Google Consent Mode v2 on a landing page: what changed for your Google Ads campaigns

Since March 2024, Google has required Consent Mode v2 on any site that runs Google Ads campaigns or measures conversions with traffic from the European Economic Area (EEA) and the UK. Without it, Google stops building remarketing audiences from your European visitors, and part of your Google Ads or Analytics conversions simply vanishes from the dashboard — not because visitors didn't convert, but because the signal proving it is no longer sent in a format Google accepts. Many landing page owners still aren't aware of this, or wrongly assume a GDPR-compliant cookie banner is enough. This guide covers what changed, what to actually install on a landing page, and what research shows about the trade-off between privacy and ad measurement.

What Consent Mode v2 is, and why it became mandatory

Consent Mode is a protocol added to gtag.js (the script powering Google Analytics and Google Ads tags) that tells Google, tag by tag, whether the visitor allowed or refused each data category — analytics and personalized advertising. Version 2, released in late 2023 and made mandatory in March 2024, is a direct response to the Digital Markets Act (DMA): Google, designated a "gatekeeper" for advertising, now has to prove it only combines advertising data with explicit, documented consent. In practice, if your landing page targets European traffic and uses Google Ads (conversions, remarketing) or Google Analytics linked to Google Ads, the absence of Consent Mode v2 blocks remarketing audience-building on EEA visitors and degrades conversion tracking — whether or not your cookie banner is otherwise GDPR-compliant.

Basic mode or advanced mode: the difference that matters

Consent Mode v2 can be configured in two modes, and the choice changes what happens even before the visitor answers the banner.

• Basic mode — Google tags (GA4, Google Ads conversion) don't load at all until the visitor has given consent. Simple to set up, but no data exists at all for visitors who decline or ignore the banner: no measurement, no basis for statistical estimation.
• Advanced mode — tags load regardless, but send an anonymized "cookieless ping" even without consent. Google then aggregates these signals through conversion modeling: a statistical model that estimates, from the observed behavior of consenting visitors, the likely conversion volume among non-consenting visitors. This modeling only activates above a certain volume (Google cites a threshold around 700 ad clicks over 7 days per country and domain) — below that, advanced mode adds nothing over basic mode.

For a landing page with a sizable ad budget (Google Ads, Performance Max), advanced mode generally recovers a real share of otherwise invisible conversions; for a small, low-traffic page, basic mode is enough and avoids needlessly complex setup. The trade-off resembles the one covered in our article on Google Ads Quality Score: past a certain volume, advanced optimizations pay off; below it, they mostly just complicate tracking.

The two new parameters: ad_user_data and ad_personalization

Consent Mode v1 only distinguished analytics_storage and ad_storage (permission to set cookies). V2 adds two signals independent of cookies themselves:

• ad_user_data — allows (or not) sending personal data to Google for advertising purposes, even without a cookie (IP address, device identifiers).
• ad_personalization — allows (or not) using that data for remarketing or similar audiences.

Both parameters must default to denied before any Google tag loads, then get updated to granted or left at denied based on the actual choice made on the consent banner — never the other way round. This ordering (default denied, then update) is exactly what most sloppy implementations get backwards, letting tags fire before the visitor has had a chance to answer the banner.

Implementing it on a Next.js landing page

On a template shipped as source code, like LanderKit templates, implementation comes down to three steps, with no need for a paid CMP if your homemade banner already meets the GDPR requirements covered in our GDPR form compliance guide:

1. Declare the default consent state — a gtag('consent', 'default', { ad_storage: 'denied', ad_user_data: 'denied', ad_personalization: 'denied', analytics_storage: 'denied' }) call placed before any GA4 or Google Ads script in the <head>, before the cookie banner even hydrates.
2. Update it when the visitor decides — on clicking "Accept" or "Decline" in the banner, call gtag('consent', 'update', {...}) with the matching values; that's the same event that should drive the banner itself, covered in our article on Google Tag Manager on a landing page if you're using a GTM container instead of gtag.js directly.
3. Turn on advanced mode if your volume justifies it — in GTM, check "Consent mode advanced" on the Google Ads and GA4 tags; with a direct implementation, this behavior is automatic as soon as ad_storage stays at denied while the tag keeps loading.

Deploying on Vercel doesn't change any of this — see our Next.js deployment guide — but make sure the gtag('consent', 'default', ...) call runs client-side before any tracking component, which rules out placing it only inside a component that loads after hydration.

What research says about the privacy/measurement trade-off

Consent Mode addresses a real tension, documented by applied economics research on GDPR as a whole rather than on this specific protocol (too recent to have its own literature yet). A study by Aridor, Che and Salz, published in 2023 in The RAND Journal of Economics, measured GDPR's effect on a major online-travel ad intermediary: the number of trackable cookies fell by roughly 12.5%, but the prediction algorithm's ability to estimate the remaining visitors' behavior didn't significantly worsen — the average value of still-identifiable visitors even rose, offsetting much of the loss (see the study on Google Scholar). That's precisely the mechanism Consent Mode v2's conversion modeling exploits: statistically estimating what's missing rather than measuring it directly.

Conversely, a study by Goldberg, Johnson and Shriver, published in 2024 in the American Economic Journal: Economic Policy, measured a 12% drop in page views and recorded revenue across 1,084 sites in Europe after GDPR enforcement began, with an effect twice as large on small e-commerce sites as on large ones (see the study on Google Scholar). In other words: compliance carries a real measurement cost, and it falls hardest on small operations that have neither the traffic volume needed for reliable modeling nor the technical team for a careful implementation — one more reason to start from clean source code rather than piling scripts on over successive campaigns.

Common mistakes to avoid

• Loading GA4 or the Google Ads conversion tag before consent default — even at denied, script order must put the consent declaration first.
• Confusing a GDPR-compliant banner with Consent Mode v2 — a banner can be flawless from a regulatory standpoint while never sending the gtag('consent', 'update', ...) signal Google expects: the two are related but distinct.
• Turning on advanced mode without the necessary traffic — below the modeling threshold, it adds nothing and just complicates the tracking audit for no gain.
• Forgetting ad_user_data and ad_personalization — an implementation that only handles the old ad_storage parameter (Consent Mode v1) still fails Google Ads' 2024 requirements.

LanderKit templates ship as Next.js source code, with no pre-installed tracking sprawl — which makes exactly this kind of clean implementation easier, controlled from the start rather than bolted on afterward. On a page with a sizable Google Ads budget like the single-product e-commerce template, or a lead-generation page like the real estate template, a poorly wired Consent Mode v2 shows up directly as a cost per lead that climbs for no apparent reason. The 10 LanderKit templates (€89 each, €229 for the full bundle) start from a base clean enough that this wiring takes a few lines of code, not a project.