Find out more at btsub.sg/promo
You are here
Staying a step ahead of fraudsters
IF I were to ask you which are the top 10 organisations globally which spend the most on research and development, you would probably think of technology giants such as Google, Apple and Amazon. However, you would be wrong. It's not the technology giants, but instead, global criminal fraud organisations. They are well-resourced, more organised and their staff are more highly trained than you can imagine; and they spend an enormous amount of money on research and development to always stay one step ahead of the game.
It is imperative for the rest of the legitimate business world to employ innovative combinations of data to authenticate transactions made by consumers and businesses in trying to combat fraud. For example, Amazon uses FraudNet to reduce the number of transactions that need to be referred to manual review. They now only refer 2 per cent of all transactions to a call centre. Compare this with Singapore, where credit can be refused due to something as simple as not providing the correct address.
When this recently happened to me, as my credit card has a Brazilian address, I didn't get an answer until 24 hours later. It was via email and I had to call a US number to verify that the credit card was mine. The questions asked were of the same ilk as "what is your mother's maiden name?", which have been the standard identity verification questions used since more than a decade ago!
Imagine that same experience if the e-retailer was able to access a combination of real-world data and randomised verification questions that would give a more specific perspective of my identity, and more importantly my behaviour. For example, they might see that my last check-in on Facebook was in downtown Singapore. They could send a question online to me via Facebook to ask me to verify my last three check-ins. A good question is to verify the colour of the car I am driving, something a fraudster would not know.
Another real example is what we are deploying with some customers in Indonesia. We have access to GPS locations via consumers' smartphones. I can verify a consumer's address with a database of all the GPS tracking from his mobile phone. If the phone remains at that address for 50 per cent of the time, it is likely to be the consumer's home address.
Analytics can help to validate real information (more than protect) and perform real-time authentication. Using data that is available in the markets, aggregating the information and coming up with real time questions that will authenticate customers in an easy and secure way will improve customer experience.
Take this example: a customer decides to shop online based on an emotional drive; he wants to conclude the transaction quickly. Normally, he would choose the e-commerce platform that has the best offer. If a transaction is denied due to a false-positive flag from not having a secure authentication, this will affect both the business (lost sale) and the customer (poor experience). By having access to different kinds of data, real-time authentication can come into play and certify in a secure environment that the data provided belongs to the person doing the transaction.
One other way is device information. Being able to understand that the device has a good reputation also makes the transaction more secure. Customers will soon demand secure environments to transact on online sites.
The number of ways to use combinations of data to authenticate yourself is infinite. However, the best combination of data will depend on the transaction. While many organisations currently use seemingly random questions, they are typically not smart questions. They will usually be the same, or similar sets of pre-defined questions. In my view, this is the big mind shift that organisations are not yet making. There is now an opportunity to take advantage of Big Data sources and combining that will randomise questions associating that data with something the consumer has or knows - for instance, the car colour example I used earlier.
It's also imperative for small and medium enterprises to use data to understand where they can target efforts in generating growth. For example, how can SMEs use analytics to protect data and prevent fraud while enhancing customer experience?
Analytics are available not only to large institutions; small businesses can also take advantage of data and analytics. Understanding the behaviours of their customers will also help them to grow their business and achieve efficiency. Imagine if an ice-cream shop is able to find out which flavours are the best sellers in each season. It would be able to achieve efficiency by producing more of those popular flavours during those specific parts of the year.
Most small businesses think the world of analytics is far beyond their business, but things are becoming more accessible. Experian has helped small businesses across the globe to boost their growth by driving new customers to their businesses and also achieving more efficient use of capital investment by using all data formats - across structured and non-structured - voice, text and video. Experian's strength lies in the interpretation of data available in different markets and transforming them into actions that enhance customer experience, customer acquisition, customer management, collections and fraud prevention across different businesses in all industries.
Finally, Experian is also using data to enhance financial inclusion across the Asia-Pacific. For example, banks need help to bypass the traditional structured data types that typically form the basis for credit risk profiles - namely transactional and demographic information that pinpoints how an individual behaves, whether he pays bills on time, his assets and liabilities. The number of newcomers to credit is increasing substantially across Asia-Pacific. Understanding these new consumers is a challenge that emerging markets face every day.
As this market has no credit history with traditional credit bureaus, it's difficult for banks to provide financial access without knowing anything about them. On the other hand, they do have mobile phones, utilities services and are active on social media, aside from other transactions that they do with retailers.
For a bank it is crucial to have access to these sources of data to predict the level of risk by understanding the behaviour and potential credit rating of these newcomers. Understanding payments and consumption behaviours has led to the development of analytic models that provide a better level of understanding of these new markets for traditional banks.
Advanced analytics using Big Data technology, combining and aggregating all data available is the way to understand these emerging consumers who seek access to credit.
In other markets such as Latin America, Experian is already collaborating with utilities companies to build this data. Even if their customers are not participating in the financial system, by understanding where they live, their consumption of electricity and water, and overlaying this information with payment profiles, we have developed an index that provides a very accurate risk profile.
- The writer is managing director of Decision Analytics in Asia-Pacific, Experian.