Fraud Prevention

Sponsoric uses multiple layers of detection to ensure advertisers only pay for genuine engagement and publishers earn from real traffic.

How it works

Every impression and click passes through a real-time detection pipeline before it counts. Events that fail any check are discarded — they don't affect advertiser spend or publisher earnings.

Detection runs at the edge for zero latency impact on publisher sites.

What we check

Bot filtering

Automated traffic from crawlers, scrapers, and known bot frameworks is identified and rejected. We maintain a continuously updated set of detection patterns that cover common automation tools and headless browsers.

Rate limiting

Click and impression events are rate-limited per source using distributed infrastructure. Limits are enforced globally across our network, not per-server, so they persist across deployments and can't be bypassed by hitting different endpoints.

Device validation

Events include a non-identifying device fingerprint that must meet minimum entropy requirements. This helps distinguish real browsers and clients from simple HTTP scripts.

Referrer analysis

We validate that traffic originates from legitimate publisher domains. Suspicious or malformed referrers are flagged and filtered.

Behavioral signals

Patterns like rapid-fire clicks, unusual timing distributions, and repeated identical fingerprints are detected and throttled.

Payment verification

  • USDC payments are verified on-chain by reading transaction receipts and decoding transfer events. Each transaction hash can only be used once (replay protection).
  • Stripe payments are verified via webhook signature validation.

For advertisers

  • You only pay for clicks that pass all validation checks
  • Fraudulent clicks are filtered before they count toward your spend
  • Budget pacing spreads your spend evenly, so a burst of suspicious activity can't drain your budget
  • Your dashboard shows only validated metrics

For publishers

  • Earnings are credited only for validated events
  • Sites are screened by AI for content safety before admission
  • Publisher accounts that generate suspicious traffic patterns may be flagged for review
  • Earnings are tracked by funding source and paid through matching channels

API rate limits

All Sponsoric API endpoints are rate-limited to prevent abuse. Limits vary by endpoint based on expected usage patterns. If you exceed a rate limit, you'll receive a 429 Too Many Requests response. Wait briefly and retry.

Rate limits apply per IP address and are enforced globally.

Reporting fraud

If you believe you're seeing fraudulent activity on your campaigns or site, contact us at hello@trial.so. Include your advertiser or publisher ID and a description of what you're seeing.