An AISpec Product

Your Raffle Has a Problem.
We Find It in Seconds.

Raffle Guard uses fuzzy matching across 6 dimensions and Union-Find clustering to detect duplicate entries, jigged addresses, fake emails, and coordinated abuse — so every winner is a real customer.

0+
Entries Analyzed
6
Detection Dimensions
<3s
Avg Scan Time
PDF
Export Reports
One Person. Fifty Entries. Zero Consequences.

Sneaker raffles are under attack. Abusers use simple tricks to flood your raffle with duplicates — stealing spots from real customers and damaging your brand.

WITHOUT RAFFLE GUARD
Email +tag tricks
john+raffle1@gmail.com, john+raffle2@gmail.com
Address jigging
123 Main St, 123 Main Street, 123 Main St Apt "", 123 Main St #0
Phone rotation
555-123-4567, (555) 123-4567, 5551234567
Name variations
John Smith, J. Smith, Jonh Smith, Jon Smyth
50+duplicate entries slip through
VS
WITH RAFFLE GUARD
Fuzzy email normalization catches +tag, dot tricks
Address standardization detects jigging variations
Phone number canonicalization spots rotations
Name similarity via Levenshtein + Soundex matching
Cross-field correlation links entries by behavior
Union-Find clustering groups all related entities
1verified entry per real person
Four Steps. Clean Raffle.

Upload, analyze, review, export. The entire pipeline runs in seconds.

Upload CSV
Analyze
Review Flagged
Export Clean
Six Dimensions. Zero Blind Spots.

Every entry is analyzed across six independent detection layers. Even one match is enough to flag — but the real power is when multiple dimensions converge.

Email Trick Detection

Catches +tag aliases, dot insertion (j.o.h.n@gmail), domain swaps, and disposable email providers. Normalizes to a canonical form before comparison.

FUZZY MATCH

Address Jigging Detection

Standardizes street types (St/Street/Str), catches apt/unit tricks, empty unit numbers (Apt ""), PO box variations, and transposed digits.

STANDARDIZE

Phone Rotation Detection

Strips formatting, validates length, canonicalizes country codes. Catches number rotation patterns and Google Voice/VoIP indicators.

CANONICALIZE

Name Similarity Matching

Levenshtein distance + Soundex phonetic matching. Catches typos, abbreviations (Robert/Rob/Bob), and intentional misspellings.

PHONETIC

IG Handle & Cross-Field

Same Instagram handle across entries is an instant flag. Cross-field correlation detects when different fields independently point to the same person.

CORRELATION

Entity Clustering

Union-Find algorithm groups all related entries into clusters. If entry A matches B on email, and B matches C on address, all three are linked as one entity.

UNION-FIND
Live Analysis Preview

Watch Raffle Guard scan entries in real time. Click “Run Scan” to see the detection engine in action.

raffle-guard --analyze drop_entries.csv
$Click “Run Scan” to analyze 12 sample entries...
Clean: 0
Suspicious: 0
Flagged: 0
0/12 processed
Simple Pricing. Serious Protection.

Choose the plan that fits your drop frequency. Cancel anytime.

Per Drop

Occasional Raffles
$99/drop
per raffle drop
  • Up to 5,000 entries
  • 6-dimension analysis
  • Color-coded risk scores
  • CSV export (cleaned list)
  • PDF summary report
  • Email support
Choose Plan

Monthly

Regular Drops
$499/month
per month
  • Unlimited drops
  • Up to 15,000 entries / drop
  • All 6 detection dimensions
  • Live SSE monitoring terminal
  • Detailed PDF cluster reports
  • Priority analysis queue
  • Slack / Discord webhook alerts
  • Dedicated support channel
Get Started

Enterprise

Multi-Location Boutiques
$799/month
per month
  • Everything in Monthly
  • Unlimited entries per drop
  • API access for integration
  • Custom detection rules
  • Historical fraud database
  • Multi-store dashboards
  • White-label reports
  • Onboarding + training call
Choose Plan
The Cost of Doing Nothing

Drag the sliders to see how much fraud costs your operation per drop.

Your Raffle Numbers

Revenue at Risk Per Drop

50
Fraudulent Entries
=
$10,000
Lost Revenue
Raffle Guard costs $99/drop. You protect $10,000 in revenue. That's a 101x return.
That's 50stolen spots from real customers — customers who wanted to buy, who follow your brand, who waited in line.
Frequently Asked Questions

Everything you need to know before cleaning your first raffle.

What file formats do you support?+
How accurate is the detection?+
Can it handle 10,000+ entries?+
Do you support Google Forms exports?+
What does the PDF report include?+
How is this different from manually checking?+

Stop Letting Abusers Win Your Raffles

Every raffle you run without Raffle Guard is giving away spots to people gaming the system. Your real customers deserve better.

Get Raffle Guard Now