Reduce chargebacks with big data analytics

Thanks to trailblazers like Amazon, eBay and, consumers now make 51% of their purchases online (, 2016). While this massive increase in online purchases makes the buying cycle more convenient for consumers and businesses, card-not-present transactions also carry risks, raising the question of how can we reduce chargebacks without compromising the user journey?

The absence of in-person verification, such as checking official identity documents or requiring the card user’s PIN, means that the risk of chargebacks is much higher for card-not-present transactions. Chargebacks occur as a result of the consumer disputing a purchase made on their card which can be for reasons including duplicate billing, being unhappy with their purchase, fraudulent use of their card, or even just forgetting or not recognising the transaction on their bank statement. Chargebacks can take as much as 90 days to be reported, leading to unexpected losses for payment service providers, issuing banks, and merchants.

With that in mind, payment service providers can massively reduce the risk of these losses for all parties involved by directly addressing one of the reasons behind chargebacks; fraudulent card use.

Online fraud is estimated to cost merchants between 0.3% and 3% of revenue. When processing transactions online, it is just as important to verify that the purchaser is who they claim to be – but in the absence of chip-and-pin, how can we ensure this while also minimising the friction in the transaction process?

While 3D Secure is already used in practice by a number of (mostly smaller) merchants, it adds friction to the user journey - particularly when used on a mobile device - which may result in drop-offs and lost sales. An alternative option is remove 3D Secure completely, however this would leave you open to the risks discussed above.  By implementing social connect instead of 3D Secure, you can know your customer better and mitigate the risks associated with card-not-present transactions by leveraging the card user’s digital footprint to accurately verify their identity. Big data analytics tools such as PROFILE ID utilise various advanced analytics techniques including Bayesian belief networks, psycholinguistics, natural language processing, and big data analytics to verify the user’s identity based on multiple aspects while leveraging the familiarity of social connect.

The insights made available from this process can help: 

1) The merchant to confidently accept more transactions

2) Reduce fraudulent transactions

3) Reduce chargebacks


·       Reduce friction

·       Protect card user against unauthorised use

·       Leverage familiarity & convenience of social connect


Author Ahmed Amin