Why I Don't Trust Ecommerce Dashboards Until I've Seen the Data Layer

You can spend £3M on a new ecommerce platform, invest heavily in Google Ads and build beautiful dashboards in GA4.

None of it matters if the data going into those dashboards is wrong.

Over the years I've learnt to be surprisingly sceptical whenever someone proudly shows me a conversion report. Not because I don't believe the numbers, but because I've seen how easily they can become detached from reality.

I've worked on enough ecommerce platforms to know that what appears on a dashboard isn't always what's actually happening on the website.

More often than not, the problem starts with something most businesses never discuss: the Data Layer.

The meeting that changed my thinking

A few years ago I sat in a meeting where everyone was celebrating a strong month.

Conversion rates were up. Return on advertising spend looked healthy. Marketing campaigns appeared to be outperforming expectations.

Then someone compared the figures with the finance system.

They didn't match.

After a bit of investigation, the culprit turned out to be duplicated purchase events. The website had quietly been telling Google Analytics that customers were buying twice.

Nobody had noticed.

The dashboards looked fantastic.

The business wasn't.

That experience taught me an important lesson. Reports are only ever as good as the information they're built on.

So what actually is a Data Layer?

The easiest way to think about it is as a translator.

Every time someone uses your website they generate hundreds of actions.

They view a product.

They search.

They add something to their basket.

They remove it again.

They choose delivery.

They complete a purchase.

Your analytics platforms can't automatically understand all of that. Your website has to explain what has happened in a structured way.

That's exactly what the Data Layer does.

It packages up those events and hands them to tools like Google Tag Manager, which then passes them to GA4, Google Ads, Meta, Microsoft Advertising or whichever platforms you're using.

Get that foundation right and every reporting platform is working from the same version of events.

Get it wrong and every system starts telling a slightly different story.

The problems I see time and time again

When I review ecommerce implementations, the same issues crop up with surprising regularity.

Purchase values don't match the actual order value.

Discounts disappear.

Shipping is treated differently between reports.

Product IDs change between systems.

Add to Basket works on desktop but not mobile.

Purchase events fire twice.

Refunds never make their way back into analytics.

Sometimes everything appears to work perfectly until someone compares GA4 with the finance system.

That's usually when the uncomfortable questions begin.

Why this is commercially important

People often think inaccurate tracking is an analytics problem.

It isn't, it's a commercial problem.

Google Ads decides who to show your adverts to based on conversion data.

Meta does exactly the same.

If those platforms are being fed incomplete or inaccurate information, they'll optimise towards the wrong customers.

You might increase spend on a campaign that never really performed.

You might stop investing in one that actually did.

The more automated digital marketing becomes, the more important good quality data becomes.

Google Tag Manager isn't the clever bit

One misconception I still hear is that installing Google Tag Manager somehow fixes tracking.

It doesn't.

Google Tag Manager is simply the delivery mechanism.

Think of it as the courier.

If the parcel contains the wrong information, it doesn't matter how efficiently it's delivered.

The hard work happens before Google Tag Manager ever gets involved.

What I expect every ecommerce site to capture

A healthy Data Layer should reliably record the things that genuinely matter to the business.

That includes product views, basket activity, checkout steps, purchases, refunds, product identifiers, quantities, revenue, discounts, promotions and customer interactions.

Just as importantly, it should record them consistently across desktop, mobile devices, guest checkout and logged in customers.

Consistency is far more valuable than collecting hundreds of different events you'll never use.

Before you redesign your website...

Whenever I'm involved in an ecommerce project, one of my first questions is always the same.

Can we trust the data?

Not whether the dashboard looks good.

Not whether GA4 has lots of reports.

Not whether someone has installed Google Tag Manager.

Can we trust the information that's driving commercial decisions?

Because if the answer is no, every conversation about conversion rates, return on advertising spend and customer behaviour becomes little more than educated guesswork.

In ecommerce, good decisions begin long before someone opens a dashboard.

They begin with getting the Data Layer right.