GDPR for Marketers: A Practical Checklist

11 min read

Most marketers believe GDPR compliance is a legal problem solved by a legal tool. Get consent, store it, and you're done. But it’s actually a technical data problem. The moment a user clicks "Reject," a series of technical events is supposed to happen. In most setups, it doesn't. Or at least, not correctly.

SS

Simul Sarker

Founder & Product Designer of DataCops

Last Updated

May 17, 2026

Most marketers I talk to believe that when a visitor clicks "Reject All," their analytics go dark for that person. Zero data. A blank row. That belief is wrong, it is expensive, and it has been wrong since GDPR took effect in 2018.

I have spent years building analytics for companies that sell into the EU, and I will be blunt: the single most common GDPR mistake I see is not a compliance violation. It is the opposite. It is marketers voluntarily destroying data they were always legally allowed to keep, because someone told them "Reject All means no tracking" and they never checked.

This is not a fear post. It is not a "fines are coming, panic now" post. There are enough of those, and they are mostly written by people selling consent banners.

This is a triage post. GDPR does not delete your data. It sorts it into two piles: data you can use freely, and data you need permission for. Most marketers throw away the whole first pile because they never learned it existed.

The reason that mistake happens is architectural. When your consent banner says no, the third-party scripts that do your tracking simply do not fire. All-or-nothing.

But GDPR is not all-or-nothing, and your data collection should not be either. The fix is to collect anonymous, non-identifying analytics unconditionally - legally, with no consent needed - and gate only the identifying stuff behind consent. Two tiers, separated at the source. That is what DataCops is built to do, and it is the difference between a half-blank dashboard and a working one.

See the first-party consent manager platform, our GDPR and first-party data deep-dive, and the Cookiebot alternative.

Quick stuff people keep asking

What does GDPR mean for digital marketers? It means personal data - anything that can identify a person - needs a lawful basis before you process it. For most marketing, that lawful basis is consent. It does not mean you cannot measure anything.

Anonymous, aggregated measurement was never personal data and never needed consent.

Can you use Google Analytics without cookie consent under GDPR? Not in its normal, identifying configuration - that sets cookies and processes personal data, so it needs consent. But you can run analytics without consent if it collects no personal identifier: no cookie, no user ID, no fingerprint, just an anonymous, aggregated count. The tool is not the issue.

What it collects is.

What is a lawful basis for marketing under GDPR? There are six lawful bases. Two matter to marketers: consent, and legitimate interest. Consent covers cookies, pixels, and personalized advertising.

Legitimate interest can cover some anonymous analytics and security work, but it is not a free pass - you have to document it and it cannot override the person's rights.

Do marketers need a CMP for GDPR compliance? If you process personal data for marketing in the EU, you need a way to collect and record consent, and a CMP is the standard way. But a CMP is a tool, not compliance itself. And here is the catch most CMP vendors will not put in their brochure: the CMP is a third-party script, and it gets blocked too.

What is Google Consent Mode v2 and do I need it? It is Google's framework for adjusting tag behavior based on consent state. If you use Google Ads or GA4 with EU traffic and want conversion modeling and remarketing to keep functioning, yes, you need it. When consent is denied, it sends cookieless pings - anonymous signals - instead of nothing.

That is the legal-data principle, built into Google's own product.

What data can you collect without consent under GDPR? Anonymous, aggregated data that cannot be tied back to an individual. Total pageviews. Sessions per landing page.

Conversion rate as a percentage. Traffic by country at the country level. Bounce rate.

None of that identifies anyone, so none of it needs consent.

Does GDPR apply to non-EU companies targeting EU users? Yes. GDPR follows the data subject, not the company. A US-based store selling to someone in Germany is processing an EU resident's personal data and is on the hook for GDPR.

Location of your servers or your HQ does not exempt you.

What happens if your cookie banner doesn't comply with GDPR? Regulators have fined non-compliant banners - pre-ticked boxes, no real reject option, consent walls. But the quieter cost is data integrity. A banner that nags or that gets blocked produces a consent record you cannot trust, and analytics built on untrustworthy consent state is worse than useless.

The gap: "Reject All" was never a blackout

Here is the layer almost every GDPR checklist misses.

When a visitor clicks "Reject All," GDPR is telling you one specific thing: do not process this person's personal data without a lawful basis. It is not telling you to stop counting. It is not telling you the visit did not happen.

It is telling you that you may not identify, profile, or cross-site track that individual.

You can still record that a visit occurred. That a session landed on a particular page. That someone in a particular country viewed a product.

That a checkout was started. None of those facts, collected without an identifier, are personal data. They are anonymous events.

GDPR has no objection to them. It never did.

So picture the typical setup. Visitor rejects. The consent banner, doing its job, blocks every tagged script - GA4, the Meta pixel, everything.

The dashboard records nothing for that person. The marketer sees their numbers drop 30, 40, sometimes 50% after the banner went live and concludes "that's the cost of compliance." It is not the cost of compliance. It is the cost of an all-or-nothing architecture.

Compliance only required dropping the identifier. The architecture dropped the entire visit.

That is Layer 2 of how analytics quietly breaks in 2026, and it is the most self-inflicted of all of them. Marketers are not losing this data to a law. They are losing it to a script-blocking switch that conflates "no personal data" with "no data."

Now layer the next problem on top, because it is the one CMP vendors really do not advertise. The consent banner is itself a third-party script. uBlock Origin, Brave's built-in shields, and other privacy tools block consent management scripts - somewhere in the range of 30 to 40% of privacy-conscious traffic. Think about what that means.

