The Conundrum Between Packaged & Composable CDPs

If we’re living in the “decade of data,” why is data driven marketing still so hard?

The answer lies in the fractured martech landscape.

On one side we have fully featured marketing clouds. Salesforce Marketing Cloud is largely derived from an acquisition they made in 2013 of ExactTarget – a company which prided themselves as “marketers building software for marketers”. 

But, more recently, a secondary camp has emerged with the fundamental premise that data today is incredibly complex – and that maybe marketers shouldn’t be building the software that powers all this. This camp has focused on complex data requirements – but has fallen short of marketer usability north stars that the ExactTargets of the world set out to build.

Let’s dig into what happened and how this split has resulted in a real conundrum for marketing teams trying to align on their martech & CDP strategy.

Path #1: Reverse ETL & The Modern Data Stack. 

The Modern Data Stack has evolved as a set of roughly 1,000+ companies designed to clear, transform, aggregate, analyze, and integrate data. 

At the core of the Modern Data Stack is the Cloud Data Warehouse with the primary players being Snowflake, Bigquery (Google), Redshift (Amazon), and Databricks. 

These tools in some sense represent the ultimate playground for today’s data engineers, analysts, and data scientists. The tech is new, cool, fun, and interesting – and data experts spend their working hours comparing options, evaluating path forwards, and making noise on social media as well. If you’ve heard of the “Composable CDP”, these are the tools that need to be plugged together.

Path #2: Fully Packaged CDPs. 

The roots of CDP as a category lie in data collection and data infrastructure that was built specifically for Martech applications. Today, platforms such as Segment & Tealium have reached a point of maturity where they’re able to satisfy many core marketing and customer facing applications – but they do so at great cost of integration and infrastructure investment. =

The beauty of the Modern Data Stack & centralized data infrastructure is that of building once, build great, and building completely – and then benefiting downstream. Specialized data infrastructure is by its nature duplicative.

With this divergence of data & technology strategy, the decision on which path to go down is a big one – and the implications are real.

Should you go down path #1, align yourself with your enterprise data strategy, and hope you can come out the other side with something that actual works?

Or do you go down path #2, invest heavily in integration & implementation, de risk your ability to get to end value – at the cost of undermining your data investments and unlocking the untapped potential that your organization has spent millions of dollars on?

At Simon, we believe that there’s a “have your cake and eat it too solution” – stay tuned for more next week as we dive into our approach to the category and something we call “Zero ETL”.