Most GA4 setups are quietly wrong
Not broken in an obvious way. Broken in the way that still produces a dashboard full of confident numbers, which everyone then uses to make decisions.
The usual suspects: conversions counted twice, form submissions firing on page load, thank-you pages counted as goals when people bookmark them, internal traffic never excluded, phone calls invisible, and offline sales, the ones that actually pay the bills, nowhere in the system at all.
A dashboard that is confidently wrong is worse than no dashboard. At least no dashboard makes you cautious.
What good analytics consulting actually fixes
- A measurement plan. What matters to the business, before anyone touches a tag. This step gets skipped constantly and it is the whole job.
- Events that mean something. Google publishes recommended events for good reason, a well-named, consistent event structure is what makes the data usable later.
- Conversion values, not just counts. An enquiry worth $50 and one worth $50,000 are not the same conversion, and the platforms will optimize as if they are.
- Offline and phone closing the loop. For most service businesses this is where the actual revenue lives.
- One report people read. Not eleven dashboards nobody opens.
That third point quietly transforms paid media. Feed the ad platforms a value rather than a count, and their automated bidding starts chasing revenue instead of activity. Most accounts never do it.
The dashboard nobody reads
A familiar shape. A business has plenty of data (ads, website, email) but it lives in five different tools, each with its own version of the truth. Nobody has time to reconcile them, so the monthly conversation defaults to "I think it is going okay".
Pulling it into one clear view changes the conversation from a feeling into a decision. It also tends to reveal the awkward thing: two of your channels have been taking credit for the same conversions for months, and your best performer was never the one you thought.
One clear dashboard beats five open tabs and a gut feeling. That is the whole pitch, and it is why we treat this as the first job rather than the last.
When you do not need an analytics consultant
- You have almost no traffic. With a handful of visitors, you do not have an analytics problem. You have a traffic problem. Go get some.
- Your setup is already clean and you act on it. Rare, but it happens. If your reports match your bank account, you are fine.
- Nobody will change anything regardless. Data you do not act on is just expensive decoration, and better data you still ignore is more expensive decoration.
But if you are running paid ads on top of numbers you privately do not trust, this is the highest-return thing you can fix. Everything downstream (bidding, budget decisions, which channel gets cut) depends on it being right. That is what our analytics and tracking work exists for.
The audit you can run yourself this afternoon
Before you pay anyone, run these five checks. They take under an hour and they will tell you whether you have a problem worth paying to solve.
- Compare conversions to reality. Pull last month's recorded conversions. Ask your team how many enquiries they actually received. If those numbers disagree by more than a rounding error, everything downstream is fiction.
- Check internal traffic is excluded. Your own team browsing the site should not be inflating anything.
- Look for conversion values. Not just counts. If every conversion is worth "1", your ad platforms cannot tell a tyre-kicker from a whale.
- Find the phone calls. For most service businesses this is the majority of enquiries. If they are not in the data, most of your business is invisible to your reporting.
- Check event naming. Google publishes a recommended event structure precisely so data stays interpretable. Improvised names produce a dataset nobody can use in a year.
If three or more of those fail, which is common, you do not need a consultation. You need the tracking rebuilt, and you should be wary of anyone who wants to sell you a strategy session on top of data they have not verified.
The good news is that this is a bounded problem. Tracking is not an ongoing retainer by nature: it is a job that gets done, checked, and then quietly keeps working. Anyone selling it as a permanent monthly cost should explain what keeps breaking.
Why tracking keeps getting harder
If your data has quietly degraded over the last few years and nobody could explain why, this is why. The ground moved.
Browser restrictions, ad blockers, consent requirements and platform privacy changes have all made client-side tracking less reliable. None of this was a mistake on your part, it happened to everyone, gradually, without an announcement. But it does mean a setup built five years ago is measuring less than it used to, and is probably reporting confidently anyway.
- Expect gaps, and know where they are. A model that admits uncertainty beats one that fabricates precision.
- Server-side tagging recovers some of what browser restrictions cost you. It is more work and it is not magic, but for accounts spending real money it usually pays.
- Get consent right. It is a legal question before it is a technical one, and the technical implementation should follow the legal answer, not the other way round.
- Anchor everything to reality. Whatever the tools say, the enquiries your team received last month are the ground truth. Google's GA4 event guidance and Tag Manager are how you close the gap between the two.
The goal is not perfect data. Perfect data no longer exists. The goal is data that is honest about its limits and still good enough to make the decision in front of you, which, for most businesses, is entirely achievable and a long way from where they are now.
Frequently asked questions
What does a Google Analytics consultant do?
Builds a measurement plan tied to what the business actually cares about, sets up events that are consistent and meaningful, sends conversion values rather than raw counts, connects phone calls and offline sales, and produces one report people read instead of eleven nobody opens.
Why is my GA4 data wrong?
Usually a handful of common faults: conversions counted twice, forms firing on page load, internal traffic not excluded, thank-you pages triggering goals when bookmarked, and phone or offline sales missing entirely. None of it looks broken, it just quietly produces confident numbers that are not true.
Why do conversion values matter more than conversion counts?
Because an enquiry worth $50 and one worth $50,000 are not the same event, and automated bidding will treat them identically unless you say otherwise. Feeding platforms a value rather than a count is what makes them chase revenue instead of activity. Most accounts never do it.
Do I need analytics consulting if I have very little traffic?
No. With very few visitors you have a traffic problem, not a measurement problem. Get the traffic first. Analytics consulting pays for itself when there is enough spend and enough volume that being wrong is genuinely expensive.