For a chunk of your visitors, the banner never even loads. No banner, no consent prompt, no recorded choice. Your tags then either fire with no consent - a violation - or do not fire at all.

Either way, the CMP you bought to make consent reliable produced an unreliable, partly-empty consent record. And on single-page-app sites, the banner and the analytics tags race each other on route transitions, so even visitors who would have consented get measured inconsistently.

So the honest situation is: you are losing the rejected visitors to all-or-nothing blocking, and losing a slice of everyone else to the CMP script itself being blocked. The dashboard you are making decisions from is missing both groups and you cannot see the hole.

The root cause is the same one behind nearly every analytics-integrity failure. Your measurement depends on third-party scripts that fire from the browser, where ad blockers, privacy shields, and consent races all get a vote. There is no isolation.

There is no two-tier separation. It is all one pipeline, and the consent banner is a crude on-off valve in front of it.

The practical GDPR checklist for marketers

This is the actual checklist. It is organized around the two piles: what is always legal, and what needs consent.

Pile one - collect this unconditionally, no consent needed:

  • Anonymous, aggregated analytics: pageviews, sessions, landing pages, conversion rate as a percentage, country-level geography, bounce rate. No identifier, no cookie, no consent.
  • Cookieless pings via Consent Mode v2 when consent is denied - so you keep modeled conversions and aggregate trends.
  • Server-side aggregation of events, stripped of personal identifiers before storage.
  • Security and fraud signals at an aggregate level, where you can document legitimate interest.

Pile two - needs consent before you collect or process:

  • Any cookie or identifier used for analytics that ties activity to an individual.
  • The Meta pixel, Google Ads tags, and any advertising or remarketing tag - these build profiles, so consent first.
  • Cross-site tracking and audience building.
  • Email marketing to individuals: GDPR-grade consent, granular, freely given, with easy withdrawal. No pre-ticked boxes. No bundling consent into a terms-of-service agreement.

Process checklist:

  • Run a real banner with a genuine, equally prominent "Reject All" - not a buried link, not a pre-ticked box, not a wall.
  • Deploy Consent Mode v2 if you touch Google products with EU traffic. It is effectively required for conversion modeling and remarketing to function legally.
  • Keep a consent record: who consented, to what, when. If you cannot produce it, you cannot prove it.
  • Make consent withdrawal as easy as giving it.
  • Document your legitimate-interest basis for anything you run under it. An undocumented legitimate-interest claim is not a defense.
  • Keep a data processing register and know which third parties touch EU personal data.

The architectural item the other checklists leave off:

  • Separate your collection into two tiers at the source. Anonymous analytics flows unconditionally and legally. Identifiable, consent-required data is gated behind actual consent. They should not share one on-off switch. When they do, every rejection blanks data you were entitled to keep.

That last point is the whole reframe. Compliance is not "stop tracking." Compliance is "sort your data correctly, then collect each pile under its correct rules." DataCops is built around exactly that split - anonymous flows always, identifiable flows on consent - running first-party on your own subdomain, which also makes it far more resilient to the ad blockers and privacy shields that hollow out browser-based setups.

Decision guide

EU traffic, currently see a 30 to 50% data drop after your banner went live? You are blanking legal data. Move anonymous analytics out from behind the consent switch. That is your highest-value fix.

Use Google Ads or GA4 with EU visitors? Consent Mode v2 is not optional anymore - set it up so denied-consent visitors still produce cookieless aggregate signal.

Shopify store selling into the EU? GDPR applies regardless of where you are based. Audit your pixel and tags - those need consent. Audit your basic traffic analytics - those can run anonymous.

Building an email list? Granular, explicit, unbundled consent at signup, with one-click withdrawal. No pre-ticked boxes, ever.

Choosing a CMP this quarter? Fine, but go in knowing the CMP script gets blocked for 30 to 40% of privacy-tool users. A CMP records consent. It does not, by itself, give you a complete or trustworthy dataset.

Worried mostly about fines? The blunt fines come from bad banners and missing consent records. Fix those first. But understand the bigger ongoing cost is the data you are throwing away unnecessarily.

You are not over-collecting. You are over-deleting.

The mistake I see again and again: marketers treat GDPR as a list of things they must stop doing, panic, and switch everything off the moment someone clicks reject. They end up more blind than the law ever required them to be, then blame the regulation for a dashboard they hollowed out themselves.

GDPR did not take your analytics. It asked you to sort your data into two piles and handle each one correctly. The anonymous pile - the visits, the sessions, the conversion rates, the country-level trends - was always yours to keep, consent or no consent.

If your dashboard goes dark on "Reject All," that is not GDPR working. That is your architecture failing.

So go look at your own numbers. Pull the day your consent banner went live and compare traffic before and after. How big was the drop?

Now ask yourself honestly: how much of that vanished data was actually personal, identifying data the law required you to stop collecting - and how much was anonymous, aggregate measurement you were always free to keep, and simply chose to throw away?


Live traffic quality

Updated just now

Visits · last 24h

487
Real users
35873.5%
Bots · auto-filtered
12926.5%

Without filtering, 26.5% of your reported traffic is bot noise inflating dashboards and draining ad spend.

Don't trust your analytics!

Make confident, data-driven decisions withactionable ad spend insights.

Setup in 2 minutes
No credit card