← All Insights

Why Your Marketing Data Is Probably Wrong (And What to Do About It)

Sayf Sharif
Sayf Sharif
President & Co-Founder · March 30, 2026

I’ve been in more than one meeting where the following happens…

We’d all be in a big room, say doing a multi-agency annual review or the like, where everyone is showing “green” numbers. Social, performance, content, you name it, every platform metric (or the vast majority) appear to be trending up. Green lights across the board!

Meanwhile, while everyone is looking at the presentation up on the big screen I would be looking at the CMO, and/or other responsibility holders. More often than not they didn’t look satisfied, they looked uneasy. It’s only gotten worse over time.

Heck, back in the day I had someone brag about how their “bounce rate was only 4%” not realizing that they absolutely and with complete confidence had terrible data, and tracking errors on their site. Now I’m more likely to talk to people who have pretty dashboards telling them pretty numbers, and they look at me and shrug their shoulders, and basically look resigned to whatever this world is.

Maybe you’ve been in a room like that, and you can’t quite put your finger on it. The numbers might look fine, everyone is doing a good job, but at the end of the day the business isn’t growing the way the numbers you have indicate it should.

Something is wrong, and your instinct that it’s wrong is correct.

The Structure Is the Problem

This isn’t about agencies doing bad work, or platforms misreporting or misrepresenting numbers, or that your internal team is dropping the ball. Most of the time, everyone is doing exactly what they’re supposed to do. The problem is structural, and it runs alot deeper than most people realize.

Every agency you work with is accountable for their channel. Your paid search partner optimizes your paid search performance. Your SEO/GEO agency reports on your organic and reach. Your social team shows you engagement. Everyone is telling you the truth… About their slice of the pie.

(Oh and even when you have a single agency doing multiple things, in the majority of agencies, those teams are different people, and just because you did agency consolidation and have one invoice, it doesn’t mean those teams at your mega agency are actually “working together”.)

Nobody tends to be accountable for the whole picture.

When nobody owns the whole picture, you are going to end up with a set of numbers that are individually defensible, and collectively misleading. It’s not dishonesty, it’s incentive misalignment baked into how our entire industry is structured. Agencies get paid to perform on their channel. Platforms get paid when you spend more on their platform. Everyone’s metrics are optimized to tell you a story that justifies their existence.

That’s just the reality, not a criticism. It means the job of connecting all your stories together into one single coherent truth falls on YOU, and most companies don’t have anyone actually doing that.

If anything it’s probably more common to have not only multiple agencies, platforms, and teams, but to have multiple owners, stakeholders, and business leaders who have different asks of your data.

Misaligned incentives on the outside, misaligned incentives on the inside. Misaligned incentives all the way through.

Three ways your data is lying to you right now

Your attribution is probably wrong. Most companies are still running on last-click (or sometimes first-click) attribution models, which means that they’re systematically over-crediting one touchpoint and undercrediting everything else. The channel that’s “closing” the customer is your hero, but the channels that built that relationship over the past six months all look like they’re underperforming. Budget shift, the wrong things get cut.

“How are you measuring the effectiveness of that channel?” I asked someone recently.

“We’re not.” they said.

They’re not the only person that said something like that to me recently, that one just hit harder because of how much budget they were putting into the channel, with literally no real data on whether it was helping them or not.

Your CRM data is rotting. This is a slow and silent killer. Data like your contacts go stale. Fields get skipped during data entry, lead sources get miscoded, duplicate records multiply. Garbage in, garbage out. It’s a slowly compounding problem though so it’s a bit closer to the frog in the boiling water metaphor. You won’t notice it at first, but it will accumulate so slowly over time you won’t realize that you’ve been making bad decisions for months, or longer.

\You're measuring activity, not outcomes. Whether it’s impressions, clicks, opens, engagement, whatever, you’re measuring real things that have real numbers attached to them. But they aren’t revenue (or whatever your actual desired outcome is). The gap that’s between “our content performed well” and “our content drove pipeline” is where marketing budgets slip through the cracks and are left to die.

One thing you can do THIS week

Figure out if you can find a single meaningful number that you can report every month to your leadership. Just one number that you can trace all the way back to its source. Can you actually verify the number? Do you know how it’s calculated? Do you know where the data comes from, and whether it’s consistent month over month?

If just thinking about that exercise makes you feel uncomfortable, that’s useful information.

The real problem isn't the data

Bad data, garbage data, siloed good data, are all symptoms.

The real problem is that most mid-sized companies don't have anyone whose job it is to own their unified truth. Not a tool or a dashboard, a person who has the context, access, and the authority to look at your whole picture and say “here’s what’s actually happening.”

It’s a leadership and structural/organizational problem, but it’s one that’s entirely solvable.

If you want to know where your data stands today, we built Base12 a free data maturity assessment that takes less than five minutes, gives you a score across the dimensions that matter most, and shows you where the gaps are.

The first step to fixing your data is knowing which parts of it you can actually trust.

Want to discuss this topic?

Schedule a Call