Consent Management
For consent to work, that choice has to travel from the banner through the consent setup to the tags that collect and send data. If something breaks along the way, tags load when they shouldn't.
1 in 3
large and mid-sized Dutch organisations start tracking visitors before they choose. On most of them, the cookie banner is still on screen.
Cloud Nine Digital Consent Research 2026, 1,925 websites in six sectors
Is our consent setup actually working?
There's one way to find out. Check what the website does before and after someone makes a choice.
In October 2026, Cloud Nine Digital measured the homepages of 1,925 large and mid-sized Dutch organisations across six sectors. On 1 in 3, tracking had already started before the visitor chose anything. On 1 in 5, we confirmed a marketing cookie or advertising identifier.
When we test an implementation we look at three points
Consent management platform
Does the banner record the visitor's choice correctly?
Consent signal
Is that choice passed correctly to the platforms that need it?
Tag configuration
Do the tags actually respond to the signal they receive?
Our Consent Scanner checks what happens in the browser under different consent states. In other words, it shows you what your tags are actually doing.
If Google Analytics loaded before analytics consent was given, that's a problem worth knowing about.
Where are you starting from?
Most teams come to us in one of three situations. The chain is the same in all three, only the starting point differs.
Something went wrong
A letter, a complaint or an audit finding
You need to know what is really running, fast, and fix it in the right order.
Something is changing
A new platform, a redesign or a new market
The chain gets rebuilt, so it gets tested before it goes live.
You want proof
It works today. Will it after the next release?
Every new tag, vendor or release can break it again. You want evidence that it still holds.
From finding to fixed, and keeping it that way
A scan tells you what's wrong. It doesn't stop the next release from breaking it again. So we test the existing implementation, fix the technical setup, hand over ownership, and monitor it after delivery.
Audit
Banner, signal and tags tested on every domain in scope. You get a findings report, prioritized by impact.
Fix
One blueprint domain first, then rolled out across the rest. A working configuration, evidenced per domain.
Embed
An internal owner, release checks and a rulebook, so the next new tag doesn't quietly undo the work.
Maintain
Continuous checks, a named contact and monthly findings flagged by impact.
We don't sell a consent platform. We check whether yours does what it says.
We work with OneTrust and other consent platforms, and with the analytics and tagging setup that receives their signals.
What happens when someone declines?
Tags that require the declined consent should not load. Other systems work with restricted or cookieless signals, depending on the platform and configuration.
70%+ of ad-click-to-conversion journeys lost to consent choices are recovered by modeling, according to Google. Source: Google, conversion modeling documentation.
Modeling does not turn a decline into a yes.
It uses observable, consented data and other eligible signals to estimate outcomes where direct measurement is unavailable.
Will we lose data?
You collect less directly observed data when visitors decline. That is what declining consent means.
We're not going to look for a workaround to a choice someone has already made. We set up the measurement to respect it.
Consent Mode and other modeling methods fill some gaps. Server-side measurement and better first-party data collection help where they fit. None of them overrides the visitor's choice.
A large Dutch member organisation: 101 cookies, 16 that mattered
We tested 23 customer journeys on the main website, banner, signal and tags. They set 101 cookies between them, but only 16 behaved differently from one journey to the next. The journeys fell into 13 groups that behave almost identically. So instead of scanning thousands of pages, we now test one journey per group each quarter, rotating. Within a year every group is covered several times, at a fraction of the cost of a full crawl.
Check what your consent setup actually does
Test the tags that load before and after consent choices, and see where the implementation needs a closer look.
A banner tells you what visitors see. The Consent Scanner checks what the website does.
Related services
Contact details
Cloud Nine Digital
Johann Siegerstraat 24
1096 BH
Amsterdam, Netherlands
T: +31 (0)20 787 1323
E: info@cloudninedigital.nl
Need something else?
Please do reach out. We are happy to help!