Customer data platform
A customer data platform (CDP) is a central system gathering customer data from different sources, unifying it and bringing it together into a lasting, unified customer profile. Unlike a classic CRM, which manages known contacts above all, a CDP also processes behavioural data from website, app, email and advertising platforms and makes it usable for marketing, sales and service. A CDP thus solves the problem of isolated data silos, in which information about the same person lies scattered and contradictory. The result is a consolidated view of every customer serving as the basis for personalisation, segmentation and cross-channel campaigns. The term was shaped largely by the CDP Institute, which defines a CDP through three features: it is controlled by marketing, creates persistent, unified profiles and makes that data available to other systems for activation. That ability to activate is exactly what distinguishes a genuine CDP from a mere data store and makes it an operational tool that does not only gather data but makes it immediately usable for marketing and sales.
Also known as: customer data platform, CDP
How a CDP works technically
At its core, a CDP has three functional layers. The first collects data via interfaces, SDKs and tags, often orchestrated through Google Tag Manager or direct APIconnections to shop, CRMand email systems. Both real-time events and periodic batch imports from data warehouses feed in.
The second layer handles identity resolution: it links anonymous and known identifiers such as cookie IDs, email addresses and login credentials into one profile. Through matching rules the CDP recognises that a visitor with several devices and addresses is the same person and merges the data points deterministically.
The third layer makes the enriched profiles and segments available again. Through reverse ETL or native connectors the data flows back into ad platforms, email tools or Marketing dashboards. Established CDPs such as Segment, Tealium or the SAP and Adobe solutions differ mainly in the depth of this activation layer and in real-time capability.
CDP in relation to CRM, DMP and data warehouse
A CDP is no substitute for a CRM but complements it. The CRM stays the system for maintained contacts and sales processes, while the CDP additionally handles behavioural data and event streams and connects both worlds. Importantly, a CDP puts first-party data at the centre, that is, data from your own direct relationship with the customer.
A CDP differs from a data management platform (DMP) in its focus on personal, long-term stored profiles rather than anonymous, short-lived cookie segments. Compared with a pure Data warehouse the CDP brings ready-made marketing logic: it is built for activation, Segmentation and real-time response, not just analysis and reporting.
In practice many companies use a combination: the data warehouse holds the raw data, the CDP prepares it operationally and distributes it to the activating channels.
GDPR requirements for a CDP
A CDP processes personal data by definition and is therefore fully covered by the GDPR. Even at collection: behavioural data may generally only be gathered and fed into the CDP after valid consent through a consent tool such as Usercentrics or Cookiebot. A consistent, GDPR-compliant tracking is therefore a prerequisite for filling the platform lawfully at all.
Purpose limitation is central: for every category of data it has to be documented for what purpose it is processed and what legal basis applies. The CDP should carry the consent status so downstream systems receive only data cleared for release. Access and deletion requests have to be actionable across every linked profile, which a CDP's central architecture eases.
A data processing agreement with the provider is advisable, as is a data protection impact assessment, since merging many sources carries heightened risk. With US providers, transfer to third countries has to be checked and safeguarded separately.
Typical use cases
The most common use case is personalisation across channels. If the CDP sees that a customer has viewed a product several times but not bought it, it can automatically trigger a suitable email or a Retargetingsegment without copying data manually between tools.
A second area is segmentation for targeted campaigns. From the unified profiles, precise audiences can be formed, such as customers with a high basket value or those close to churning. These segments flow straight into ad platforms and feed data-driven decisions.
Third, the CDP serves as a reliable data source for marketing dashboards and analysis, because all channels are evaluated on the same cleaned profile basis. Before adopting one, it is worth a Performance audit, to clarify which sources are really needed and where data silos do the most damage.
A fourth, often underestimated benefit is the improvement of data quality itself. Because the CDP normalises incoming data streams, merges duplicates and resolves identities, it raises the level of every connected system. Instead of each advertising or email tool working with its own, partly contradictory lists, every channel draws on one shared, maintained truth. That reduces waste, avoids duplicate communication and makes the effect of individual measures measurable across the whole customer lifecycle.
Frequently asked questions
What is the difference between a CDP and a CRM?
A CRM manages known contacts and sales processes above all. A CDP adds behavioural and event data from website, app and advertising channels and merges anonymous and known identities into one unified profile. The CDP does not replace the CRM but enriches it and distributes the data to activating systems.
Is using a CDP GDPR compliant?
Yes, provided the processing rests on a valid legal basis. Behavioural data generally presupposes consent through a consent tool. The CDP should carry the consent status per profile, allow rights of access and deletion and be covered by a data processing agreement. With US providers, the transfer to a third country is to be examined as well.
How does a CDP differ from a DMP?
A DMP works mainly with anonymous, short-lived cookie segments for serving ads. A CDP, by contrast, stores personal, long-term profiles and makes them available operationally for personalisation and cross-channel campaigns. The CDP is thus oriented more to first-party data and lasting customer relationships.
Do I need a CDP if I already have a data warehouse?
A data warehouse holds raw data for analysis and reporting but brings no ready-made marketing logic. A CDP adds prebuilt functions for identity resolution, segmentation and real-time activation. Many companies combine the two: the warehouse stores, the CDP activates the data in the marketing channels.
Which data can a CDP bring together?
A CDP integrates website and app events, transaction data from the shop, CRM contacts, email interactions and campaign data from ad platforms. Through identity resolution, cookie IDs, email addresses and login identifiers are linked, creating one consolidated profile across several devices and sessions.
How much work is rolling out a CDP?
The effort depends on the number of source systems and the data quality. One to several months for connection, data modelling and identity rules is typical. A performance audit beforehand helps prioritise the genuinely relevant sources and avoid unnecessary complexity.
Related terms
First-party data is data a company collects directly from its own users, with their consent.
Dividing a heterogeneous market or customer base into homogeneous groups so that messages and measures can be targeted.
Event tracking records individual user actions — clicks, scrolls, downloads or purchases — as measurable events for web analytics.
A marketing approach that bases strategy, campaigns and optimisation systematically on data and metrics rather than intuition.
Consent management organises how cookie consent is obtained, managed and documented in a legally sound way.
Server-side tracking processes data through your own server container rather than in the browser alone.
Put AI to work for your business?
We help you integrate artificial intelligence into your processes, your marketing and your website — strategically and securely.
Your contact
Stefan
I look forward to hearing about your project and finding the best solution together.
