Non-card payment methods in Southeast Asia
From education technology, to fighting fraud for the likes of Facebook, Netflix, and Uber, Brian’s career so far has been quite a journey.
His first forays into the world of fraud were at Google, where he started off working with advertisers whose accounts had been blocked by suspected fraud.
This gave him first-hand insights into the pain caused by every single false positive. And he was hooked.
Americano (milk, no sugar), a cold brew when the mood catches me, or a dirty chai latte – if I’m feeling crazy…
Data breaches, like Target’s high-profile attack, are becoming the norm. And fraudsters are getting their hands on huge amounts of ‘clean’ shopper data. They are also getting better at separating the act of validating the data, from the act of using the card to make fraudulent purchases – something that used to be addressed by velocity checks. And so retailers relying on traditional defense techniques are suffering.
We are also seeing a trend towards targeting specific card BIN ranges. The moment a bank inadvertently lets its guard down, the entire fraudster community knows about it. And we have seen a trickle of fraud scale into a full-blown attack within a few hours. When this happens, businesses relying on manual reviews have a hard time.
It is crucial to focus on systems, rules and infrastructure that can quickly identify and respond to emerging trends, without the need for manual intervention.
There are many ways to do this. At Adyen we have focused on creating a real-time transaction relationship graph, onto which we add alerts, rules and other logics. Machine-learning algorithms are also employed to identify and react to changes quickly.
The biggest ‘bright spot’ we see for retailers is the ability to open up the ‘black box’ of generic card declines. With machine-learning logic, we give retailers insights into the reason behind declines, so they can respond in real-time, and turn things around.
For example: In some markets 3D Secure is mandatory, and all transactions without 3D Secure will be declined. With this information, the retailer can retry the transaction with 3D Secure, saving what otherwise would have been a lost conversion.
One word: omnichannel. The most exciting thing I’ve been working on lately is the ability to leverage multichannel data to learn more about shopper behavior. With this we can help retailers see which shoppers are visiting their stores, which are visiting their ecommerce site, and how to design customer journeys around their specific preferences.
An example: One recent study we did with a retailer showed they had blocked several hundred online transactions of shoppers that had spent over $1,000 in store. Now they have access to cross-channel data, they can roll out the red carpet for loyal in-store customers online too.
Two years ago, when I had a meeting with a retailer’s online and in-store teams, it would often be the first time those teams met. But now retailers are consolidating their teams.
Amazon’s recent announcement of AmazonGo (its cashier-free convenience store) is a great example of how payments in store are beginning to resemble one-click, or zero-click, payments online.
This might be way down the line for many retailers. But the sooner they start capturing and securely storing payment details, using tokenization, the easier it will be to support this kind of experience in the future.
Our focus for 2017 is to further develop our machine-learning capabilities, and create a data science infrastructure able to cope with billions of dollars worth of transactions easily and securely.
The ability to freeze time – mostly to catch a nap every once in a while.
By submitting this form, you acknowledge that you have reviewed the terms of our Privacy Statement and consent to the use of data in accordance therewith.