Eight websites · checked in a real browser
Almost nobody is measuring anything.
Over two weeks in September 2026 I ran the same audit across eight small-business websites — loaded every page in a real rendered browser and checked what was actually installed. Half of them had no way whatsoever to tell where a customer came from. Here is exactly what was missing, with the real counts and the real denominators.
had no Meta pixel. Any money these seven spend on Facebook or Instagram cannot see what happens after the click — no retargeting, no lookalikes, no idea which ad produced the customer.
had no measurement at all — no Meta pixel and no Google Analytics. Not underperforming. Genuinely unmeasured.
What was missing
Each bar is the share of sites where the check came back as a gap. Denominators differ because the audit gained checks as it was built — the exact count is printed on every row, and nothing is averaged across different questions.
Two of them cannot run the ad they want
Meta will not run lead ads for a business without a reachable privacy policy. It is a hard requirement, checked at the point the campaign goes live.
So two of these businesses could build the campaign, write the copy, fund the budget, and be refused — for the absence of a page that takes an afternoon to publish. Neither of them knows that yet.
Why this keeps happening
None of these are unusual businesses, and none of them are badly run. The pattern is simpler than that.
Tracking is invisible. A missing pixel breaks nothing — the site still loads, the form still sends, the phone still rings. There is no error message for you have no idea which half of your money worked, so it never reaches the top of anybody's list.
It only becomes visible at the moment somebody starts spending. Then the absence is retroactive: the first ninety days of data that would have taught you something were never collected, and cannot be collected later.
How this was checked
- Eight websites, audited between 16 and 28 September 2026. Two to eight pages loaded per site.
- Every page was loaded in a fully rendered browser, not read as raw source. A pixel injected by JavaScript is invisible to a source-code check, and reporting it missing would be wrong.
- Public sources only. No logins, no account access, nothing a customer could not see themselves.
- A gap means not detected on the pages scanned. It is not a claim about pages that were not checked.
- This is not a random sample. Eight is a small number, and these are businesses I approached or was asked to look at — not a survey. Read it as a field note, not a statistic.
- Some checks are narrower than their label. The email check detects Klaviyo specifically, not every email platform — it is reported that way above rather than dressed up as something broader.
- Businesses are unnamed and every finding here is free to fix without hiring anyone.
Sectors covered — coffee & events catering · home services · packaging & design · ranch & venue · residential building · electrical supply distribution ×2
Want to know which of these is true of you?
There is a free live check on this site that runs the visibility half of this audit against your own domain in about fifteen seconds. No login, no account access, real screenshot included.
Run the free checkOr book twenty minutes and I’ll walk you through the full review — free, whether or not we work together