Skip to content

Your conversions dropped.
Your sales didn’t.

If your Google or Meta reporting fell off a cliff after you added a consent banner, the orders are almost certainly still arriving. What stopped is the measurement. Those are different problems, they have different fixes, and only one of them is an emergency.

Reads your live store in under a minute and tells you which of the causes below applies to you.

Does this match what you are seeing?

  • Purchase events stopped arriving in GA4, or arrive for a fraction of real orders
  • Meta reports far fewer conversions than your store’s own order count
  • Google Ads shows a Consent Mode warning on your conversion actions
  • It started right after you added a cookie banner
  • Your reported ROAS collapsed while actual revenue did not

Why it happens

Three separate mechanisms, usually stacked. A scan tells you which ones are live on your store.

A banner doing its job stops browser-side conversions

When a visitor declines, the tags that report the sale are supposed to stop firing. That is the banner working, not breaking. But nobody tells your ad platforms the difference between "this sale did not happen" and "this sale happened and you may not hear about it" — so both look like zero.

Without Consent Mode v2, Google cannot model the gap

Consent Mode v2 is how you tell Google that a conversion happened without consent to attribute it. With it, Google models the missing journeys. Without it, there is nothing to model from — the gap is simply absent, and your bidding optimises against a picture with holes in it.

And consented conversions still get lost in transit

Ad blockers, ITP and browser restrictions drop a share of tags that had every right to fire. Sending the event from your server instead of the browser is how that share stops disappearing — and it is the part most stores have not done.

It looks the same on every platform

A pixel added through your store’s own marketing settings loads on its own schedule and does not know a consent banner exists. The scan identifies which platform you are on and reads it accordingly.

Salla

Pixels connected through Salla’s own marketing integrations fire on their own schedule. Adding a banner to the theme afterwards does not reach them.

Zid

Same shape: the store settings inject the tag, so the tag predates anything you add to the storefront later.

Shopify

Apps and channel integrations load their own scripts, and checkout is a separate surface from the rest of the store.

WooCommerce

Marketing plugins write tags straight into the theme, usually above whatever consent plugin was installed later.

There is no backfill

This is the part worth acting on. A conversion that was not measured today cannot be measured later — no platform reconstructs a signal it never received. Whatever you install next month starts working next month, and the weeks in between stay missing permanently.

It also compounds in the wrong direction: your bidding is being optimised right now against a picture with holes in it, so the budget keeps moving toward whatever still happens to report.

Source: Google — about Consent Mode and conversion modelling

What fixing it looks like

Start with the scan either way — it is free and it tells you which of the three mechanisms is actually costing you. What you do next depends on whether you would rather do the work yourself.

Do it yourself

Pro adds server-side forwarding to ten platforms, the scanner and the audit log. $29.99 a month.

Look at Pro

Have us do it

We install it, wire Consent Mode v2, test your pixels in every consent state and re-scan to prove it. $299 to install and $99.99 a month, per domain.

Look at Managed

Find out what is actually broken

Scan my store — free, no signup