A clean dashboard can still lie to you.
That is the problem.
You log into GA4. You see charts. You see sessions. You see revenue. Everything looks polished enough that nobody questions it too hard.
Meanwhile:
- revenue does not match the platform
- purchases are duplicated
- attribution feels weird
- paid traffic looks better or worse than it should
- nobody can say with confidence where customers are dropping off
This is more common than people think.
And it is exactly why a tracking audit matters.
Because once bad data starts getting used in meetings, forecasts, and channel decisions, the damage multiplies fast.
The biggest mistake people make with GA4
They assume “installed” means “working.”
It does not.
You can connect GA4, see traffic flow in, and still be missing the data that actually matters.
Ecommerce businesses do not just need page views. They need clean events. Clean attribution. Clean purchase data. Clean funnels.
The live Ecomm Decoded Tracking Sprint is built around this exact problem: GA4, event tracking, custom dashboards, conversion funnels, and figuring out where customers drop off. It is for stores that cannot trust the numbers enough to make decisions.
If that sounds familiar, start with this audit.
1. Does revenue roughly match the platform?
I am not looking for perfect twins.
Different tools will always have small gaps.
What I want to know is whether the data is directionally trustworthy.
If Shopify says one number and GA4 says something wildly different, you have a problem.
Common reasons:
- duplicate purchase events
- missing purchase events
- timezone mismatches
- wrong property or stream
- broken checkout tracking
- consent setup interfering with data capture
If revenue is off, stop there first. Do not move on to advanced reporting until this is clean.
2. Are the core ecommerce events actually firing?
I want to see whether the basics are there:
- view_item
- add_to_cart
- begin_checkout
- purchase
And I want them firing when they should, not randomly, not twice, not missing.
This matters because ecommerce reports are only useful if the event chain is real.
On Ecomm Decoded’s tracking intake form, the live problems listed are very specific: revenue not matching the platform, duplicate purchases, missing events, attribution feeling wrong, ad signals looking off, and consent banners interfering. That is exactly how a good audit should think — starting with symptoms that affect decisions.
3. Are you double-counting anything?
This is one of the fastest ways to make a dashboard useless.
Sometimes a platform connection is firing. Then a manual tag is also firing. Then another app is helping “improve tracking.”
Now everything is doubled.
Twice the purchases. Twice the revenue. Twice the false confidence.
If your numbers look weirdly inflated, do not assume it is a good month.
Check the setup.
This is why I liked making GA4 content for Ecomm Decoded in the first place. One of the live articles already points out that duplicate events will destroy your data because you stop being able to trust the numbers. That part is not dramatic. It is true.
4. Is attribution believable?
Attribution is where a lot of teams quietly lose the plot.
Paid says one thing. GA4 says another. Shopify says something else. Now everyone starts defending their channel instead of fixing the setup.
Here is what I check:
- are UTMs being used consistently?
- are paid traffic sources coming through cleanly?
- is there self-referral noise?
- are campaign names readable and consistent?
- are channels getting bucketed properly?
You do not need perfect attribution to make smart decisions.
You do need attribution that is believable enough to compare channels honestly.
5. Is the funnel telling a clear story?
You should be able to answer simple questions fast:
- how many people viewed a product?
- how many added to cart?
- how many began checkout?
- how many purchased?
- where is the biggest drop-off?
If it takes 20 minutes and three tools to answer that, your tracking setup is not helping enough.
This is where clean event structure and a few useful reports matter more than endless dashboards.
6. Are your ad signals getting clean data?
If Meta, Google, or TikTok are connected to messy data, they optimize on messy signals.
That means you are not only reading the wrong numbers. You are also feeding the platforms the wrong picture of performance.
Check:
- pixel and tag installation
- purchase signal quality
- duplicated events across ad tools
- consent impact
- basic event alignment between platforms
This is not just a reporting issue. It is a performance issue.
7. Is consent setup blocking more than you realize?
Consent banners matter. Privacy matters.
But a poorly implemented consent setup can also break tracking in ways teams do not notice until much later.
If the banner or consent tool is interfering with page views, purchases, or tag firing, that has to be fixed properly.
Not worked around. Fixed.
8. Are you tracking too much useless stuff and not enough of the right stuff?
This one is subtle.
Some setups are not broken. They are noisy.
Too many events. Too many dimensions. Too many reports. Not enough clarity.
I would rather have:
- a smaller set of reliable events
- a clean funnel
- channel data you can trust
- a simple dashboard people actually use
That beats a giant analytics setup nobody believes.
9. Can the team explain the setup without guessing?
This matters more than people think.
Ask the team:
- what tools are installed?
- which tags are live?
- who owns changes?
- what happens when something breaks?
- where is the documentation?
If nobody knows, the setup is fragile.
That is one reason documentation is built into the Ecomm Decoded sprint model. The point is not just to fix things once. It is to hand them back clean so the team can maintain them.
10. Are you making business decisions off reports you have never validated?
This is the last checkpoint.
Open a private window. Go through the site. Trigger the important actions. Check whether the right events fire. Then compare that to the reports.
Validation beats assumptions every time.
My simple GA4 audit order
If I had one hour, I would review GA4 in this order:
- revenue versus platform
- purchase event quality
- core ecommerce event chain
- duplicate tracking issues
- source and medium cleanliness
- ad platform alignment
- funnel visibility
- consent impact
- dashboard usefulness
- documentation and ownership
That order matters.
Do not start with pretty reports when the foundation is shaky.
Where to go next
If your dashboard looks fine but nobody trusts it, start here:
At the end of the day, the question is simple.
Do you trust the numbers enough to make decisions with them?
If the answer is “not really,” that is your answer.
Ready to fix what’s broken? Book the Tracking Sprint and get your data clean before it costs you another month of wrong decisions.
Ready to fix what's broken?
Stop guessing and get a real diagnosis. Book a Growth Audit or jump straight into a Sprint.
Find this useful? Make Ecomm Decoded a preferred Google source → so it shows up first when Google answers your ecommerce questions.