Case Studies

Disambiguating Unclear Merchant Names

About the Financial Services Company

  • Operates one of the largest payment networks in the world
  • Facilitates transactions in 190+ countries and territories at 50+ million merchant locations
  • $1B+ cards issued worldwide


One of the largest financial services companies in the world was struggling to resolve edge cases when attempting to resolve merchant names from unclear billing descriptions. Accurate merchant data, which includes merchant names, addresses, DNS numbers and other details, are a foundational component for any credit card network. Through no fault of the financial services company, keeping this data accurate and current is a breathtaking challenge for several reasons:

  • Merchant data is constantly changing
  • Massive number of merchants (tens of millions)
  • Massive volume of transactions (hundreds of billions annually)
  • Data may contain errors
  • Formats and languages vary by country
  • Data may be obscured by a merchant aggregator (e.g., Shopify)
  • Data may be hidden behind a third-party payment solution (e.g., PayPal)

This financial services company partnered with super.AI to automate merchant name disambiguation, demonstrating how artificial intelligence (AI) can solve complex problems that would be impossible for software or humans to resolve on their own. To achieve more precise merchant disambiguation, AI was used to classify unknown billing descriptors and improve the accuracy of the merchants mapped to the company’s ever-growing and changing merchant list.

Project Highlights

  • No-code AI solution delivered 99%+ accurate merchant name disambiguation.
  • Scalable solution capable of processing up to 8.3M records hourly.
  • Rapid time to value with kickoff to first stage processing output in just 3 weeks.
  • Super.AI’s unified AI platform quickly achieved 99.7% process automation.



The financial services company selected super.AI's unified AI platform for unstructured data processing to automate merchant name disambiguation. The initial dataset included one million merchant records, which was divided into two datasets to train and refine the AI models. At a high level, the custom solution works as follows:

The company rapidly achieved 99% accurate merchant name disambiguation, including precise mapping to a single merchant even for duplicate entries—all with zero coding. Super.AI's platform scaled horizontally to process the entire dataset with capacity for 8x data volume per day. Additionally, the company was able to achieve 99.7% AI-automated processing.

Get a customized demo with your documents

Book a free consultation with our experts.

Related Case Studies

Button Text