Privacy
Last updated 2026-05-14
SpectraQC is a broadcast quality-control service. We process media files you upload to check them against industry delivery specifications, and return a PDF report. This page tells you what we store, where it lives, and how long we keep it.
What we collect
- Your account email + name (via Clerk)
- Uploaded media files (video, audio)
- QC results: which checks passed, measured values, generated PDF
- Transcripts (where audio is decoded for speech-related checks)
- Billing details (via Stripe — we never see card numbers)
- HTTP request metadata for security + abuse prevention
How long we keep it
| Data | Retention |
|---|---|
| Uploaded media | Deleted on job completion |
| QC reports + job metadata | 1 year |
| Transcripts (anonymous-keyed) | 90 days |
| Billing events | 7 years (UK/EU accounting law) |
| Account itself | Until you delete it |
Uploaded source media is deleted by the analysis worker as the final step of every job (success and failure paths alike); a 1-day S3 lifecycle rule on the ingest bucket sweeps anything the worker fails to delete.
Where it's stored
Customer media + reports live in AWS S3 in the EU (eu-west-1, Ireland). DynamoDB tables and Transcribe jobs run in the same region. Access logs and backups stay in-region.
Cross-border transfer: Visual QC analysis uses Anthropic's Claude API, which processes decoded frames in the United States under EU SCCs. No audio, no transcript, no filename, no customer identifier accompanies the request. Broadcaster customers requiring zero-retention or EU-region routing can opt out — see sub-processors.
Your rights
- Access — request a copy of everything we hold on you.
- Erasure — request deletion of your account and all linked data. We comply within 5 working days (35 days for the backup-snapshot scrub).
- Portability — your QC reports are downloadable as PDF from the app at any time.
- Objection — refuse a new sub-processor we add. Email dpo@spectraqc.io; we'll discuss alternatives, and your contractual exit rights apply if no resolution is reached.
Contact dpo@spectraqc.io to exercise any of these rights.
Sub-processors
We use AWS, Anthropic, Vercel, Clerk, Stripe, and Slack as sub-processors. Each handles a specific slice of data — the full list with data categories and regions is public.
Analytics
We use Vercel Analytics and Vercel Speed Insights to measure aggregate page views and Core Web Vitals on the marketing site. Both are cookie-less, do not store IP addresses, and do not build a profile of individual visitors. Dynamic path segments (job IDs, spec IDs) are stripped client-side before any event is sent, so the analytics dashboard only sees route shapes like /jobs/[id], not the underlying identifiers.
Questions: dpo@spectraqc.io. Data controller: SpectraQC Ltd., Dublin, Ireland.