data Platform (CDP), you use one database with customer data from which you can retrieve a central customer profile. A CDP works mainly with first-party data.

This is data from customers and leads that you know. You obtain this data through your own channels, such as email, a webshop, a shop’s point-of-sale system, marketing automation system, CRM, Facebook page or mobile app. The platform was originally developed to provide optimal support to customers (based on insights into their behaviour) in their customer journey irrespective of the phase they are in.

As you analyse the behaviour of individual customers on the basis of all the channels, you get an insight into various things such as repeat purchases and up-selling and cross-selling information. Using the data you collect, a CDP therefore provides you with a good understanding of the behaviour of individual customers. With this, you can also realise a constantly updated database of customer profiles.

You can further enrich the individual customer profile in your CDP by linking this first-party data with third-party data that a DMP collects (I will explain DMP below). The data that you collect in your CDP will always be available.

Of course, you do have to take into account the retention period stipulated by the GDPR.

A DMP: an extra source of data for accessing (currently) unknown leads

DMP stands for Data Management Platform. Not every organisation needs a DMP in order to take initial advantage of the combination of data. After all, you can often (if on a smaller budget) already take major steps by combining sources of data with one another.

Whereas a CDP works with first-party data, a DMP works mainly with third-party data. This is data from anonymous leads, the profiles of which are based on a cookie-ID. A DMP contributes to the creation of general behavioural profiles in order to be able to target unknown leads more effectively using online advertising campaigns.

A CDP builds a temporary profile based on a cookie-ID. This allows you to communicate more effectively and in a more targeted manner in online campaigns. This information cannot, in contrast to a CDP, be linked directly to an existing customer. Using the data it collects, it provides insight into the behaviour of defined target groups (also referred to as audiences).

You can link third-party data collected by a DMP with, for example, the profiles of loyal customers in a CDP. This makes it so you can effectively target so-called look-a-like audiences. As it involves the storage of large amounts of data, and the aim is to feed brief online advertising campaigns, data in a DMP is often retained no longer than 90 days.

A DMP is therefore an extra data source, which you can link to a CDP or CRM in order to be able to add data to the central customer profile.

Data warehouse

A data warehouse, as the name suggests, is a place that stores large amounts of data. In contrast to a CDP and a DMP, a data warehouse itself doesn’t do anything with the data; it only makes that data available to other systems such as business intelligence tools. Popular data warehouses include Google BigQuery and Microsoft Azure. It also collects data from email marketing/marketing automation solutions, such as Webpower. From here, this data is available for strategic and tactical analyses, reports and dashboards in business intelligence tools such as Power BI.

What is best for my organisation?

Often, a customer already made the choice for either a CDP, DMP, data warehouse – or some other solution –  by the time they approach Webpower for advice on (email) marketing automation. However, if this is not the case we will happily share what we think is best for your organisation. When doing so, we use, for example, our functioning links with Nominow (CDP) and Datatrics (DMP).

“The data puzzle has been laid out”

This was the final article of the data series in which I explained everything about data. I explained why data integrations are used and described the relevant tools. I then outlined the importance of segmentation and data enrichment for email marketing in its goal for conversions, realising commitment/loyalty and/or saving time. Lastly, I explained the differences between a CDP, DMP and data warehouse.

The data puzzle has now been laid out. I hope that I have helped you to better understand the process from A to Z. If this is the first article you have read, why not download the entire series? I will take you step-by-step through data and email marketing.

If you have any questions, please feel free to contact me. I’ll be happy to tell you more about my favourite puzzle and the power of Webpower.

Share this article

Questions about this article?

Sean Barten

​Client Service Director +31 85 773 99 90