# Engineering Support for an iGaming Fraud Queue

> TechSupport Team supported PayLucky’s Python and Kafka fraud-review application. In the five-month comparison, the client’s fraud console shows the chargeback rate falling from 0.92% to 0.50%. The client’s risk team approved changes to fraud rules.

- Publisher: TechSupport Team
- Canonical page: https://techsupport-team.com/case-studies/igaming-payments-fraud-queue-engineering/
- Updated: 2026-11-16

Client: PayLucky · iGaming & Betting · Maltese iGaming payments provider

## Client and engagement facts

- **Client:** PayLucky
- **Industry:** iGaming & Betting
- **Business context:** Maltese iGaming payments provider
- **Delivery model:** Engineering support for fraud review
- **Application:** Fraud detection and review for iGaming merchants
- **Supplied geography:** Maltese
- **Coverage:** 24/7

## Application stack and tools

### Application stack

- Python
- Kafka
- PostgreSQL
- AWS

### Tools used

- Client fraud console
- Datadog
- Zendesk

Related services: [Application support](https://techsupport-team.com/services/application-support/)

## Why Support Was Needed

Analysts were handling 1,400 reviews a day, and rule changes could affect legitimate transactions. The work focused on routing velocity, location mismatch and card-testing signals for review.

## Technical Work

- Shadowed 40 hours of analyst review to map the actual decision tree.
- Built a triage lane flagging velocity, geo-mismatch and card-test patterns.
- Added a card-test detection rule after observing a five-day attack in the console.

## Representative Incident

In month two, the new triage lane flagged a card-testing wave. The case records 340 suspicious transactions blocked in 40 minutes. Client analysts retained oversight for edge cases.

## Results and how they were measured

Table: Reported results and their measurement scope

| Metric | Baseline | Result | Units | Measurement window | Sample / population | Source and scope |
| --- | --- | --- | --- | --- | --- | --- |
| Chargeback rate | 0.92% | 0.50% | % | 5 months | Client fraud-console chargeback measure; the client risk team approved rule changes. | Client fraud console; Scope: The case excludes a nine-day chargeback spike after a card-scheme rule change. |
| Review queue | 1,400/day | 620/day | /day | 5 months | Client fraud-review queue. | Client Datadog |
| False positives | — | -38% | Result: % | 5 months |  | Client Zendesk |

## Evidence and Limitations

The case excludes a nine-day chargeback spike after a card-scheme rule change. False positives are reported as a relative change, with no absolute baseline supplied.

## Fraud Rule Approval

TechSupport Team proposed changes to fraud rules. PayLucky’s risk team approved each change.

## Scope a Similar Engagement

Describe your review application, recurring queue failures and the person who approves rule changes. [Request a support plan](https://techsupport-team.com/contact/) or explore [application support](https://techsupport-team.com/services/application-support/).

## Related pages

- [Application support](https://techsupport-team.com/services/application-support/): The related service for this engagement.

## Contact

Request engineers from TechSupport Team or ask a question: https://techsupport-team.com/contact/
