State privacy laws produce a pattern we see repeatedly in US accounts: reported conversions fall by a third over a fortnight. Spend is flat, creative is unchanged, rankings are unchanged, and the business says revenue has not actually moved. Everybody starts blaming the platform.
It is almost always measurement, and the usual culprit is consent.
What state privacy laws changed in the US
California gave people the right to opt out of the sale or sharing of personal information, and requires businesses to honour the Global Privacy Control — a signal the browser sends automatically, with no banner interaction at all. A growing list of other states has followed with their own laws.
The practical effect for advertisers is simple: a meaningful share of your US visitors now arrive with tracking restricted before they have clicked anything.
How that shows up in your dashboards
- Fewer conversions reported, while the business sees the same number of orders.
- Audience sizes shrinking as fewer people are added to retargeting pools.
- Attribution shifting toward direct and organic, because paid channels can no longer claim what they used to.
- Bidding getting worse, which is the expensive part — smart bidding optimises on the signal it receives, and a thinner signal produces worse decisions.
How to tell if this is your problem
Three checks, in order:
- Compare platform conversions against your own back end over the same window. If the business recorded the same revenue and the platform recorded less, it is measurement, not performance.
- Ask who touched the consent banner. Legal and web teams update these without considering the analytics consequences, because from their side nothing broke.
- Check event match quality in Meta’s events manager and the diagnostics in Google Ads. A sharp fall there dates the change for you.
What actually helps
Configure consent mode properly, not just install it. Google’s consent mode and Meta’s consent signal both need to receive the real state from your consent platform. An installed-but-unwired setup is common and does nothing.
Server-side events, with real matching. A Conversions API recovers events the browser loses — but only where consent allows, and only if enough customer information is attached for the platform to match the person. A CAPI sending events with nothing to match on is doing very little.
Deduplicate. Browser and server events must share an event ID or you count the same purchase twice, which produces the opposite problem: a flattering ROAS and a scaling decision built on a fiction.
Design the banner honestly and test it. A banner that makes rejecting hard may be a compliance exposure. One that makes rejecting the default costs you data you were entitled to. Get it right once with your own counsel rather than tuning it quarterly.
Measure the business, not the platform. If revenue is flat and reported conversions fell, nothing is wrong with your advertising.
The structural answer
Accept that platform-reported conversions will keep drifting downward and stop treating them as truth. Hold a blended view at business level — total revenue against total ad spend — which cannot be affected by any of this. Our blended ROAS and MER calculator does that in one screen, and it is the number worth putting in front of a board.
Use platform numbers for decisions inside a channel, where the bias is at least consistent. Use blended numbers for decisions about money.
And when a consent banner changes, annotate it in your reporting. Six months later, nobody remembers — and that annotation saves a fortnight of investigating the wrong thing.
Questions people ask
Why did my US conversions fall with nothing changed?
Something did change, just not in the ad account. A privacy banner, an opt-out signal being honoured, or a tag now waiting for consent will all cut reported conversions while actual revenue stays flat.
What is Global Privacy Control?
A browser signal that tells a site the visitor opts out of the sale or sharing of their data. Several US state laws require it to be honoured, and browsers and extensions send it by default for a growing share of visitors.
How do I tell if this is my problem?
Compare your own order or CRM records with platform-reported conversions over the same weeks. Flat revenue against falling reported conversions is a measurement problem; if both fell, it is a performance problem.
What helps?
Server-side tagging with good match data, consent mode wired so modelling can operate, and moving your judgement to blended figures — total revenue against total spend — rather than platform-reported return on ad spend.
Working with US brands: Performance marketing for US brands