The Dutch regulator stopped writing letters and started acting.
The Autoriteit Persoonsgegevens has been checking Dutch cookie banners systematically since April 2025. By November 2025 it had warned more than two hundred websites, and roughly three quarters of them changed something. This is the market where the consent question became real.
Start with the free 15 point site check. Two working days, no call required.
Featured case study
373
products migrated
- 822feed offers, zero errors
- 4.3%conversion rate across all campaigns, Jul to Aug 2026
What is actually different about selling into the Netherlands
The specifics that change the build, not a list of countries I will work in.
The regulator
The Autoriteit Persoonsgegevens, and it is currently among the more active authorities in Europe on cookie banners specifically rather than on data protection in general.
What the site must publish
Article 3:15d of the Dutch Civil Code, the Dutch end of the same European provider information duty that gives Germany its Impressum. Plus your KvK number and VAT number.
Checkout
iDEAL is the default a Dutch buyer looks for, and it is moving onto open banking infrastructure under the Wero scheme. A checkout without it loses sales it never sees.
The trust signal that matters
Thuiswinkel Waarborg. For Dutch consumer retail the trustmark carries weight that no amount of design does, and its absence is noticed.
What usually goes wrong on a Dutch site
A banner that fires before the choice
Exactly what the AP has been writing to people about since April 2025. It is provable from outside in about thirty seconds, which is why the enforcement campaign works.
A cookie wall with no real refuse option
Accept prominently, refuse buried two clicks deep or absent. The Dutch regulator has been specific that refusing has to be as easy as accepting.
No iDEAL at checkout
Cards and PayPal only, on a site selling to Dutch consumers. The basket gets built and abandoned at the payment step, and analytics records it as a pricing problem.
English pages for a Dutch search market
Dutch buyers read English comfortably, which is exactly why this mistake survives. They still search in Dutch, so the English page is not in the race.
Not sure which of those it is? Send the web address and the written 15 point check says so in two working days, free.
What the work actually is
The same list whether you are in the Netherlands or anywhere else I work.
The 15 point technical check
Speed, indexing, structured data, redirects, consent and the rest, written out in the order worth fixing rather than the order a crawler found them.
Pages for what buyers search
Built from real search data in the language your buyers type, not translated from an English page and hoped over.
A redirect map before anything moves
Every old address mapped to a new one and tested before launch, because this is the step that loses rankings when it is skipped.
Tracking that waits for consent
Nothing loads before a choice is made, and the banner says what it does. This is the part regulators are actually acting on.
Google Ads on a flat fee
A fixed monthly amount whatever the spend, so the advice does not get more expensive when the budget does.
Shopify and WooCommerce moves
Products, variants, media and URLs moved with the search positions held, on the same domain where that is possible.
Custom systems when software is the problem
Order flows, stock feeds and internal tools, when the honest answer is that no amount of marketing fixes the process underneath.
Plain monthly notes
What was done, what it changed and what is next, in a page you can read without a glossary.
Want the number first? Every price is published, including the ones I would not charge you.
What the enforcement campaign actually asks of you
The Dutch campaign is not about having a banner. Almost every site has one. It is about whether the banner does anything, and the test is mechanical: open the page with a clean browser, make no choice, and see what has already loaded. If analytics, a pixel, a map or a font from a third party fired before the click, the banner is decoration. That is the finding behind most of the two hundred plus warnings.
Fixing it is engineering, not paperwork. The scripts have to be genuinely held until consent rather than loaded and told not to report, the refuse control has to be as reachable as the accept control, and the privacy policy has to describe what actually runs. I have written the full reasoning, including which analytics tools survive the test and which do not, in the guide to GDPR and analytics.
Dutch checkout expectations, which are specific
The Dutch market has the most distinctive payment behaviour in Western Europe. iDEAL is what a Dutch buyer reaches for by default, and the scheme is migrating onto newer open banking infrastructure, so a checkout integrated years ago is worth re-checking. For consumer retail the Thuiswinkel Waarborg trustmark does work that design cannot, because it answers the only question a cautious buyer has about an unfamiliar shop.
I have done this work on a Dutch retailer: three brands, three storefronts, three ad accounts and a CRM, with an order cycle of ten to twelve weeks that made the online figures lie about how the business was doing. The write up is at the furniture retailer case study, and the measurement problem it describes is the one most Dutch retailers with a long lead time have without knowing it.
A furniture retailer: three storefronts, three ad accounts and a CRM
A Dutch furniture retailer with three brands and a ten to twelve week order cycle, where the online numbers alone made a busy business look like a failing one until the measurement was fixed.
What I will not do
Said before you ask, because it is usually the deciding part.
Promise a position
Nobody controls Google's results, and anyone who promises a ranking is either guessing or lying. I will tell you what is winnable and what is not, with the search data in front of you.
Hold your accounts
Analytics, Search Console and every ad account are registered in your name from day one. If you stop working with me you lose nothing but me.
Bill a percentage of ad spend
A percentage pays me more when you spend more, which is the wrong incentive. Advertising is a flat monthly fee whatever the budget.
Take the work if it is the wrong spend
If the free check says your problem is not a website problem, the report says so and I tell you what would actually move the number.
Rather just ask? One question by email or WhatsApp is fine.
The rest of the Benelux
Neighbouring markets, each with its own regulator, language and checkout.
Every market I work in is listed on where I work, with what differs in each one.
Fair questions before you get in touch
You are not based here. Does that matter?
It has not so far. Everything runs by email and WhatsApp, the accounts are in your name, and the work is the same work wherever it is typed. I am in Pakistan and I say so on every page, because finding that out later is worse than reading it now.
Has the Autoriteit Persoonsgegevens actually fined anybody over a banner?
The 2025 campaign has mostly produced warnings, and roughly three quarters of the sites warned changed something. Treating a warning as the worst case is a bet on staying in the first group rather than the second.
Can we keep Google Analytics on a Dutch site?
Behind genuine consent, with the tag held until the choice is made, it is a question about consent quality rather than about the tool. If you would rather not depend on consent for your basic numbers, a self hosted or EU hosted measurement tool removes the problem. The guide sets out the trade.
Do we need Dutch pages if our customers speak English?
They do speak English, and they search in Dutch. The pages that bring buyers have to be in the language of the search, which is why this is the one place I will not compromise on translation.
Why are the prices on the website?
Because the alternative is a quote form, and a quote form exists to get you on a call before you know the number. The free 15 point check, the 290 euro audit and the monthly and rebuild prices are all published. You can decide before you ever write to me.
How long before any of this shows up in search?
A new page on a site that already carries some authority can move within weeks. On a site with very few links pointing at it, expect months, and expect the first movement on specific questions rather than on the obvious commercial terms. I will not give you a date, because nobody controls the result. What I will give you is the order of the work and what each step is meant to change.
Send your web address. That is the whole first step.
The written 15 point check comes back within two working days, free. It says what is holding the site back, in the order worth fixing.