MAGGIE PAGANO: There are lessons from my being robbed, including having faith in British innovation and to praise the banks for tracking down criminals
- Until recently, most fraud losses came from stolen payment cards
- But lockdowns have stopped criminals getting the cards
- The focus has shifted to what’s called authorised push payment fraud
- Cracking whether a real-time transaction is genuine or not is tricky
- UK’s Featurespace created adaptive behavioural analytics to solve the problem
A few days ago, I shopped at my local Waitrose, leaving at around 5pm. Walking to my car in the parking area, I was approached by a bearded man with gleaming teeth, who said someone had just reversed into my bumper.
He bent down to show me where the driver had hit my car, urging me to look.
There was nothing to see, other than a mark where he had wiped away the dirt. He said I was lucky the other vehicle hadn’t done any damage, and after a quick chat, a young woman appeared, and they were off.
Under lock and key: Until recently, most fraud losses came from stolen payment cards – but lockdowns have stopped criminals getting the cards
Back home, an hour or so later, a flurry of texts from Barclays asked me to confirm purchases on my card. They were not mine. I went to get my card from my wallet to ring the Barclays helpline.
But all my cards – debit, credit and business – had vanished. Soon more texts came from Barclaycard, asking if I was having PIN problems taking money from an ATM? Something clicked. The man who stopped me must have been a conman. Whilst he distracted me, the woman had taken my wallet. Like a magician, she unzipped it, took the cards, and returned it to my handbag. What a fool I felt.
The bank called, asking me to confirm more transactions; one for £999.00 at a local shop, four transactions at the local Tesco ATM for £250.00 each – timed at 5.12pm – and several small Barclaycard purchases at Sainsbury’s in a nearby town. None were mine so Barclays blocked my cards.
What brilliant detective work by the bank. Not only did they stop these purchases, but overnight the £1,000 taken from the ATM was refunded. Other repayments are due.
Most banks use a cocktail of solutions to catch out criminals but the pandemic has seen fraud soar and change. In the first half of 2021, £754m was stolen through fraud but the security systems I saw stopped another £736m disappearing. Until recently, most fraud losses came from stolen payment cards. But lockdowns have stopped criminals getting the cards.
The focus has shifted to what’s called authorised push payment fraud, where the customer is tricked into sanctioning a payment to a criminal’s account.
Cracking whether a real-time transaction is genuine or not is tricky. It’s a problem that has been addressed by Featurespace, the company which created what’s known as adaptive behavioural analytics.
This alerts within seconds if a customer’s behaviour doesn’t fit patterns: it’s how Barclays knew it wasn’t me at the Tesco ATM. Featurespace was born out of the engineering department of Cambridge University. It is based on the work of the late Professor Bill Fitzgerald, and co-founder Dr David Excell, who were approached by Betfair to stamp out fraud in its real-time betting operations.
The result was Adaptive Real-Time Individual Change (ARICs), a behavioural analytics engine aimed at flagging up anomalies as they occur. This was in 2008, when betting companies were the only ones dealing with real-time money flows.
Now most UK financial institutions have caught up. They too deal with real-time money transfers and use the machine-learning processes developed by Featurespace. It is a private company, though one of its biggest investors is listed firm IP Group, which supports university spin-outs.
There are many lessons from my being robbed. Most obviously, to ignore anyone telling you your bumper has been dented, but also to have faith in the brilliance of British innovation and, for once, to praise the banks for doing so much to track down the criminals. Credit where it’s due.