Checking functions can live in two places: inside the platform running the business, or in a separate system connected to it. The choice looks technical and turns out to be operational, because it determines what the checks can see, how quickly they see it, and how much work an operator spends keeping two views of the same account aligned.
Where AML compliance software sits in the architecture shapes everything about how it performs in daily use. Soft2Bet, a leading iGaming turnkey solutions provider delivering high-quality products and services for online gambling operators, builds these functions into the platform itself.
Two Places the Function Can Live
An external arrangement receives data from the platform, evaluates it, and returns findings. Each of those three steps introduces a point where something can be delayed, mismatched, or lost, and all three have to be maintained indefinitely. It works, and it introduces a boundary: information has to be exported, formats have to be reconciled, and both sides hold a version of the same account that agrees only as often as the exchange runs.
A built-in arrangement removes the boundary entirely. The checking functions read the same record the rest of the operation reads, at the moment it changes. There is no transfer, no delay, and no second copy to keep aligned, which eliminates a category of problems rather than managing it. The distinction between eliminating and managing is worth dwelling on. Managed problems consume attention indefinitely; eliminated ones stop appearing in the operator’s week at all.
What Proximity to the Record Provides
Sitting next to the account record rather than at the end of an export gives the checking layer several practical advantages.
Checks built into the platform can:
- evaluate activity as it happens instead of in batches
- see the complete account history without assembling it
- apply the same standard across every brand at once
- surface findings directly where the account is managed
- reflect changes to the account immediately
One Standard Across a Portfolio
An operator running several brands has a particular interest in this. Portfolio growth is where architectural decisions made for a single brand tend to reveal what they cost. Built into the platform, the checking layer applies one consistent standard everywhere, and a new brand inherits it on launch without a separate setup.
External arrangements tend to fragment as a portfolio grows. Each brand acquires its own connection and its own configuration, they drift apart over time, and the operator ends up maintaining several versions of what should be a single approach. Consistency becomes an ongoing effort instead of a property of the system. Anything requiring ongoing effort to stay consistent will eventually stop being consistent, usually during a busy period when nobody has time to check.
The Cost of the Boundary
The work created by a boundary is easy to underestimate because it never appears as a single task. It shows up as data mapping to maintain, exchanges to schedule and verify, discrepancies to investigate, and a second interface for people to learn.
None of that produces anything of value on its own; it exists purely to keep two systems agreeing. Removing it does not just save the effort. It removes the possibility of the two views disagreeing at exactly the moment an operator needs to trust what it is looking at. Confidence in the record is difficult to quantify and immediately obvious in its absence, particularly when a decision has to be made quickly.
Conclusion
Where the checking functions live is the kind of architectural decision that never appears in a demo and shapes daily work for years. Built into the platform, they see everything as it happens and apply one standard across the business; connected from outside, they see a copy, on a delay, and generate maintenance of their own. Neither arrangement is difficult to set up; only one of them stays simple as the operation grows.