Transaction Monitoring
Detect suspicious activity with configurable rules and an AI risk score on every transaction, turn alerts into cases your team can resolve, and support suspicious matter reporting.
30 days of full access with 500 test credits. No credit card required.
- Payments
- Transfers
- Account activity
- Review
- Escalate
- Report
- 0 to 100AI risk score on every transaction
- API + batchReal time by API, or CSV batch upload
- Audit trailCase, rule and model changes are logged
- 30 daysFree trial, 500 test credits, no credit card
From the first transaction to the report.
- IngestSend transactions in real time by API, or upload a CSV batch.
- DetectRules and scenarios flag patterns while the AI scores each one.
- ScoreAlerts are ranked by risk. A high AI score can open a case on its own.
Judgement needed? It moves to your team
- ReviewAnalysts work each case with evidence, notes and a decision trail.
- EscalateSenior review where the risk warrants it.
- ReportFindings compiled for suspicious matter and activity reports.
MemberCheckFraudShieldNew · AI rules engine
An AI engine that scores every transaction alongside your rules.
Rules catch what you define in advance. The AI engine scores every transaction too, with models trained on your finalised rule plans.
- AI risk scoringA 0 to 100 score on every transaction, whether or not a rule fires.
- Cases from the AI scoreNo rule fired but the score is in a range you enable? A case opens at the matching risk level.
- Scores you can explainEvery score shows the rule breakdown behind it: impact, influence and direction.
71AI risk score
- ↑New counterpartyHigh influence, raises the score
- ↑Unusual for this customerMedium influence, raises the score
Case opened from the AI score
Illustrative. Scores and ranges are examples.
See how a transaction is handled.
Pick a scenario to see what the rules and the AI score do with it.
A regular supplier payment, in line with the customer's history.
- Matches expected behaviourHigh influence, lowers the score
No rule fires and the score is low, so no alert is raised.
Nothing for an analyst to review.
Nine transfers, each just under a limit, sent within a short window.
- Structuring pattern
- Velocity
- Structuring patternHigh influence, raises the score
- VelocityMedium influence, raises the score
Two rules fire and the AI score is high, so the alert is ranked near the top of the queue.
An analyst opens a case, adds evidence and decides, escalating to senior review where the risk warrants it.
A large payment to a first-time payee, unlike this customer's history.
- New counterpartyHigh influence, raises the score
- Unusual for this customerMedium influence, raises the score
No rule fires, but the score falls within a range you have enabled, so a case is created at the matching risk level.
The case arrives with the score and the breakdown behind it.
Illustrative scenarios, not live data. Scores and rules are examples; yours are set to your own risk appetite.
Detection, investigation, reporting and control in one platform.
Detect
- Rules and scenarios. Default rules from domain specialists, or your own in the Rule Builder.
- Reference lists. Country risk, monitored keywords, IP and wallet-address blacklists, plus your own by CSV.
- AI risk score. An independent 0 to 100 score, explained by impact, influence and direction.
- Customer profiles. Built automatically from each transaction, with changes between transactions highlighted.
Investigate
- Case workflow. Open, Investigating, Escalated, On Hold and Closed, at Urgent, High, Medium or Low risk.
- Assignment. Manual, round robin or least assigned, with a reason recorded on every escalation, hold and closure.
- Evidence. Attach up to 10 files to a case and keep notes on the record.
- Dashboard. Case volume, risk mix, case age and workload by assignee at a glance.
Report
- Regulatory report templates. Start from a goAML-based template, adapt it, and generate reports from a case.
- PDF, CSV or XML. Generated reports are fixed: editing a template never alters one already produced.
- Case reports. A styled PDF with the transaction details and the AI risk assessment.
- Reporting flag. Flag rules for regulatory reporting, and filter cases by that flag.
Govern
- Role-based access. Nine roles, so rules, cases, billing and user management sit with different people.
- Audit logs. Case, rule and AI model changes are logged, searchable by date, user, entity and event.
- Locked rule plans. A finalised plan can never be edited or deleted. Optional closure review holds a case at Proposed Closure until a reviewer approves.
- Test and production. Isolated environments, each with its own active rule plan, and API keys that expire and can be revoked.
An alert nobody can clear is not a control.
Rules are easy to add and hard to retire, so alerts pile up while the team stays the same size. Scoring every alert keeps the ones that reach a reviewer worth their time.
- Rules set per customer segment, not one threshold for the whole book
- Alert scoring puts the strongest signals first, not the oldest
- Tuning decisions are recorded, so you can explain them to a supervisor
- 78Structuring pattern, repeatedCase open
- 71New payee, large amountCase from AI score
- 52Velocity above segment normIn review
- 12Routine supplier paymentCleared
Illustrative. Scores and ranges are examples.
