TL;DR
Two published numbers describe the same European banner and they differ by twenty points, because one divides by recorded choices and the other by banners shown [3]. Your pixels see the second one. That gap does not widen in November, but your traffic does, so the same percentage hides several times the revenue. Google will model around a denial once you clear 700 ad clicks per country over seven days [4]. Meta's Conversions API has no consent parameter at all [9]. Qualifying for Google's modelling takes days to register, which makes this a September job.
Key Takeaways
- Consent rate divides opt-ins by recorded choices; opt-in rate divides opt-ins by banners displayed. The difference is the visitors who ignored the banner entirely [3].
- A store quoting an 80% consent rate may be running browser pixels for closer to 56% of its European traffic, because a visitor who made no choice has no marketing consent as far as a pixel can tell [1][11].
- Google Ads conversion modelling requires correctly implemented consent mode or TCF v2.0 plus 700 ad clicks over 7 days, per country and domain grouping [4]. A four-market EU store can qualify in one market and not the other three.
- Meta's
data_processing_options: LDUapplies to US state privacy terms and names 14 specific states [8]. The server event has no consent field at all; the nearest thing isopt_out, which demotes the event to attribution only [9]. - Nobody has published a dated figure for how consent behaviour shifts during Black Friday specifically. Argue the season from volume and from qualification windows.
- On the Plus plan, seven server-side integrations in WeltPixel Conversion Tracking can be set to skip visitors who declined marketing consent, off by default and one destination at a time, while Google Ads passes the consent state your site reports through with the conversion instead.
Which number do your pixels actually obey?
The opt-in rate, and most stores quote the other one.
Didomi's State of Data Privacy benchmark, published January 2026 on data from across 2025 and hundreds of millions of European consent interactions, reports both metrics side by side. Consent rates by region run from 75.1% in Western Europe, with France at 71%, up to 89.3% in Eastern Europe. Opt-in rates for the same regions run from 55.7% to 67.6%. The no-choice rate sits between 21.7% and 27.4% [1]. Two numbers, one banner, twenty points apart.
The definitions explain it. Consent rate is opt-ins divided by total recorded choices. Opt-in rate is opt-ins divided by total notices displayed. No-choice rate is the banners shown where nobody interacted [3]. The consent rate quietly excludes everyone who scrolled past.
Your pixels have no such exclusion. A Shopify storefront exposes four consent purposes through the Customer Privacy API, and a pixel asks questions like marketingAllowed() before it fires [11]. That function returns a boolean. There is no third state for "has not decided yet" that a marketing pixel can treat as permission. So the visitor who ignored your banner is, to every browser pixel on the page, a visitor without marketing consent, and your real browser-side coverage sits nearer the opt-in rate.
One caution on borrowing a sector figure: the benchmark publishes 16 industry rows, Fashion and Jewelry at 78.0% consent and 55.4% opt-in, Home Equipment at 80.7% and 56.4% [2], and no "ecommerce" row. Read the closest category and say so.
Google gives you a vocabulary, and a 700-click floor
Google's side of this has documented rules, which makes it the side you can actually plan against.
Consent mode lets your banner tell Google's tags "denied" rather than staying silent, and Google will model some of what the denial cost you. The conditions are published: "You have correctly implemented consent mode or the IAB Transparency & Consent Framework (TCF v2.0)" and "You have a daily ad click threshold of 700 ad clicks over a 7 day period, per country and domain grouping". Both quality checks precede a model training phase [4].
Read that scoping clause twice, because it is where multi-market EU stores get surprised. The floor is scoped to a country and domain grouping. A store running Germany, France, Netherlands and Belgium can clear 700 clicks in Germany and miss it in the other three, so the modelled top-up lands in one of the four markets you are comparing against each other. The Belgian campaign then looks worse than the German one for a reason that has nothing to do with the campaign.
The second consequence is a calendar problem. Because it is a seven-day condition that gates a training phase, a consent mode fix made late in November cannot be modelled in time for the sale. There is no expedited path.
One configuration choice decides whether you get the good model at all. Basic consent mode holds Google tags back until the visitor interacts and qualifies only for a "General model (less detailed modeling)". Advanced mode loads tags with denied defaults, sends "consent state and measurements without cookies" on a denial, and is eligible for an "Advertiser-specific model (more detailed modeling)" [5].
