What is transaction data enrichment and how can you leverage it to create exceptional customer experiences? The ability to access and analyze high-quality financial data is the future of financial services. Yet transaction data currently collected by financial institutions is often ambiguous and cluttered with letters, numbers, and symbols, making it difficult for consumers to recognize their transactions and for financial institutions to fully utilize the data to improve the customer experience. As a subset of ‘Big Data,’ transactional data is broadly defined as information that records exchanges – usually of money – between people and companies. Transaction data can include purchases, payments, debits and credits, hotel reservations, interest paid and much more. Transaction Data Enrichment provides categorized and simplified transaction data that enables our customers – financial institutions and fintech innovators – to provide clean data that can be shown on different digital channels, such as websites and mobile apps. Now, many of our bank customers are using this cleansed data for digital assistants and chatbots, as well as smart speakers. This enriched data provides details needed to make conversations easier on those channels by providing simplified information. In the past, a simple purchase, such as a coffee at your favorite coffee shop might look like ##STARB180201%#***, and if your consumers are not paying attention, they might not know what that was, resulting in a call to the call center. Once that transaction is cleaned – Starbucks Coffee, Portland, OR – it's easy to identify when and where the purchase was made. Additionally, we're able to categorize it – whether it's a restaurant, online purchase, bill, utility payment, etc. – and then we identify the merchant in the description. Through that merchant, we can also find information about the geolocation, which can sometimes include a full street address, but most of the time includes a city and state. Our latest update has evolved the machine learning techniques to broad-base the structural learning for scale. The update also provides more coverage of the long tail of available merchants. In our previous version, we provided details on 1,900 of the top merchants usually found on credit card statements – McDonald's, Starbucks, Arby's – including the location of a particular store. Now we support around 2,500 top merchants with an accuracy of 97% and higher because they're well-known merchants, as determined by the frequency of occurrence in our transaction data. But that's not all. In the U.S. alone, there are over 20 million unique merchants; we're able to identify another one million merchants. We're continuing to grow that long tail as we identify more and more smaller merchants. We're using machine learning to expand these capabilities by learning the patterns and formats of the many types of financial transactions. Once our financial data intelligence platform understands that, it can determine which merchants are already in the system, and add the new ones it finds. By learning the structure of the financial transactions, we're learning how to find and identify the merchant name and geolocation on new transactions from brand new companies. Next, we identify what the transaction's category is, what it's used for, and we use customer feedback to continuously improve the system. We also proactively monitor whether we are maintaining the same level of accuracy in merchant identification and categorization. Some of it is done through automated tools, and others involve our experts looking through the data to identify and fix any anomalies. Ultimately, this can help banks reduce their fraud research costs. The reliability of fraud assessment increases because you have accurate data, which means fewer fraudulent purchases get through. It also helps to identify the correct names and transactions for a particular customer so you can more easily identify suspect purchases. Our data accuracy rates in international markets is growing as well, specifically in the United Kingdom and Australia. We will be adding support for approximately 6,000 long tail merchants and 700 top merchants in the UK; as well as 2,000 long tail merchants and 200 top merchants in Australia. Recently we hosted a webinar on Transaction Data Enrichment to help our partners better understand consumer transaction data. During this webinar, I discussed how financial service providers can use Transaction Data Enrichment to:
- Translate jumbled transaction data into clear, easy-to-read information
- Reduce customer support calls and fraud research costs
- Clarify transactions by adding simple descriptions along with merchant names, dates, and amounts
- Put financial data into context for your organization and your customers
- Leverage the data to gain insights, predict trends, and target the right customers with the right products and services
To learn how to reduce customer confusion by bringing context and clarity to transactional data and gain valuable insights that you can use to transform your marketing efforts and customer interaction, contact our sales team.
