Monitoring on an open engine
Each transaction from your channels is evaluated against the rules and typologies you manage, on an open-source transaction-monitoring engine hosted by the Linux Foundation. Your team can inspect how an alert was raised.
AML and fraud detection software for banks, SACCOs and MFIs in Kenya
Transaction monitoring, typology rules you can test before they go live, and case management for your analysts. Delivered through IntegWatch, our financial crime intelligence product, on an open-source monitoring engine.
What this practice covers
Rules you test before go-live
Write typology rules for patterns such as structuring, round amounts and high-risk jurisdictions, then run them in simulation on sample or historical transactions before they touch the live alert queue.
Alerts your analysts can work
Alerts open as cases with a priority, a score and a plain-English reason, so analysts work a ranked queue instead of a spreadsheet. Each decision stays on the case with its evidence.
Connected to your channels
Transactions are mapped in from core banking, mobile money and payment gateways, so monitoring covers the channels your customers actually use.
Support for regulatory reporting
Cases keep the evidence your compliance team needs to prepare suspicious transaction reports. Your money-laundering reporting officer decides and files; we set up the workflow.
Honest about what is still in development
Sanctions and PEP screening, a single customer view and network analysis are in development. We say so plainly, and we do not publish detection rates.
How we can work together
Monitoring implementation
Deploy and configure the open-source monitoring engine in your environment: channel connections, typology rules, case management, testing and handover to your teams.
IntegWatch early access
Evaluate IntegWatch, our financial crime intelligence product, on a scope, data set and success measures you agree with us.
Rules and tuning support
Ongoing rule writing, simulation and tuning with your compliance team, as your products, channels and regulatory guidance change.
Why monitoring falls behind
Most institutions do monitor transactions. The trouble is what happens next. Rules written years ago fire on the same harmless customers every day, analysts clear alerts from a spreadsheet, and nobody can say with confidence why a rule exists or what changing it would do.
Meanwhile money moves faster than ever: mobile money, agency banking, instant transfers and payment gateways. Monitoring that only sees the core banking ledger misses most of it.
What we deliver
- Transaction monitoring. Every transaction from your channels is evaluated against the rules and typologies you choose. Fraud and AML typologies can each have their own thresholds, queue and owner.
- Typology rules and simulation. Your team writes rules in Rule Studio and runs them in simulation on sample or historical data, so you see what a rule would flag before it reaches the live queue.
- Case management and alert triage. Alerts arrive as cases with a priority, a score and a reason written for people, not engineers. Analysts investigate, decide and record why.
- Regulatory reporting support. Cases hold the evidence your compliance team needs for a suspicious transaction report. Filing stays with your institution.
- Integration with your channels. Connection Studio maps data from core banking, mobile money and payment gateways into the engine. See Integrations for the wider integration work.
Who it is for
Banks, SACCOs, microfinance institutions, payment service providers and mobile-money operators in East Africa. Transaction monitoring for a SACCO looks different from a bank’s: fewer channels, smaller teams, and members who expect their savings to be safe. We scope the rules and the queue to the team that will work them.
An open engine, and IntegWatch
Our monitoring runs on an open-source transaction-monitoring engine hosted by the Linux Foundation. Because the core is open, your team can inspect how transactions are evaluated instead of trusting a black box.
We deliver in two ways. We implement the engine in your environment, or you evaluate IntegWatch, our own financial crime intelligence product built on it, which is now onboarding early-access partners.
What we will not claim
We do not publish accuracy or detection rates. The results that matter are the ones measured on your own data, with your own analysts, during an evaluation we agree together. Screening, a single customer view and network analysis are in development, and we will tell you exactly where each one stands.
How an engagement runs
- Discovery. Your channels, volumes, current rules, alert handling and reporting duties.
- Design. The typologies to start with, the data each needs and who works each queue.
- Build and simulate. Connections, rules and case management, tested in simulation before anything goes live.
- Go live and tune. We review alerts with your analysts and tune rules as patterns change.
If you also run core banking or payments with us, see financial technology. For security assessments, penetration testing and hardening of the systems around your monitoring, see Cybersecurity.
Reviewing how you monitor for fraud and AML?
Tell us about your channels, your volumes and how alerts are handled today. We scope transaction monitoring with your compliance, risk and IT teams, starting with a discovery phase.