3 Ways that Big Data is changing fraud prevention.

Posted by Pat Phelan on April 9, 2014

Few things are likely to have a greater influence on the future of the fraud prevention industry that the emergence of ‘Big Data’ as a core part of up and coming startups in the industry.

What is ‘Big Data’? It's nothing more than a popular term used to describe the exponential growth and availability of data these days and the fact that technology has finally caught up with the ability to store, manage and analyse this data in a time frame that’s incredibly useful for decision making in business.

Imagine a simple spreadsheet full of rows, columns and numbers - most of us can manage reading and understanding them until they get to a certain size at which point they become incredibly difficult to work with. That’s been the problem industry has faced for years, as the volume of data you want to store and analyse got larger and larger it became harder and harder to work with. 'Big Data’ uses new ways of technically storing data, new ways of connecting up the information and new ways on how it's accessed and analysed that removes much of the tedious, mind numbing aspects of the work.

These innovations in how you work with and manage huge volumes of data are starting to have a significant impact on the fraud prevention industry. Here’s a few reasons why;

1. Dynamic sources of data

Traditional identity screening was pretty old school in terms of the data sources that it used as part of the anti-fraud process. Name, address, phone numbers, date of birth were all key as they were easy to access and pretty easy to confirm. The only problem being of course that they’re also very easy to fake, or even get wrong. Take something as simple as an address: a lot of folk these days live in rented accommodation, or still use the family home for ‘official’ mail, the days of one or two addresses across your entire life are well behind us and therefore basing an entire identity check on validating address is setting yourself up to fail. You hit similar problems with phone numbers, in a world of mobiles and pre-pay phones, how long would it take you to get a brand new phone number that wouldn’t be linked to your existing identity?

As well as being able to take a variety of data into account when making a decision, a proper system, like Trustev, can now take live up to date information into account, instead of relying on cached or stored information from datasets. This means that identity decisions are made based on the best possible information and not on facts and figures that might be out of date and not relevant.

2. Speed

Take ecommerce for example - an online retailer is looking for an instant identity decision - a customer is standing there offering them money and they need to know whether its safe to accept it or not. In the past the length of time that would pass in accessing and processing information meant that you had to limit the amount of data you processed dramatically so as to limit the ‘wait’ time.

The technology and techniques behind ‘Big Data’ eliminate these wait times. Huge volumes of data can be processed and analysed in microseconds, delivering real time results that businesses can instantly act on.

3. Complexity

Most importantly, today's data decisions are far more complex than businesses previously wanted to handle. The amount of information they want to cross reference is well beyond the ability of older systems to handle. What was a dozen or so checks of individual pieces of data has turned into extraordinarily complex decision trees using algorithms, cross referencing multiple instances of data, double and triple checking results - and its all needed faster than ever before.

At Trustev, we believe that identity is crucial for the future of the internet economy. We built our technology from scratch, from the ground up to harness this new Big Data world. We use new database technologies like HADOOP to give speed, decision accuracy and incredible flexibility to our platform and to how our data and technical teams access and model data to constantly improve our systems ability to detect fraud..

We work closely with one of the leaders in the analysis space, Datameer, a company founded by one of the ‘father figures’ of the HADOOP community. In fact you can see our CTO Chris Kennedy and Head of Fraud and Data Strategy Diarmuid Thoma, talk about it here.


If protecting your revenues from fraud and knowing who you do business with is a concern for your business, drop us a line at sales@trustev.com

Topics: Technology, fraud, Identity