How we rate casinos

We verify every casino's regulatory licences against official public registries — but the licence is a starting point, not the whole picture. A strong regulatory regime matters, but so does the operator's actual track record: how long they've been running, how they handle complaints, and whether they've survived scrutiny at scale. A Curaçao casino operating reliably since 2017 is a different proposition from one that launched last month. Longevity earns a bonus because time is the hardest test to fake.

We read each casino's record at an independent complaint mediator. What counts is how decided complaints ended: those resolved against those left unresolved. A complaint the mediator rejected as unfounded, or one still open, counts neither for nor against. A casino with 90%+ resolved scores full marks; fewer than five decided complaints is too few to score and takes the standard 1.6.

How quickly can you actually get your money? We track reported withdrawal times. Same-day payouts score the maximum. Anything over 5 days is penalised heavily.

This is where Frankly goes deeper than anyone else. Wagering is the hidden cost of every bonus, so we work out what it takes to clear $100 of bonus — the multiplier and what it applies to (bonus only vs deposit + bonus) — and place that on the Frankly index of every live welcome bonus. A bonus at the index median scores 1.5: standard terms are not a penalty. Sticky structure, cashout caps, short expiry and zero table or live contribution cost points; non-sticky earns one. Every bonus is placed on the Frankly bonus index rather than an industry figure: across 28 live welcome bonuses (November 2026) the median wagering is 40x, and clearing a $100 bonus takes a median $4,000 of turnover (quartiles $4,000–$4,500). A bonus around that median is standard; sticky or deposit-inclusive terms are what cost more.

Does the casino publish Return to Player percentages? A casino that runs its own provably fair games, where every bet can be verified on its site, scores the maximum; a few third-party provably fair titles in a slots lobby don't count. Per-game RTP data gets strong credit. A general average gets partial credit. No RTP data published scores zero.