The GA4 side of modelling has its own documented floors [6], and those belong to the guide on whether your consent banner is erasing your analytics, which walks the diagnostic properly. Context for this season: since June 15 2026 Google Signals "will only control the association of your Google Analytics sourced data with signed in user information for behavioral reporting", and consent mode is the single control for data [7]. The audit for that change is already written.
Meta has no consent parameter at all
This is the asymmetry people do not expect, and it changes what the EU decision even is. The Conversions API server event reference has no consent field, and the parameters that sound like one are not one. data_processing_options accepts a single value, LDU, and it applies to US state privacy terms, naming 14 states from California and Texas to Rhode Island. Meta's own page warns that when it is active, "businesses may notice an impact to campaign performance and effectiveness, and retargeting and measurement capabilities will be limited" [8]. It is a US mechanism, and it does nothing for a Dutch shopper.
The nearest server-side control is opt_out, described as "A flag that indicates we should not use this event for ads delivery optimization. If set to true, we only use the event for attribution" [9]. That is a demotion, not a consent signal. Browser-side there is fbq('consent','revoke') and fbq('consent','grant'), a client-side pause that has to be called on every page, and Meta's position on the legal question is plain: "Each company is responsible for ensuring their own compliance with the GDPR" [10].
So the per-channel reality for one declined European visitor: Google can be told "denied" and may model around the hole once your market clears its threshold, while Meta can be sent the event or not sent it. Nothing in between exists, so the Meta decision is yours and it is binary.
If the decline rate is stable, why does Black Friday change anything?
Because the percentage holds and the denominator does not.
Start with the honest limitation: there is no published, dated figure for how European consent behaviour shifts during Black Friday specifically. The CMP benchmarks are annual and regional [1][2], and Shopify's BFCM report measures volume rather than consent composition [12]. Anyone quoting you a seasonal consent percentage is guessing, and a guessed percentage survives into next year's slide deck.
The seasonal argument is arithmetic instead. Shopify's BFCM 2025 report puts platform sales at $14.6 billion, up 27% on 2024, peaking at $5.1 million per minute, with cross-border orders at 16% of all global orders [12]. Sale traffic skews toward first-time visitors, and a first-time visitor is the one meeting your banner for the first time. Whatever your decline rate is in September, November applies it to several times the traffic.
Run the arithmetic on an illustrative store to see the shape. A store taking 2,000 EU orders a month, if its buyers opt in at the same 58% its banner does, has roughly 840 orders whose browser events never fire. The same store taking 6,000 orders across BFCM week has about 2,520. Same banner, same percentage, three times the hole, in the week your bidding decisions are most expensive.
Three things to check before the freeze, in about twenty minutes:
- Open your CMP's analytics view and write down both metrics, consent rate and opt-in rate, for your European markets. If your dashboard reports only one, find out which denominator it uses [3].
- In Google Ads, check the last seven days of clicks per country against the 700-click floor [4]. Note which markets clear it and which do not.
- Confirm whether your consent mode setup is basic or advanced, since only advanced feeds advertiser-specific modelling [5].
Fix the load order too, if your tags are racing the banner: why your tags fire before Shopify answers the consent question covers what that race does to your numbers, and the consent mode v2 implementation guide covers the setup. Change it before the dates in the freeze calendar, because a consent change mid-sale is indistinguishable from a tracking failure afterwards.
The decision you make per channel, before the freeze
The browser side of this is not a decision: browser events respect Shopify's Customer Privacy API signals [11]. Server-side is where you choose. In WeltPixel Conversion Tracking on the Plus plan, each server-side integration can be set to skip visitors who declined marketing consent. It is off by default, it is per integration rather than global, and the app states the cost on the setting itself: reported conversions drop by roughly your marketing decline rate. Skipped orders stay visible per pixel as skipped, so what you lose is a number you can read.
Google Ads sits outside that switch by design, because the consent state your site reports is passed with the server-side conversion instead.
Which way to set it is a legal and commercial judgement, and the answer differs per destination. What matters operationally is making it in September, with the twenty-minute check in front of you, and writing down the date you changed it. A step in your conversion count on November 14 with no note against it is a week of somebody's time in December.
The per-integration consent skip and the per-pixel skipped-order counts live in WeltPixel Conversion Tracking [13], one setting per destination.
FAQ
Is my consent rate the same as the share of visitors my pixels can track?
