Suite, smart hub, and composable: the many faces of a customer data platform

‘Suite vendors’ and ‘smart hubs’

Companies like Microsoft, Adobe, and Salesforce offer entire customer experience platforms, of which a CDP is only one component. The advantage of these ‘suite vendors’ is that they reduce complexity and increase visibility across the board. The downsides: limited integration, and the much-dreaded vendor lock-in. 

Smart hubs like Lytics, Segment, and bloomreach instead offer a unified solution for collecting, cleaning, and activating customer data. Here, customers are free to integrate their CDP with third-party applications of their choice, including a marketing automation platform.

The composable CDP

Another solution that emerged in early 2021 is the composable CDP: a set of integrated, best-in-breed tools assembled by an organisation as alternatives to the key components of a traditional CDP. For example, using Snowplow for behavioural data creation, Databricks for storage and modelling, and Hightouch for data activation. This idea was born from the fact that reverse ETL made it possible for organisations to build their own customer data platform on top of their data warehouse, instead of relying on off-the-shelf solutions.

Proponents of the composable CDP argue that this leads to improved data governance and data quality. Because of their inherent modularity, composable CDPs are also considered more future proof.

What makes a CDP a CDP?

Like most concepts that cross the IT-business barrier, the term CDP has been heavily misused by vendors and marketeers alike in a variety of contexts. However, there are a few key components that make a CDP a CDP.

These include:

• a customer data infrastructure (CDI) to collect behavioural data from first-party data sources;

• an ETL or ELT solution for extracting data from secondary data sources and loading it into the data warehouse;

• a data storage or warehouse to which all CDP components are linked;

• a tool for identity resolution that makes it possible to unify user data from multiple sources;

• a data modelling tool that allows users to build audiences or segments;

• a reverse ETL solution for moving data from the data warehouse to third-party tools or an internal database;

• data quality tools that ensure you’re not basing your activations and decisions on crap;

• data governance tools that enable the necessary checks and facilitate compliance with data and privacy regulations.

Most smart hub CDPs already have the above features, components and capabilities baked in. If you decide to set up a composable CDP, however, you’ll need to stitch all of these together. Here, a capable team of experts is indispensable. Lucky for you, we know some very skilled people.

There are additional complexities as well. For example, a composable CDP will require you to write the unification code for identity resolution yourself. The benefit is that analysts can use whatever methodology works best.

CDP as a service: subscribe to greatness

In brief, a composable CDP is more budget-friendly and provides more transparency and flexibility. But it’s also a very complex endeavour. A smart hub, on the other hand, is expensive, and, well, still pretty complex and time-consuming to deploy.

But what if you could benefit from the features and benefits of a smart hub CDP, without the worries? Create 360-customer views, segment audiences, personalise customer experience, test and optimise your webpages and applications, predict churn, push next-best offers and whatnot without any complexity concerns, and for the fraction of the price of a full-blown, off-the-shelf CDP?

Enter CDP as a service: our subscription solution powered by Lytics and MultiMinds. All the capabilities of Lytics’ composable CDP, backed by a full-service team, for a fixed monthly fee and a 6-month commitment.