Transaction monitoring for regulated businesses.
Test it with your own scenarios.
30 days of full access with 500 test credits. No credit card required.
- Full product accessRules, alerts, cases, dashboards, APIs and workflows.
- Your own dataStart with a sample feed or upload your own transactions.
- CustomisationTest the default rules and the custom rules engine.
- SupportAPI documentation and help centre. No sales call or setup.
Planning an enterprise rollout? Contact sales about custom arrangements.
Questions about transaction monitoring.
Does monitoring run in real time or batch?
Both. Submit a single transaction through the API and it is assessed immediately against your active rule plan, with the rule results, the AI risk assessment and any case created returned in the response. Or upload a CSV batch in the app, which is processed in the background, shows live status and error files for any rows that fail, and can email you when it completes.
Can we configure our own rules and scenarios?
Yes. Rules, thresholds, and scenarios are configurable to your products, customer types, and risk appetite, from velocity and structuring patterns to peer-behaviour comparison, so monitoring reflects the activity you actually see rather than a fixed template.
How does it reduce alert noise?
MemberCheck combines threshold and scenario rules with behavioural comparison and scores each alert by risk, so analysts work the highest-priority cases first instead of losing time to low-value alerts from fixed limits alone.
Does it support suspicious matter and transaction reporting?
Yes. The activity, findings and analyst decisions behind a case are kept on the case, and regulatory report templates generate reports from it as PDF, CSV or XML to support suspicious matter or suspicious activity reporting to your regulator. The starter template follows the widely used goAML format, which differs by region and industry, so templates can be adapted. A generated report is not altered by later edits to its template.
What access controls does it have?
Access is role-based, with nine roles including Org Owner, Org Admin, Business Manager and Viewer, Rules Manager and Viewer, Billing Manager and Viewer, and User Manager. Rules, cases, billing and user management can sit with different people, and read-only roles give reviewers visibility without the ability to change anything. API keys are specific to an environment, can carry an expiry date and can be revoked. For questions on single sign-on, data residency or availability commitments, contact sales.
Is there an audit trail?
Yes. Audit logs record events across the system, such as case status, assignee and risk level changes, AI model training and activation, and rule reversions, and can be searched by date, user, entity type, event type or entity ID. A finalised rule plan is locked permanently and can never be edited or deleted, so the rules behind a past decision stay as they were.
How do we integrate it with our systems?
Through a REST API that returns JSON. You can submit a single transaction or a batch, and check the status of a batch. Without an integration, upload CSV files in the app, using a reusable mapping template for your own column names. Customer profiles are built automatically from the transactions you submit.
Can we test safely before going live?
Yes. Test and Production are separate environments with fully isolated transaction data, each with its own active rule plan and its own API keys. Test credits are included, so you can keep testing after you go live.
Can the AI score block a transaction?
No. The AI risk score is a complementary signal to your rules, not a replacement for them, and it never blocks transaction processing. If it is unavailable, transactions are processed on your rules alone.
Does transaction monitoring use AI?
Yes. The AI rules engine trains models on your finalised rule plans and scores each transaction from 0 to 100, independently of rule results. When no rule triggers but the AI score falls within a range you enable, a case is created at the matching risk level.
Can we see why a transaction received its AI score?
Yes. Each AI risk assessment is shown with the rule breakdown behind it, covering impact, influence and direction, and the assessment is included in case report PDFs.
What is included in the free trial?
Full product access with nothing gated, covering rules, alerts, cases, dashboards, APIs and workflows, plus 500 test credits over 30 days. You can test the default rules and the custom rules engine, and use API documentation and help centre support. No sales call or setup is required.
How do we try it?
Use the Start free trial button on this page. The trial gives you 30 days of full access with 500 test credits and needs no credit card. You can start with sample transactions or upload your own. Transaction monitoring runs on its own platform with its own sign-in, separate from your MemberCheck account.
Further reading.
- AnalysisThe AI revolution against money laundering
- GuideMastering compliance challenges in transaction monitoring
- Best practiceTuning monitoring to cut false positives safely
- ExplainerThe benefits of real-time transaction monitoring
- Sector viewTransaction monitoring across diversified industries
- GlossaryAML terminology, in plain English
See MemberCheck in action.
A short walkthrough of screening, verification and ongoing monitoring in one platform, set to the thresholds and jurisdictions your programme actually uses.
Transaction monitoring runs on its own platform, with its own credentials, separate from the MemberCheck screening app. An existing screening account does not sign you in to it. Already a FraudShield customer? Log in (opens in a new tab)