What do you like best about Cookiebot by Usercentrics?
As a compliance product, the consent tool does its job. It scans the site, categorises cookies, and serves a working consent banner, and the initial setup was straightforward. If the billing and support matched the product, I would rate it well. Review collected by and hosted on G2.com.
What do you dislike about Cookiebot by Usercentrics?
The billing model, and the handling of a billing dispute. Cookiebot charges by the number of subpages its scanner finds, but the scanner cannot distinguish a genuine page from a filtered view of the same page. A filter plugin on our WordPress site briefly produced several thousand query-string URLs (e.g. /resources/?filter=…), and a scan on 8 March 2026 counted each as a distinct subpage – over 10,000 "pages" on each of two domains that in reality hold well under 1,000. Both were pushed onto the Extra Large tier, and our monthly charge for the two jumped from £50 (2 × £25) to £156 (2 × £78). Under their terms a charge based on that scan may be technically permissible, and I accept that – though it is neither fair nor client-oriented.
What followed is the real complaint. The fault existed only briefly in March and was fixed soon after; it did not persist for months. Yet Cookiebot kept charging the Extra Large tier through April, May and June, with no sign of a fresh scan in any of them. I caught the overcharge only in June. When I raised it, Cookiebot conceded the page count was wrong but offered to refund a single month as a "gesture of goodwill" – it was not goodwill, it was a plain overcharge.
I then asked for the dated scan behind each disputed month, the one document that would prove the charge. I was told the scan was "the exact same" for every invoice and that sending them would be "redundant"; later, that each scan "accurately reflected the site at the time each invoice was generated". Both cannot be true. If the fault was fixed after March, the later counts were not identical to March – so either no fresh scan was run, and the charges rested on no current data at all, or the claim is simply false. By Cookiebot's own rule that pricing reflects the scan at the time of billing, one March scan cannot justify three separate later months. And the tier only ever moved one way: it escalated automatically when the count rose, but never came back down when the real count fell – always in Cookiebot's favour.
In the end Cookiebot refunded only the final month, refused April and May on identical facts, and declared the matter "closed" because I had raised it months later. Raising a charge late does not make it correct, and a vendor does not settle a dispute by announcing it is closed.
Above all, Cookiebot has refused to produce the actual scans for the disputed months – the only evidence that could show whether those charges were ever justified under its own terms. It insists the charges are correct while declining to prove it. A company confident in its billing would simply show the scans. That refusal, more than anything, is why this is a zero from me. Review collected by and hosted on G2.com.
