Here’s something that keeps me up at night: we’re bombarded with data, yet making worse decisions than ever before.
That’s a bold claim, I know, but stay with me. Over three decades of working with data, I’ve watched a concerning trend unfold. The democratization of data – which in many ways has supercharged our industry – has inadvertently created a false sense of expertise that’s leading marketers astray.
I’m not suggesting we go back to the days of limited data. That would be ridiculous! But I am suggesting that giving everyone access to data without also equipping them with the necessary critical thinking skills to interpret it has created a rather spectacular mess.
Click and Click-Through Rates
Take clicks and click-through rates. These metrics were comprehensively debunked over a decade ago. Comscore ran a correlation analysis that showed – wait for it – a negative correlation to actual outcomes. Nielsen’s data supports this. Yet here we are in 2026 and I’m still having conversations with incredibly smart people at agencies and brands who are using CTR as a primary success metric.
I understand the appeal. More clicks should mean more sales, right? It’s intuitive. Unfortunately, it’s also not true. And the data proving it’s wrong has been available to us for years.
This isn’t a matter of opinion or my personal bias (though I’ll confess I find the persistence of CTR particularly maddening). This is an observable, measurable reality that is being collectively overlooked.
The problem is that today’s marketers are expected to do more than ever before, juggling functions that weren’t traditionally in their remit. A short module on data analytics at university is a great foundation but probably doesn’t qualify the student as a Data Scientist.
And this skills gap creates a dangerous situation: marketers under pressure, short on time and lacking the specialist training needed to interpret complex data properly. So what happens? They default to metrics that feel familiar, even when those metrics have been proven unreliable.
I understand the appeal. More clicks should mean more sales, right? It’s intuitive. Unfortunately, it’s also not true. And the data proving it’s wrong has been available for years.
This is exactly how well-established metrics like ad recall get misunderstood and misused. At a recent event, we ran a quick poll asking marketing professionals which measurement technique was most reliable: passive exposure measurement or ad recall-based measurement. A massive 68% chose ad recall. These weren’t junior coordinators – these were brand managers, insights directors and CEOs.
What the Research Tells Us
When we recently analyzed over 2,000 brand lift studies comparing ad recall responses to passive exposure tracking, ad recall was wrong in eight out of 10 cases. Put simply, people said they saw ads they didn’t actually see or forgot ads they did see – meaning your measurement groups are contaminated from the start.
When we also looked specifically at customer bias, we found that customers were 3.1 times more likely to claim they’d seen an ad compared to non-customers, despite both groups having comparable actual exposure.
Think about what this means. If you’re using ad recall to measure brand lift, roughly 7 out of 10 people in your “exposed” group are likely existing customers. You’re not measuring campaign impact – you’re measuring existing brand relationships. It’s like asking Cold Storage customers if they’ve heard of Cold Storage and calling it research.
Of course, you might expect me to bang on about measurement given my role (and I realise I’ve done exactly that) but this isn’t really about measurement methodology. It’s about business outcomes.
The fundamental question our industry asks itself should always be: what are we trying to achieve? Am I building brand, or driving sales? Because “this metric moved” is not an outcome. It’s just noise unless you can link it to something that matters to the business.
Rebuilding Critical Thinking
I’ve sat in too many meetings where people present data showing improvement on some metric or another, and when you ask “What does that mean for the business?” The room goes quiet.
We’ve become so focused on generating reports that prove something happened that we’ve forgotten to ask whether that something was actually valuable.
The solution is to rebuild critical thinking into our decision-making processes. Question the metrics you’re using. Ask for proof that they correlate to real-world outcomes.
The solution isn’t to restrict access to data – that ship has well and truly sailed. By and large, most of our roles would not be possible without the wealth of data we now have.
The solution is to rebuild critical thinking into our decision-making processes. Question the metrics you’re using. Ask for proof that they correlate to real-world outcomes. Don’t accept “we’ve always measured it this way” as justification for anything.
Here’s what I’ve learned: knowing when to bring in specialist expertise isn’t a sign of weakness, it’s a sign of good judgment. We all have our strengths and data analysis is a specialist skill like any other.
The data is out there. The insights are available. We just need to be honest about whether we have the expertise to interpret them properly – and humble enough to ask for help when we don’t.
