The challenge: matching transaction strings to merchants
Fable Data approached the Local Data Company as they required high-quality, robust and legitimate data to validate merchant information against transaction data which can often be difficult code to crack. Merchant transaction strings (the information on each transaction on bank or card statements) can be a series of letters and numbers and can vary widely across each brand or vendor.
Examples of transaction data strings for a Premier convenience store (source Fabledata.com)
In order to match transactions to vendors, Fable Data required comprehensive data on all the merchants trading in the UK including lists of stores for brands. Avoiding web-scraped data was important in order to protect the integrity, reliability and accuracy of their insight product. Fable Data uses Local Data Company records to match transaction strings against known merchants to build a what is known as a ‘training set’ which helps analysts identify a list of transaction strings which are a true match for a vendor, including all sites across a store estate. This work also helps Fable to determine whether spending was within retail stores, cashback services, at ATMs or online and helps them to reduce the prevalence of ‘false positives’ for transactions.
Matching against Local Data Company data on 5,000 multiple brands, Fable Data has been able to match transaction strings to vendors and organise volumes of spending into retail categories such as grocery, clothing and home & garden, boosting the insight available from the transaction data. This data forms a critical part of Fable Data’s offer and provides relevant, impactful and timely evidence for hedge funds making investment decisions.
COVID MARKET RECOVERY TRACKING
During the COVID-19 pandemic, Fable Data was able to provide real-time evidence on the recovery of the consumer economy in the UK, using the powerful combination of Local Data Company data and transaction data. They measured the daily, weekly and monthly recovery of different corners of the market to evidence the real-time trends as they happened. The data was so strong it was featured in the Financial Times eight times within a two-month period as an indication of the speed of recovery following lockdown.