American Express “Amex” charges the merchants almost 50% more than their competitors. How can it be that they are still so successful? Why do they charge so much and why do retailers agree to pay that?
It is very simple; Amex is the only credit card company that also owns a bank, which means they have a closed loop of data. Amex, with its Business Insights Consulting Division, is turning this data into a valuable source for the retailers they work with and for their customers, and all this value is then translated into a revenue stream.
How does this work?
Amex collects data from the 6 billion transactions processed every year. This data includes not only the fact that a transaction was made, but also the amount of the transaction and the exact vendor. This information is unique to Amex, and is not available for other credit companies who are only a chain in the loop of payment. Amex is the only one that sees the data both on the consumer side and on the merchant side of the business.
With the data, Amex is able to analyze trends and information on cardholder spending and build algorithms to provide customized offers.
For customers: AmEx offers rewards to incentivize higher spending and also offers value added services such as merchant financing to select merchants, providing an attractively priced source of financing. They are able to attract and retain customers to these merchants and to increase the amount of spending.
For merchants: The company is also able to create targeted marketing to match merchants with the right customers, who are more likely to spend more and stay loyal. AmEx has modeled audience segments for use in online ad targeting. The credit-card firm compiles audience segments based on its cardholders’ transactions, pegging people as likely to shop at a sporting goods store or specialty retailer, for example.
In 2010, Amex decided to further leverage their data by using big-data tools and machine learning, to improve the revenue streams the understanding of customer behavior and the firm’s decision making.
They applied machine learning to all parts of the business, more specifically in these use cases:
- Fraud detection and prevention – the goal is to stop fraud activity before there is a substantial loss, while allowing “business as usual”, hence – not slowing down the transaction speed.
- New customer acquisition
- Improving customer experience – by creating push recommendations for customers based on their past spending profile.
This contrast between American Express and its competitors is evident in the numbers; Visa has more than 2 billion cards in use worldwide and processes more than 60 billion transactions per year, while AmEx has just 107 million cards in force and processes just 6 billion transactions per year. Despite this disparity, American Express has annual gross revenues of $33 billion while Visa earns just $14 billion per year. For sure, it is amazing that American Express, a company that is operating since 1850, is continuously innovating itself into the data era.