How we test and rate casinos

Why the method comes before the score

A score is compact, and that compactness is exactly why it gets believed too easily. It says nothing about what was measured, for how long, or with what interest attached. We publish the method first and the score after: only then can you check whether our conclusion rests on something you would have checked yourself.

What we measure is narrower than it looks. We assess how an operator behaves toward one ordinary player: at sign-up, when setting a limit, on a withdrawal, and when asking customer service a question. We have no access to an operator's own records and none to a regulator's supervisory file. A score from us is an observation from outside, not a seal of approval.

The six components we weigh

The assessment consists of six components, listed below in the order in which they count: licence and duty of care weigh heaviest, presentation and terms weigh lightest. That weighting is fixed before a test starts and is identical for every operator, so a weak component cannot quietly be made lighter afterwards.

We do not publish the weighting as a percentage split. A percentage suggests a precision the underlying judgement does not have: that customer service answers one question clearly is an observation, not a measurement to two decimal places. What is fixed is the ranking — and that it does not shift between operators.

What our own account shows us

At every operator on our list, the editorial team opens an ordinary player account, at our own expense. We do not request a press account and do not accept credit from an operator, because a prepared account does not go through the route you go through.

We record that route first: what data is asked for, how identification runs, and at what point the exclusion register is checked — before or after the first stake. See iDIN identification explained. Then we look at the tools inside the account: where the deposit and play limits sit, how many steps it takes to find them, and whether lowering one takes effect immediately. That is where two operators that look identical on paper start to differ.

How we measure a payout

For every withdrawal we record four things: the time of the request, the time the money lands in the account, the method used, and whether additional verification was asked for along the way. That last point causes delay more often than the payment method itself, and it is exactly where an operator can make you wait without doing anything wrong.

One measurement is not an average. We do not publish an average payout time across operators and present our measurement as what it is: one withdrawal, with one method. What the method itself allows sits alongside it: with iDEAL we reckon on one to two working days, and that window belongs to the method, not the operator — see the page on iDEAL among our payment methods. If our measurement lands well outside that window, that is an observation about the operator.

We check the licence in the register, not with the operator

An operator stating it holds a Dutch licence is not a source. We look the operator up in the Kansspelautoriteit's Kansspelwijzer and record what stands there: the authority, the licence number, the holder and the status. The holder is often a different name from the brand — behind Kansino sits Play North Limited, behind JACKS.NL sits JOI Gaming Limited — and that name matters more than the logo, because the duties attach to the holder. See what a Ksa licence means.

This is also where our data collection visibly stops. Ten of the twelve operators carry the status "actief" in our records; for the other two that field is empty, and we do not fill it with a plausible-sounding status. Where our own record and the register differ, the register is the one that counts.

What a weak point means in our reviews

For every operator we note one strong and one weak point. That is not a balanced scale, it is the shortest way of saying what you would notice first. A weak point in our reviews is an observable trait, not a warning: a small live selection, customer service that replies slowly by email, bonus terms tucked away deep in the site.

Whether that affects you depends on you; a narrow live offering is no drawback for someone who plays one game only. What we do is keep the weak point worded the same until we retest it — even if the operator reports that it has since changed. Such a report is a reason to look again, not the outcome itself.

Why our scores sit close together

The scores across our reviews run from 7.0 to 8.4. That narrow band is not carelessness, it is a consequence of the selection: an operator without a Dutch licence does not get a low score from us, it does not appear at all. The score distinguishes within a group already bound by the same legal requirements.

Read it accordingly. A 7.0 does not mean an operator is untrustworthy, and an 8.4 does not mean playing there is favourable. The score is about handling: how fast the money comes back, how easy the limits are to find, how clearly the terms are written. It says nothing about the odds of the game itself — those do not change with the operator.

When a score changes or disappears

A score belongs to a test, and a test belongs to a date. Every review carries the date we last updated the data behind it; without that date, a score does not tell you whether it concerns the current situation.

We retest when something changes in what we weigh: the payout, customer service, the tools in the account, or the licence status. If a licence no longer shows as active in the register, the score disappears and the page stays up with a note explaining why — a review without a score is more honest than a score attached to an invalid licence. At present, all twelve operators in our records show as active.