Overview: Correlation vs. causation in marketing

You've likely heard the phrase "correlation does not imply causation" — in a high school stats or science class, perhaps.

For marketers, correlation is often a key part of their strategy. Consider this directive: "Create a campaign with tactics based on what high-value customers did in the past."

Causation, on the other hand, would consider what caused those customers to become high-value — maybe a specific message or web experience.

For most of the modern martech era, marketers only had access to retrospective and fragmented views of customer behavior. Data arrived late, was often aggregated or sampled, and lived across disconnected systems owned by different teams. The tools built on top of that data were designed to answer descriptive questions like “What happened?” or “What do our best customers look like?” rather than “What action will change this outcome?”

But, with the rise of AI, we have more power than ever to change this methodology.

In our new video series, From Data to Decisioning, we explore the concepts of causation and correlation in marketing, and why causality data is the missing ingredient your team needs to drive meaningful growth with agentic AI.

Watch episode 1 below:

Marketing has gotten really good at predicting customer behavior, but not necessarily at changing it. And that's surprising given that marketing at the end of the day is a causal language.

So let's unpack this a little bit. What is correlative thinking, or correlation, and what is causal thinking? And why are these disconnected? Most importantly, how can we solve for that disconnect and start with correlational thinking?

Correlational thinking

Correlational thinking, correlational prediction means what happens to this customer if we do not change the way we interact with that customer? If that customer lives under the dictatorship of the status quo.

We can already see that this is probably not what ultimately leads to growth.

Causal thinking

Causal thinking asks the other question. Causal thinking asks what happens to the outcome for this customer. 

If we change a lever, for example, what happens to the outcome of that customer? If, instead of showing a generic product, we show a certain product when that customer comes to the site.

Why this matters

Now, why do these two feel disconnected?

They're disconnected because marketers have only had correlational predictions, predictions based on associations at their fingertips to make decisions.

Now, with the rise of AI, we are able to integrate causal thinking into how decisions are done.

And we call that causal decision making.

How causal thinking influences AI

If we want AI to drive growth, to cause growth, it needs more than associative patterns. It needs traces and evidence of causal effects. That is what we call causality data, and it is the key for marketers to achieve that growth thereafter with the help of artificial intelligence.