BrightCheck’s Great Wall of Fraud Prevention

Data Dive: November 2024

What We Did and Why It Matters

We rolled up our sleeves, donned our fraud-fighting capes, and got to work. This study dives into how BrightCheck’s systems performed when tasked with keeping out the bad guys (and gals). We looked at 544 user records for 2 weeks, following their journey through our fraud-busting machine.

Here’s how it worked:

1. Automated Waterfall Analysis: We analyzed appeal data from users who cried, “I’m innocent!” Spoiler: Many weren’t. The only data requested was First Name and Mobile Phone
2. Enhance authentication: By requesting additional user information, including First Name, Last Name, City, State, and Year of Birth.
3. Human Backup: Our Risk Mitigation team went full detective mode on Level 2 step-up authentication failures, digging into external data to uncover the truth.

The goal? Spot patterns, tighten the screws on fraudsters, and make your platform safer than Fort Knox.

Waterfall Process: The Tale of Triumph and Rejection

Level 1: Enrollment

• Users Who Knocked on Our Door: 544
• Approved Without Drama: 294 (54.04%)
• Failed Miserably: 250 (45.96%)

Level 2: Second Chances

For the 250 users who didn’t pass Level 1, we extended an olive branch—a chance to appeal.

• Second Attempt Pool: 250 (100%)
• Appeals Submitted: 43 (17.2%)
• Deferred for Later Consideration: 82 (32.8%)
• Approved After Appeals: 94 (37.6%)
• Failed Again: 64 (25.6%)

Final Scorecard

• Total Approved (All Levels): 388 (71.3%)
• Lift from Appeals (Delta): 94 (17.3%)

Colorful Characters of Fraud: Key Findings & Insights

• Sex Offenders: One identified—already one too many.
• Criminal Backgrounds: Four applicants with records tried their luck. All denied.
• Double First Names: 28 users submitted names like "John Michael." A scammer staple, not the U.S. Open.
• Name-Email Mismatches: 188 cases, such as "Dylan Annabel" using "[email protected]."
• Generic Locations: 34 applicants claimed small towns like "California City, CA" (population 14,000). Highly unlikely fraud hubs.
• Landline Numbers: 6 listed landlines—a suspicious relic.

Scammer Tactics and Lessons Learned

• Double First Names: A common scammer tactic. We now provide an explainer video to clarify proper First/Last Name formatting.
• Fake Geography: Locations like "Illinois City, IL" don’t fool fraud filters.
• VOIP Numbers: 14 Level 2 failures used VOIP lines, coupled with mismatched names and high-risk scores.
• Fake Emails: Cases like "Dylan Annabel" highlight the importance of email/name consistency.
• Landlines: Rarely used, often signaling fraud.

By spotting patterns and addressing vulnerabilities, we’re turning fraud into an increasingly uphill battle.

User Feedback: The Good, the Bad, and the Fake

We scoured the Apple and Android app stores and spotted a recurring theme in 1-star reviews: fake profiles and escorts. These gripes prove that our fight against fraud is as critical as ever.

Recommendations: How to Build an Even Better Fraud Fortress

1. Ask for the Legal First Name and Mobile Number: It works for rideshare giants; it’ll work for you too.
2. Sniff Out VOIP Numbers: Implement VOIP detection tools to flag these sneaky fraudsters before they enter the system.
3. Pre-Check with 2FA: Add two-factor authentication early in the process. Fraudsters hate it; we love it.
4. Geography 101: Validate city and state entries. A little logic goes a long way (e.g., LA becomes Los Angeles).
5. Sex Offenders? Bye. Immediate removal. No second chances.

A Final Word

Fraudsters may be clever, but BrightCheck shines brighter, illuminating the path to a safer platform. With every step, we refine our processes to outsmart bad actors and protect what matters most—your users and your brand. Have questions or need support? We’re here to light the way.

Warm Regards,

Scott Bright

Founder / CEO

BrigthCheck

Get 100 free API calls with our ScamDetect API today: BrightCheckpro.com