Transparency
Transparency report
We've been running Fray in our own ad blocker since early 2025. We will report quarterly on the evidence we file, the action platforms take and the disputes we receive.
Quarterly figures have not yet been published.
Filings and outcomes
| Metric | Value | Notes |
|---|---|---|
| Reports filed with platforms | Not published | Each filing identifies an advertiser account and creative set, with evidence from our own machines' impressions where reproduction succeeded, and from sanitized partner sightings where cloaking or targeting defeated it. Filings state which. |
| Advertiser accounts removed or suspended | Not published | Removed or suspended by the platform following our report. |
| Campaigns disrupted | Not published | A campaign may span several accounts; returns of the same campaign are counted once. |
| Decline rate | Not published | The share of filings where the platform took no action. Declined cases are never published. |
Disputes
| Metric | Value | Notes |
|---|---|---|
| Disputes received | Not published | Received at [email protected]. The dispute form is not connected. |
| Median resolution time | Not published | Target first response: 2 business days. |
Watchlist, register and partners
| Metric | Value | Notes |
|---|---|---|
| Watchlist size | Not published | Number of brands monitored. The public watchlist lists the domains currently covered by the registry; it is not a count of investigations or outcomes. |
| Brands with registry records | Not published | The public lookup contains three fictional examples: example-fashion.com, example-bank.co.uk and example-coffee.com. These are not a count of brands Fray works with. |
| Contributing partners | Not published | Each runs its own token issuer with monthly published epoch keys. |
| Sightings contributed | Not published | Aggregate count only; raw envelopes expire after 90 days. |
Detection details
We do not publish detection signals, time from sighting to detection, or live latency dashboards. Those details would help scammers evade detection.
Methodology questions: [email protected].