01
Conversions that fire more than once
A form thank-you page counted on every refresh, or a button tracked twice through two tag managers. Your conversion rate looks great and none of it is real.
How I fix itNumbers you trust, on one page you will actually open
GA4 and Search Console set up properly first, then a Looker Studio dashboard that answers the question you actually ask instead of the twelve you never did.
Most accounts I inherit are counting the wrong things
Not slightly wrong. Wrong in ways that change decisions, like a thank-you page counted as three conversions, or paid traffic filed under direct. A dashboard built on that just makes bad data prettier.
02
GA4 revenue that never matches the back office
The two numbers disagree, so people stop trusting both and go back to guessing. There are usually five reasons and two of them are worth fixing.
How I fix it
03
A report nobody reads
Forty two charts delivered monthly and opened never. If the report needs a call to explain it, the report is the problem.
How I fix it
1%
Gap between reported revenue and the back office after a GA4 rebuild
6
Numbers at the top of the report, which is all most months need
1
Page. If it does not fit on one page it does not get read
2 to 3
Weeks to configure, test and verify a full GA4 setup
Figures come from client accounts between 2023 and 2026. Tracking work is the one thing here with a definite finish line rather than an ongoing curve.
Nothing gets built on numbers we have not tested
Before a dashboard exists I go through every conversion you care about and trigger it myself, in the browser, watching what arrives. It is unglamorous and it is the only way to know.
Six numbers, a landing page table, and a paragraph
The layout I use everywhere. Six numbers at the top with last month beside them, the landing page table underneath, and a short written note explaining what changed and what I am doing about it.
How a reporting engagement runs
STEP 01
Audit
Every existing event, conversion and tag catalogued, with the ones that are wrong marked and explained.
STEP 02
Rebuild
GA4 reconfigured, tags cleaned up, and each conversion triggered by hand to confirm it lands correctly.
STEP 03
Build
The Looker Studio dashboard assembled on top of data we have already verified.
STEP 04
Hand over
A walkthrough of what each number means, so you are not dependent on me to read your own report.
I had stopped looking at our analytics because I did not believe any of it. Now there is one page I check on the first of the month and it matches what the bank says.
Dan Reyes
Owner, Northside Heating and CoolingReporting questions I get asked
Why does GA4 never match my Shopify dashboard?
Different definitions, mostly. GA4 attributes to the session that led to the order, Shopify counts the order. Add ad blockers, consent banners and refunds and a gap of a few percent is normal. A gap of thirty percent is a bug.
Do I need to move off Universal Analytics data?
If you want year on year comparisons, yes, and it needs doing before the historical data is gone. I can export it to BigQuery so the old numbers stay queryable next to the new ones.
Can you work with the tracking my last agency set up?
I audit it first. Sometimes it is fine and needs three corrections. More often there are two tag managers fighting each other, and starting clean is faster than untangling it.
Do I need BigQuery?
Only if you have more data than the GA4 interface samples comfortably, or you want history beyond the retention window. For most businesses under a few hundred thousand sessions a month, no.
Not sure your numbers are real?
Give me read access to GA4 for a day. I will record a 10 minute walkthrough of what is being counted wrong and what it is costing you in decisions.