In this special guest feature, Kevin Lee, a Trust and Safety architect at Sift, discusses how Digital Trust & Safety, powered by machine learning, is taking the lead with fraud prevention efforts. Kevin helps customers implement strategies that cross-functionally align risk and revenue programs. He has lead various risk, chargeback, spam/scams, and trust and safety organizations at Facebook, Square and Google.
Businesses must reckon with a new reality: some of today’s top digital innovators are fraudsters. Online fraud isn’t as simple as using a stolen credit card to make online purchases anymore. Fraudsters learn quickly and share information freely; after all, they don’t have to worry about GDPR regulations like the rest of us. Their high-tech and sophisticated toolbox now includes machine learning algorithms, bots, and other instruments used to commit fraud at scale.
At the same time, customers demand excellent experiences. When checkout is bogged down in CAPTCHAs and security questions, they’ll abandon their cart. When their order doesn’t arrive on time, they’ll voice their displeasure on social media.
In this new digital world, fighting fraud and delighting customers might seem mutually exclusive. Businesses tend to err on the side of fighting fraud, to mitigate the risks that fraudulent behavior will hit their bottom line. The legacy approach to fraud-fighting involves static systems like rules engines. Rules engines are rigid and hard-coded. Once implemented, rules treat fraud as though it is black and white: a transaction either breaks a rule or not, and there no context, history, or gray area. Fraud analysts who use rules-based systems examine suspicious actions as discrete events.
Many businesses have realized that rules-based systems are inadequate: they don’t scale, they’re infexible, and they’re easily reverse engineered. Those businesses have turned to machine learning, which adapts to fraudulent behavior as it happens. This is especially crucial in an era when entirely new types of fraud arise almost every day: account takeover attacks, fake content, bots, and more. Rules can only apply to a single vector of fraud, whereas machine learning is successful across multiple channels and types of fraud. Equally important: machine learning grows with your business. These algorithms are capable of leveraging massive data sets to fight fraud — and the more data they can access, the more accurate the output will be.
But by focusing solely on fraud prevention, businesses are missing a key part of the picture — even businesses that use machine learning. They continue to treat fighting fraud and delighting customers as though these goals are at odds with each other. Businesses roll out a new product or app first and worry about fraud prevention later. That means risk teams are often left picking up the pieces after an attack has already occurred and everyone has heard about it on social media. To make the situation even more dire, fraudsters have started to adopt the same advanced technologies that fraud-fighters are using, such as machine learning and bots. Fighting fire with fire is more important than ever. Companies that fail to do so — that focus on fraud mitigation rather than prevention — get left behind.
The answer is Digital Trust & Safety. The term “Trust & Safety” was coined by disruptive, digital-first companies that recognized trust is woven into their DNA. Google, Airbnb, Patreon, Uber, Twitter, and other top innovators have reframed their business model to include Digital Trust & Safety: a holistic framework for approaching your business’s mindset, processes, and technologies. Legacy methods drive a wedge between revenue and risk; Digital Trust & Safety builds a bridge between them.
Digital Trust & Safety is the only method that combines risk and revenue decisions, which are underpinned by machine learning technology. Instead of creating a product and slapping on a fraud prevention band-aid later, Digital Trust & Safety invites fraud teams to act as stakeholders throughout the entire product life-cycle. Companies that operate with a Digital Trust & Safety mindset proactively build a strategy and technology stack that that can stop fraud before it happens. They leverage machine learning to customize checkout experiences for each user, introducing varying amounts of friction for each user depending on the risk they pose to the business.
The bottom line: legacy methods hamper innovation. Digital Trust & Safety is the only scalable and effective approach for the digital era. Companies that adopt Digital Trust & Safety are empowered to join the ranks of leading companies building the most groundbreaking products of our time.
Sign up for the free insideBIGDATA newsletter.
It will be very helpful.