No. Consent rate divides opt-ins by recorded choices, so it excludes the 21.7% to 27.4% who never interact with the banner [1][3]. Your browser pixels see something closer to the opt-in rate, because a visitor with no recorded choice has no marketing consent as far as the Customer Privacy API is concerned [11].
Will Google Ads model the conversions I lose to consent denials?
Only if you clear the published conditions: consent mode or TCF v2.0 correctly implemented, and 700 ad clicks over 7 days per country and domain grouping [4]. Multi-market stores often qualify in their largest market only.
Can I tell Meta that a visitor declined marketing consent?
Not through the Conversions API. There is no consent parameter on the server event, data_processing_options: LDU is a US-state mechanism [8], and opt_out only demotes the event to attribution [9]. The choice is to send the event or not send it.
Does consent behaviour get worse during Black Friday?
There is no published, dated figure showing that it does. What is documented is the volume: Shopify recorded $14.6 billion in BFCM 2025 sales at a peak of $5.1 million per minute [12]. A steady decline rate applied to several times the traffic hides several times the revenue.
If I turn on the server-side consent skip, what happens to my reported numbers?
They fall by roughly your marketing decline rate on the channels where you switched it on, which the app tells you before you save. Skipped orders stay visible as skipped in the per-pixel view, so the drop is measurable.
Sources
- Didomi, State of Data Privacy benchmark, "Average consent rate in Europe" (consent rates 75.1% Western Europe to 89.3% Eastern Europe, France 71%; opt-in rates 55.7% to 67.6%; no-choice rate 21.7% to 27.4%; January 2026, 2025 data), didomi.io/blog/benchmark-average-consent-rate-europe, accessed September 21, 2026
- Didomi, "Consent rate by industry in Europe 2026" (Fashion and Jewelry 78.0% / 55.4%, Home Equipment 80.7% / 56.4%; 16 sectors, no ecommerce row), didomi.io/blog/benchmark-consent-rate-by-industry-europe-2026, accessed September 21, 2026
- Didomi support documentation, CMP metric definitions (consent rate, opt-in rate, no-choice rate and their denominators), support.didomi.io, accessed September 21, 2026
- Google Ads Help, "About consent mode modeling" (consent mode or TCF v2.0 correctly implemented; 700 ad clicks over 7 days per country and domain grouping; quality checks precede training), support.google.com/google-ads/answer/10548233, accessed September 21, 2026
- Google for Developers, "Consent mode overview" (basic mode gets a General model; advanced mode sends consent state and measurements without cookies and is eligible for an Advertiser-specific model; updated July 30, 2026), developers.google.com/tag-platform/security/concepts/consent-mode, accessed September 21, 2026
- Google Analytics Help, "[GA4] Behavioral modeling for consent mode" (1,000 denied events per day for 7 days; 1,000 daily granted users for 7 of the previous 28), support.google.com/analytics/answer/11161109, accessed September 21, 2026
- Google Analytics Help, "Updates to Google Analytics Data Controls" (June 15, 2026 Google Signals change; Consent Mode becomes the single control), support.google.com/analytics/answer/17016975, accessed September 21, 2026
- Meta for Developers, "Data processing options" (
LDU, 14 named US states; performance and measurement impact when active), developers.facebook.com/docs/marketing-apis/data-processing-options, accessed September 21, 2026 - Meta for Developers, "Conversions API server event parameters" (
opt_out; no consent parameter exists), developers.facebook.com/docs/marketing-api/conversions-api/parameters/server-event, accessed September 21, 2026 - Meta for Developers, Meta Pixel GDPR implementation notes (compliance responsibility;
fbq('consent','revoke')per page), developers.facebook.com/documentation/meta-pixel/implementation/gdpr, accessed September 21, 2026 - Shopify developer documentation, "Customer Privacy API" (four purposes;
marketingAllowed();visitorConsentCollected), shopify.dev/docs/api/customer-privacy, accessed September 21, 2026 - Shopify, "BFCM 2025 data" ($14.6 billion, up 27%, peak $5.1 million per minute, cross-border 16%), December 2, 2025, shopify.com/news/bfcm-data-2025, accessed September 21, 2026
- WeltPixel Conversion Tracking, Shopify App Store listing, apps.shopify.com/weltpixel-conversion-tracking, accessed September 21, 2026