FraudGuardian is a tool we developed to help you screen incoming orders to help reduce the costs of fraud. Contrary to the name, FraudGuardian does not screen for fraud. Instead, it looks for orders that are probably not fraud and automatically approves them, helping to automate your business one step further. Without FraudGuardian, all incoming orders are set to pending so that you can inspect them yourself before approving them and setting the client status to active. For a larger discussion of fraud in the hosting industry, please read this.
You can sign up for a FraudGuardian account here. Once you've done that, you just need to enable the module, then configure it. There are two main parts to configuring FraudGuardian: Fraud Sets and Thresholds.
Thresholds are your way of telling the system what variables you do and don't want, and how badly you do and don't want them. For example, you may decide that free email is something you don't mind, and assign it zero points. But it may be that you don't want anyone coming through a proxy, and so assign this ten points. You also set how many points a client can have, and what behavior the system will take depending on the points total assigned to each incoming client. These options are discussed in greater detail in the field explanations below.
You can have a single threshold, where all new clients have the same standards applied to them, or create various different thresholds depending on countries or regions. To create a new threshold, enter a name for the threshold in the field and choose to apply to ALL countries, ONE country, or a REGION of coutries. Then click "define new country threshold." You will see this threshold appear as part of a list under "Current Country Thresholds." To set your parameters for this threshold, click "Configure."
This section allows you to set the system behavior based on a client's final score.
One of the common tactics employed by fraudsters is to charge as much as possible once they find a card that is live. It might be possible that the client's billing account and address information matches well enough to not stand out as being fraud, but the amount they are purchasing is. For example, if a fraudster steals a card from someone in the same town, the system may not be able to tell that this is not who they say they are. But if the fraudster is trying to charge $500 worth of hosting, that might tip you off that something is amiss (depending on how much you charge for hosting). Like the section above, you can set three differetn set of actions based on the amout being charged. For example:
ModernBill uses what it calls Fraud Socre Triggers to assign points to incoming client accounts in order to determine whether or not the order is likely to be legitimate. The system has these triggers set to default values upon installation, but you are free to adjust them as you like. For example, you may decide that you don't care if the client's IP matches their country for billing purposes. In that case you can set this value to 0 instead of the default: 2.5.
This is the message that will be displayed to people who trigger your soft rejection notice for this threshold.
This is the message that will be displayed to people who trigger your hard rejection notice for this threshold.
A fraud set is a group of thresholds. The fraud set allows you to tie various thresholds with order forms, and specify the email you wish to have notices sent to. If you are selling in one currency, you will like just need one fraud set, as you will be using a single order form. If, however, you are using multiple order forms because you are selling in multiple currencies, you may want to create a unique fraud set for each order form. It is also possible to have a single threshold apply to several order forms, via the fraud set.