Measure conversions when most visitors reject tracking
In a consent-gated market a large share of visitors opt out before a single analytics event fires. I measure the rest with server-side truth and consented journeys, model the opted-out share honestly, and label it as modeled. No fabricated numbers.
Top Rated · 100% Job Success · 5.0 ★ · 22 Upwork contracts
Quick answer
Yes, you can measure conversions and revenue when most visitors reject tracking. You count the sale server-side from the commerce backend (a fact regardless of marketing consent), measure consented journeys with PostHog, and model the opted-out share transparently, always labeled as modeled. It works across multiple brand sites on one stack.
Consent gating quietly breaks your analytics
Standard analytics assumes the tag runs on page load. Under a consent manager, it stays dormant until the visitor opts in, and in regulated categories a lot of people never do. So your conversion count drops to whatever slice consented, and that slice is biased, because the people who accept tracking are not a random sample of buyers. You end up making decisions on a skewed sample and calling it the truth.
A measurement layer that survives consent
Server-side truth for the sale
The order is a fact your backend already knows. I capture conversions and revenue server-side from the commerce backend, so the number is complete even when the browser tag was blocked.
Consented journeys, measured fully
For visitors who opt in, the full funnel is measured with PostHog, gated on consent so nothing captures before a yes. That is your defensible, first-party layer.
Model the gap, and label it
The opted-out share is estimated by scaling consented data against the known consent rate, and it is always presented as a modeled figure, never as measured. Honest beats confident and wrong.
Multiple brands, one picture
Several brand sites on one stack roll into a single view: revenue per brand with consent gating in place, instead of separate accounts nobody reconciles.
Reporting you can defend
One report with three clearly separated layers: measured revenue, measured consented journeys, and the labeled modeled total. A number you can stand behind on a call with a regulator or a CFO.
It stays working
Tracking breaks silently. I add monitoring so a broken feed or a dropped consent-mode update alerts you, instead of quietly skewing every decision for weeks.
From a skewed sample to a number you can defend
- 1
Audit what you can actually see
I measure your real consent rate and how much of your conversions and revenue your current setup is missing. In a consent-gated market that gap is usually far bigger than teams expect.
- 2
Move the source of truth to the server
Conversions and revenue get captured server-side from the commerce backend, not a browser pixel, so the total is complete regardless of marketing consent. This is the single biggest fix.
- 3
Set up consent-gated capture
PostHog is wired to fire only after opt-in, EU-hosted or self-hosted where it matters. Nothing is captured before consent, and the data stays on infrastructure you control.
- 4
Model the opted-out share honestly
The gap is estimated from the consented data and the consent rate, and it ships clearly labeled as modeled. You get a realistic total without pretending it was measured.
- 5
Report and monitor
You get one report that separates measured from modeled, plus alerts so a silent break gets caught before it costs you a month of decisions.
Measurement that holds up under a hard look
- I run consent-gated measurement across multiple brand sites, where opt-in is a minority of traffic.
- Server-side truth plus honest modeling: I label what is measured and what is estimated, and never dress one up as the other.
- Top Rated on Upwork, 100% Job Success and 5.0 across 22 contracts.
- One engineer who owns the pipeline, the analytics, and the reporting, so the measurement actually connects end to end.
Common questions
Can you measure conversions if most visitors reject cookies?
Yes, but not the way most setups try to. You measure consented users fully with server-side events, count the sale from the commerce backend where it is a fact regardless of marketing consent, and model the opted-out gap transparently. You never present a modeled number as if it were measured.
Why does GA4 undercount so badly under consent gating?
GA4 and browser pixels fire only after a visitor opts in. In regulated categories the opt-in rate is often well under half, so the dashboard shows a fraction of reality, and the missing part is not random. Every ratio built on it inherits that bias.
Is server-side tracking a way around consent?
No, and it should not be treated as one. Server-side tracking reliably captures what you are lawfully allowed to process, mainly the sale itself as a first-party business record. Consent still governs marketing and analytics identifiers. Confirm the lawful basis with your DPO.
Why PostHog instead of GA4 for a consent-gated site?
You control when capture starts, you can keep data in the EU or self-host it, and you own the event stream. GA4 is built to fire early and send data to Google by default, which is the opposite of what a consent-gated setup needs.
Can you roll this out across several brand sites?
Yes. Each server-side event is tagged by brand, so several sites on one stack become one query: revenue per brand with consent gating in place, instead of separate analytics accounts that never reconcile.
What do I actually get at the end?
A number you can defend: complete measured revenue from the server, a clearly labeled modeled total for the opted-out share, and a consent rate tracked over time. Plus monitoring so a silent break gets caught early.