How a Fraudster Acts
How a fraudster acts? Besides the specifics details of what a fraudster is trying to purchase and the details they have given it is also worth considering what a fraudster does before they begin a transaction. Being able to capture behavioural data may assist you in determining if a booking is fraudulent.
If we take the example of an online retailer then we can consider behaviour in the following ways:
How has the customer arrived at your site?
There are many ways a customer can arrive at your site namely:
– Direct traffic
– Search engine
– Mailing list
What needs to be considered is how a fraudster in likely to act? It is common sense to assume that a fraudster is unlikely to be booking a holiday promoted on a newsletter, but there is a chance that a fraudster trying to make themselves look honest may have signed themselves up to your newsletter.
In the case of advertising you would feel that a fraudster would be attempting to access your site directly rather than clicking on a banner, so you would presume that advertising would be the least likely route of a fraudster.
A user who has bookmarked your site could be a loyal customer or could be a fraudster who wants quick access to your site.
A fraudster who has reached you via a search engine may suggest that they are not directly targeting your site, but having knowledge of the search term used might be useful. Are you more likely to be a fraudster if you have entered the exact name of a website in to the search engine or phrase cheap-flights?
Most online retailers measure how users get to their site purely because they want to know what parts of their marketing are performing the best. Don`t be afraid to consider how a fraudster is getting to your site and see how this compares to the norm.
What searches have then done?
Depending on the type of fraudster using your site you will see different patterns in how they use your site. If you have data on how a standard customer uses your site the activity of a fraudster is sure to highlight suspicion.
If you are dealing with an organised fraudster who has orders to fulfil you will be dealing with someone who is unlikely to be searching for a range of options in terms of destination, departure point and departure date. The fraudster rather than looking for recommendations or the latest deals is likely to hit the search box straight away and type very specific details not making use of the +/- options.
If you are dealing with an individual fraudster then they are more likely to look like a normal customer so the use of a search box may include looking at a variety of options. In this case it is worth noting how the user is entering data onto your booking engine. Fraudsters are likely to have documents or spread sheets stating credit card details and postal addresses etc, the cut and pasting of this data into your booking engine highlights suspicion so is worth measuring.
Can measuring the behaviour of users be helpful in the fight against fraud?
While the behaviour of a fraudster is unlikely to give you the definitive answer to the question is this customer a fraudster having some information of the behaviour of a user may assist you in closing down a loop hole and may be the extra piece of information you are looking for to reject a booking.
Behavioural data is best used in conjunction with other fraud prevention techniques, it may well come low down in any rule base used but if this data is available to you, use behavioral data to your advantage.