Marketing Is the Worst at Using Its Own Data

Marketing is the discipline built to understand people better than they understand themselves. We study behavior, model audiences, and track how a stranger moves from curiosity to conversion.

Then we turn around and make our own budget decisions based on gut instinct and a dashboard that three different platforms built three different ways.

The Cobbler's Children Have No Shoes

There is an old saying about the cobbler whose children go barefoot. He spends all day making shoes for everyone else and never gets around to his own family.

Marketing has the same problem.

A Qualtrics study of more than 700 global senior marketing executives found that many rely only on their intuition for critical business decisions because they lack timely and relevant data. These are not junior marketers guessing. These are the people deciding where millions of dollars go.

The irony is hard to ignore.

Fragmentation Is the Real Issue

If you have run any kind of multi-channel operation, you already know the feeling. Ads are in one platform. Email metrics live in another. Website behavior is in GA4. SEO is in Search Console. That lead who clicked an ad, read three blog posts, opened two emails, and converted three weeks later? You probably credited the last touch.

Today's consumers interact with brands across an average of 9.5 touchpoints before converting. Each platform operates within its own measurement ecosystem, making comprehensive campaign effectiveness analysis nearly impossible.

And every one of those platforms has a financial incentive to claim the credit. A 2024 Wharton study found that channel silos lead to an average overestimation of marketing performance by 23 to 31 percent.

That is not a rounding error. That is the difference between scaling a channel and cutting it.

False Positives Are the Dangerous Part

When data is fragmented, you do not just miss information. You act on the wrong information.

A channel that looks like it is underperforming might be doing the early work that something else eventually closes. A campaign that looks like a winner might be capturing existing demand rather than creating new demand. The story your dashboard tells depends entirely on what your tools can see, and right now most of them cannot see very much.

Privacy changes and fragmented user journeys have blurred the picture of what truly drives growth. Companies expect AI to solve it all, but it does not.

Marketers cut what looks bad and scale what looks good, based on an incomplete read of both. That is not strategy. That is noise with a budget attached.

Where Decision Intelligence Fits

The traditional response to this problem is to add another tool. Another attribution layer. Another report. But more reporting does not fix fragmentation, it compounds it.

Decision intelligence connects the underlying systems and surfaces what the data actually says, with the source attached. In my own work, the value is not another place to look at metrics. It is having a single place where context across channels is actually connected. A question like "what drove signups this month" should not require opening six tabs and triangulating manually. That question should have a traceable answer.

It is a small example. But it points at a real shift in how marketing decisions get made.

The Gap Is Not Going Away on Its Own

According to The CMO Survey from Spring 2025, board pressure on marketing leaders rose 21 percent from 2023 to 2025, with CFO pressure up 52 percent. The expectation is not just good campaigns anymore. It is a clear explanation of why the business grew or did not.

That expectation is fair. The infrastructure most teams are working inside is not built for it.

Marketing has spent years making the case that it is a revenue function. Making that case requires being able to read your own results with the same rigor you apply to understanding your customers.

The cobbler's children are still barefoot. The shoes exist. Someone has to decide to make them.

Composition and research assisted by AI. Observations and opinions are my own. I would love to hear others thoughts on this topic and the research behind it.