Project Summary
When a FinTech company set out to build a real-time fraud detection and transaction risk-scoring platform capable of identifying fraudulent activity across card-not-present, ACH, and digital wallet payment channels in milliseconds, the engineering challenge was significant from the start. Fraud patterns shifted constantly as bad actors adapted their tactics, and the scoring models had to hold to strict latency requirements for real-time payment authorization without driving up false declines on legitimate transactions. The team didn’t fully realize how much of that development work qualified under the IRS R&D tax credit until a formal study was completed. By satisfying the four-part test, the company unlocked $174,000 in federal R&D tax credits and $64,525 in state credits, amounting to $238,525 in total savings.
Project Overview
To qualify for the R&D Tax Credit, each activity must satisfy the IRS four-part test. CSSI’s analysis confirmed that the qualifying activities identified for this company met all four criteria:
- Business Component: The company developed a proprietary real-time fraud detection and transaction risk-scoring platform, integrating machine learning models, behavioral analytics, and adaptive rules engines designed to identify fraudulent transactions across card-not-present, ACH, and digital wallet payment channels, a direct effort to develop a new or improved business component under IRC 41.
- Elimination of Uncertainty: At the outset, it was unknown whether the fraud-scoring models could maintain accuracy as fraud patterns evolved and bad actors adapted their tactics, and whether transactions could be scored within the strict latency window required for real-time payment authorization without materially increasing false declines. The team worked systematically to resolve those uncertainties.
- Process of Experimentation: Data scientists and engineers ran iterative model training cycles against historical and synthetic transaction data, A/B testing scoring thresholds and feature sets in production-like sandbox environments, refining each model version based on measured precision, recall, and latency outcomes at each stage.
- Technological in Nature: The work relied on machine learning and data science, software and systems engineering, and applied statistics.
|
Employee Wages |
$1,020,000 |
|
Supply and Contractor Costs |
$430,000 |
|
Total QRE’s |
$1,450,000 |
|
Total State Credit |
$64,525 |
|
Total Federal Credit |
$174,000 |
Study Results
The analysis identified a total of $1,450,000 in Qualifying Research Expenses (QREs) across the tax year. Employee wages accounted for the largest share, with $1,020,000 attributable to data scientists, machine learning engineers, risk analysts, and software engineers directly engaged in qualifying research activities. Supply costs contributed an additional $210,000 in qualifying expenses, primarily from cloud computing infrastructure, transaction data licensing, and sandbox testing environments used in development and validation. Contractor expenses added $220,000, representing the 65% allowable portion of third-party research costs under IRC 41. Based on those qualifying expenses, the study produced a federal R&D Tax Credit of $174,000 and a state R&D Tax Credit of $64,525, bringing the company’s total tax credit benefit to $238,525.
Key Takeaways
- Core fraud-detection R&D qualifies. The uncertainty this company faced, whether models could adapt to evolving fraud tactics while holding to strict real-time latency requirements, is exactly the type of technical uncertainty the IRS credit is designed to reward.
- The workforce is the biggest driver. The $1,020,000 in qualifying wages reflects data scientists, machine learning engineers, risk analysts, and software engineers doing their normal technical work, work that generates credit eligibility hourly.
- Supply and infrastructure costs compound the benefit. The $210,000 in supply costs, covering cloud compute, transaction data licensing, and sandbox testing environments, added meaningfully to the total credit.
- Contractor costs are recoverable. The $220,000 in contractor expenses shows that outside data science consultants and specialized testing vendors can contribute significant qualifying costs, a benefit many FinTech companies leave on the table.
- FinTech fraud detection is a credit hotspot. Fraud tactics, transaction volumes, and regulatory requirements shift constantly, so high technical uncertainty is the norm rather than the exception, making R&D credit analysis especially valuable for FinTech and risk-technology companies.
Ready to Discover Your R&D Tax Credits Potential?
If your company is developing or improving products, formulas, or processes, you may be leaving significant tax credits on the table. CSSI’s engineering-based approach ensures every qualifying activity is identified, documented, and defensible, so you capture the full value of the work your team is already doing.
Request a Free Analysis today and find out what your business could qualify for.
