We do not publish client names or results without written permission. Instead, here is a full sample audit report for a demo store, previews of the documents you receive, and how we fix the problems we see most.
Sample audit report mockup for Northwind Outdoor Co. (demo store), a demo store with invented data. Signal-quality score 38 out of 100, needs work. 4 high, 3 medium, 1 low severity findings. Top findings: F1, Meta Purchase sent by both the store app and a GTM tag (High); F2, No shared event_id between Meta pixel and Conversions API (High); F3, Consent Mode defaults to granted before the visitor chooses (High). Event parity for the sample week: Store orders 212, GA4 purchases 236 (+11%), Meta purchase events 389 (+83%), Google Ads conversions 401 (+89%).
audit-report · northwindSample
Tracking audit report
Northwind Outdoor Co.
Shopify · Meta and Google Ads · demo store
38/100
Signal quality
✗ Needs work
3 of 5 areas broken
4 High
3 Medium
1 Low
Top findings
HighF1 Meta Purchase sent by both the store app and a GTM tag2h
HighF2 No shared event_id between Meta pixel and Conversions API3h
HighF3 Consent Mode defaults to granted before the visitor chooses4h
Event parity purchases vs store, sample week
Store212base
GA4236+11%
Meta389+83%
Google Ads401+89%
✓ Target every platform within 5% of the store
Sample audit report
A full audit report, written for a store we made up.
Score, method, eight ranked findings with evidence and fixes, an event-parity check across the store, GA4, Meta and Google Ads, consent test results and a phased fix plan. The same format a paying client receives.
Sample dashboard mockup with invented numbers for a demo store, comparing what platforms report with the store. Meta purchases vs store orders: audit week 389 vs 212 (+83%, broken); after the fixes 203 vs 198 (+3%, clean). Google Ads conversions vs store orders: audit week 401 vs 212 (+89%, broken); after the fixes 194 vs 198 (−2%, clean). GA4 revenue vs store revenue: audit week $36,210 vs $31,480 (+15%, broken); after the fixes $29,870 vs $29,410 (+2%, clean). Within 5% of the store counts as clean.
tracking-health · northwindWeeklySample
Tracking health
Platforms vs the store
PlatformStore
Store orders
212
audit week
Audit week
+89%
✗ worst gap
After the fixes
3%
✓ worst gap
Meta purchases vs store orders
Audit week✗ +83%
389
212
After the fixes✓ +3%
203
198
Google Ads conversions vs store orders
Audit week✗ +89%
401
212
After the fixes✓ −2%
194
198
GA4 revenue vs store revenue
Audit week✗ +15%
$36,210
$31,480
After the fixes✓ +2%
$29,870
$29,410
Clean = within 5% of the store. Invented numbers.
From the sample report: purchases each platform received against the store, in the audit week and after the fixes.
Sample deliverables
What you keep after an engagement
Short excerpts of the documents we hand over. Each one is yours to keep and share with any developer.
Tracking plan
Every event, when it fires, what it carries and where it goes.
tracking-plan · v1.23 of 14 eventsSample
Tracking plan excerpt (sample)
Event
Fires when
Key parameters
Sent to
add_to_cart
Item added
value, items
GA4, Meta
begin_checkout
Checkout opens
value, items
GA4, Meta
purchase
Order completed
transaction_id, event_id
GA4, Meta, Google Ads
Excerpt with invented values.
QA log
Each test, what should happen, what did, and the result.
qa-log · round 12 pass · 1 retestSample
QA log excerpt (sample)
Test
Check
Result
QA-04
One Meta Purchase per test order
✓Pass
QA-05
Same event_id on pixel and server
✓Pass
QA-06
No purchase on thank-you reload
✗Fail, retest
Excerpt with invented values.
Consent test matrix
Which tags run for each consent choice, tested in fresh browsers.
consent-matrix · fresh browsersSample
Consent test matrix excerpt (sample)
Tag
Refused
Accepted
Ignored
GA4
Cookieless ping
Full
Cookieless ping
Google Ads
Cookieless ping
Full
Cookieless ping
Meta pixel
Blocked
Full
Blocked
Excerpt with invented values.
Monthly report outline
For Growth Retainer clients. Revenue first, clicks last.
monthly-report · contentsSample
Revenue and orders from the store, against ad spend
Tracking health: platform counts vs store orders
What we changed in campaigns and why
Creative tests: what won, what is next
Next month: what to scale, fix or stop
Outline only.
Typical patterns we fix
Case notes, without the client
These describe how common problems happen and how the fix works. They are patterns, not stories about a particular store, so there are no names or numbers.
A store where purchases counted twice: how it happens and how the fix works
What you see
Meta or Google Ads reports far more sales than the store took, and the reported return on ad spend looks better than the bank balance.
How it happens
Two things send the same purchase. Usually a store app was installed after someone had already added a pixel tag in Tag Manager, or the browser pixel and the Conversions API both send the sale without a shared event ID, so the platform cannot tell they are the same order.
How the fix works
We pick one browser source, remove the other, and give every order one event ID that both the browser and the server send. Then we place test orders and confirm each one appears once in every platform.
A banner that asks for consent but does not wait for the answer
What you see
The cookie banner looks right, yet analytics and ad cookies are already set before anyone clicks.
How it happens
The banner and the tags are configured separately. Consent Mode defaults were never set to denied, or the tags load on a page-view trigger that runs before the banner has a chance to answer.
How the fix works
We set denied defaults in a tag that runs before everything else, tie each tag to the consent it needs, and test refused, accepted and ignored visits in fresh browsers. Wording goes to your legal adviser.
Sales credited to the payment page instead of the ad
What you see
A payment provider shows up as one of the top sources of revenue in GA4, while paid campaigns look weaker than they should.
How it happens
Buyers leave the store to pay and come back to the thank-you page. GA4 treats the return as a new visit from the payment site and gives it the credit.
How the fix works
We add the payment domains to the unwanted referrals list, check the checkout flow keeps the original session, and confirm with a test order that the campaign keeps the sale.