Evaluate Your Fraud Approach With the Sift Science Analyze Page
By Swaroop Sham /
12 Sep 2016
You may already be using Sift Science Workflows to optimize fraud review, but are you giving your fraud practices a regular health check? As your business scales, you may be left pondering questions like, “Should I be blocking lower?” or “Can I review fewer users without affecting the bottom line?”
A powerful way to manage your existing fraud approach is by leveraging the analytics available on the Sift Science Analyze page. You can review your existing policies, or experiment with thresholds to find the optimum risk acceptance for your organization.
Reviewing existing fraud practices
The Analyze page provides an estimate of the percentage of events that are accepted, deemed risky, or rejected. These events can include transactions, create account, create order, create content, or add promo. You can then apply a timeline to review the performance of these events over time.
Experimenting with fraud thresholds
Additionally, by sliding the threshold boundaries, you can get a sense of how much fraud you can stop at each threshold. For example, you can see the impact to your fraud review operations based on an aggressive blocking policy.
When reviewing your company’s Analyze page, remember that the accuracy of the data represented is dependent on how much feedback you provide to Sift Science about your users. Frequent feedback equals more accurate scoring.
If you’re using Sift Science to prevent payment fraud, you can also leverage List Analytics in addition to the Analyze page to look at order value, further analyze average Sift Scores, check out labeling breakdowns, and see Sift Score distributions.
How to improve your score distribution
Here are two easy tactics that can significantly improve the quality of your analytics. First, make sure you are giving Sift Science regular feedback in the form of labeling and Decisions. Second, double check that you’re sending all the correct data. Take advantage of reserved fields along with custom fields to provide data points that are unique to your business.
We understand that as businesses scale, fraud teams require tools that scale in the growing battle against fraudsters. Sift Science’s analytics toolset of Analyze pages and List Analytics provide these teams with powerful tools to solve some of these challenges. Have questions, comments or feedback? Don’t hesitate to email email@example.com and we’ll be happy to help.
Swaroop was a Senior Product Marketing Manager at Sift and is passionate about products that keep fraudsters and cyber criminals away. In a previous life, he built security products and learned his trade at cybersecurity firms including Proofpoint, FireEye, and F5 Networks.