Enter to payments ·

Try2Check

RNG certification renews yearly; the audit trail goes dark in March

Annual RNG certificates renew on a twelve-month cycle, leaving eleven months of unverified play and a visible audit gap each March

5 min read

RNG certification renews yearly; the audit trail goes dark in March
RNG certification renews yearly; the audit trail goes dark in March

Every online casino you play at publishes an RNG certificate with a date on it. Almost none of them publish what happens between those dates. The certificate renews on a twelve-month cycle — a lab re-tests the generator, signs a new PDF, the operator swaps it onto the footer of the site — and for the other eleven months the evidence that the RNG still behaves the way it did at testing time lives inside systems players can't see. The gap is most visible in March, when a stack of certificates issued the previous spring quietly expire and get replaced by documents that look identical but describe a different twelve months of operation.

That's the claim worth pulling apart: certification is an annual event, but randomness is a continuous property. The audit trail has a seam, and the seam is seasonal.

What a certificate actually attests to

A typical RNG certificate from a lab like GLI, BMM, or eCOGRA covers a defined test at a defined moment. The lab takes the operator's (or supplier's) random number generator, runs it against statistical batteries — chi-square, runs tests, serial correlation, poker hands, gap tests — and confirms the output is indistinguishable from random within stated confidence intervals. It also checks the implementation: that the seeded generator feeding the slot reels or the card shuffle is the one being tested, not a decoy.

What it does not do is sit inside the live system for the following year. The certificate is a snapshot with an expiry date. Between snapshots, the operator is responsible for change control: if the RNG code, the seeding method, or the game math changes, the certificate should be re-issued. Whether that happens on schedule is exactly the part nobody publishes.

The practical consequence for a player reading a certificate is that you're trusting two things at once — the lab's test, and the operator's discipline in not silently changing the thing that was tested. The first is documented. The second usually isn't.

The March seam

Certificates cluster. Suppliers tend to schedule recertification in batches, and a lot of them land in Q1 because it follows the calendar-year compliance cycle: annual internal reviews in January, lab slots booked for February and March, new certificates live by spring. The result is a predictable wave of expiries and renewals in March, and a corresponding wave of certificate footers changing on operator sites within a few weeks of each other.

Here's the numerical anchor that makes the seam concrete: a certificate issued on 14 March 2024 and renewed on 14 March 2025 covers 365 days of claimed compliance, but the public artifact for that period is two PDFs and a gap. If you visited the site on 13 March 2025, the live certificate had one day left on it. If you visited on 15 March, it had a fresh one. Nothing in between told you whether the RNG was tested during the twelve months, or just at the two endpoints.

That's not an accusation of wrongdoing. It's a description of what the public record contains. The labs test what they're given, when they're given it. The operator holds the continuous monitoring — and continuous monitoring is the part that stays in-house.

What continuous monitoring looks like on the inside

Operators and suppliers that take this seriously run their own statistical checks on live output: RTP tracking against theoretical, hit-frequency monitoring, and alerts when a game drifts outside tolerance bands. Some publish summary RTP reports; most don't. When they do, the reports tend to be aggregated across a game library rather than broken out per title, which makes them hard to compare against the per-game RTP figures players actually use to choose games.

The asymmetry is the point. The annual certificate is public and standardized. The continuous monitoring is private and bespoke. A player can read the first and has to take the second on faith.

Why the gap matters more for some products than others

For a slot with a fixed math model and a well-tested generator, a year between lab tests is a reasonable interval — the code doesn't move, and the operator's monitoring catches drift. The risk concentrates in products where the RNG inputs change frequently or where the math is bespoke.

  • Live dealer and hybrid games. The RNG may only govern card shoe shuffling or bonus triggers, and the certificate scope can be narrower than players assume. Read what's actually covered, not just the logo.
  • Provably fair crypto casinos. These publish hash seeds and let players verify individual outcomes, which is a different and in some ways stronger transparency model — but it verifies fairness of the shuffle, not the RTP of the paytable. The two are separate claims and get conflated constantly.
  • Custom or white-label slots. If the math is built by a smaller studio and the certificate is held by the platform rather than the studio, the change-control question gets murkier. Who re-tests when the studio patches the game?

None of this means avoid these products. It means the certificate tells you less than the footer implies, and the products with the most moving parts are where the annual snapshot covers the least.

What to actually check, and when

If you care about this — and if you're reading an iGaming article about audit trails, you probably do — the useful habit is to look at the certificate date rather than the certificate existence. A few things worth doing:

  1. Note the issue and expiry dates, not just the lab name. If the certificate is more than eleven months old, you're near the seam.
  2. Check the scope line. Does it cover the specific game, the platform, or the supplier's generator generally? A platform-level certificate is weaker evidence for an individual title than a game-level one.
  3. Look for a published testing-frequency statement. Some operators say they re-test quarterly or on every major release. That's a stronger claim than annual recertification, and it's usually buried in a responsible-gambling or fairness page rather than the footer badge.
  4. Cross-check the RTP figure against the game's help screen and the operator's game page. Discrepancies are rare but they're the kind of thing annual testing is supposed to catch and continuous monitoring catches sooner.

And the usual caveat that belongs here rather than in a footnote: none of this changes the house edge. A perfectly certified RNG at 96.2% RTP still grinds you down over time, and the audit trail is about fairness of the deal, not about whether the deal is in your favor. Set limits accordingly.

The question the industry hasn't answered

The open question isn't whether annual certification is adequate — for most established products, with in-house monitoring, it probably is. The question is why the continuous-monitoring layer stays private when the annual snapshot is public. If an operator can publish a certificate, it can publish a one-line statement of what it monitors and how often. Almost none do.

So the next time you see a fresh certificate appear in March, the useful thought isn't "good, it's certified." It's "certified as of when, covering what, and what happened in the eleven months since the last one?" The industry has a standard answer for the first two. The third is where the audit trail goes quiet — and where a bit more disclosure would cost operators almost nothing and tell players considerably more.