Over the past year we made the biggest set of changes to compliance screening in Business Radar’s history. The best ideas did not start on our roadmap. They came from compliance teams telling us, in demos, support tickets and quarterly reviews, exactly where screening tools fall short. Here is what we changed, and the thinking behind it.
We built our own compliance database
One change happened entirely behind the scenes. Business Radar now collects sanctions, PEP and government-linked entity data directly from the source, in our own compliance database, rather than licensing it from a third-party data vendor.
We did it to get two things back: speed and control. When your data comes from a vendor, you inherit their refresh cycle, their entity model and their limitations. When a customer asked us why a list change took days to appear, the honest answer used to be “because that is when our vendor ships it.” That answer was not good enough for teams whose exposure changes the day an entity is listed.
With our own database, list changes reach monitoring alerts the same day they occur. It also means that when a customer needs a source we do not yet cover, we can add it ourselves instead of filing a request with someone else. Since making the switch we have extended screening beyond the core lists to sources such as the Offshore Leaks database and court cases from official registries, and the source set keeps growing, driven by what our customers actually need to screen against.
Screening now follows your risk appetite, not ours
Every compliance team we work with has a risk framework. Almost every screening tool ignores it and applies the same rules to everyone.
We changed that. Compliance settings in Business Radar now let your organisation define what gets screened, monitored and reviewed. If certain sanction lists fall outside your scope, you exclude them and they stop generating hits. If your PEP policy demands precision, you require a full last name match before a person even reaches review. And you decide what happens to false positives our AI review identifies: keep them out of the queue entirely, or keep them visible if your policy requires eyes on everything.
How much of this is fixed centrally is itself a choice. Organisation admins can set the policy once so every screening runs to the same standard, or give analysts room to adjust settings themselves where the organisation allows it. Either way, results are shared across the team, so everyone works from the same picture. The effect our customers describe is twofold: consistency for the compliance lead, and review queues the analysts can actually clear.
If our AI makes a call, it shows its work
Automated false-positive detection is only useful if your team can trust it, and trust does not come from accuracy alone. It comes from being able to check.
When Business Radar now rates a match as a false positive, it shows the reasoning behind that decision. Analysts can review the logic, agree, or override it. Every decision, automated or human, lands in the audit trail. We think this is what AI in compliance has to look like: not a black box that filters your hits, but a colleague that explains itself and accepts being overruled.
The audit trail itself was redesigned around the same principle. Screenings now document risk signal scores per category, a full breakdown per entity, AI review notes with confidence levels, and reviewer notes and decisions. When an auditor or regulator asks how a match was assessed, the answer is on the page.
Smaller changes with large impacts
Some of the most appreciated changes were requested by our users. Compliance checks now cover aliases, which matters more than it sounds: the same entity can appear under different spellings, in different languages, even in different alphabets, and a screening that only knows the Latin registered name will walk straight past a hit in Cyrillic or Arabic transliteration. Portfolio compliance screening went from one-by-one to a few clicks, so periodic reviews of hundreds of counterparties no longer consume days of analyst time. And compliance results now carry notes and mentions, so the discussion about a hit happens on the hit itself and ends up in the audit trail, instead of living in an email thread nobody can find a year later.
One request turned out to be bigger than it first sounded: screening individuals without a company affiliation. It started as a practical ask, but it changes what Business Radar can be used for. Due diligence does not stop at directors and shareholders. Compliance teams also need to screen private individuals: prospective clients, investors, board candidates, business partners. Business Radar now handles a standalone person with the same rigor as a company, sanctions, PEP and enforcement checks included, without needing a corporate entity to hang the screening on.
These did not stem from internal product strategy sessions. Customers walked us through their actual workflows and pointed at the friction. We have come to see this as the way our roadmap should work: our customers have a need and we see if we can make it possible.
What this adds up to
Sanctions screening has a reputation as a box-ticking exercise: run the check, export the report, move on. We think that reputation exists because the tools made it that way. Screening built on stale vendor data, generic matching rules and unexplained automation produces exactly the noise and distrust that compliance teams have learned to live with.
We are trying to build the opposite: screening on data we control and update the same day, rules that reflect your risk appetite instead of our defaults, and automation that explains every call it makes. Judging by what our customers ask us to build next, we are on the right track, and we intend to keep working this way.
If you want to see what this looks like on your own counterparties, book a demo.
