Privacy Policy
Last updated: 2026-08-04
CRWDS is a campaign-management and delivery platform for brand advertising inside livestreams, operated by Hyped, Lda. This Privacy Policy explains what personal data we collect, why, how long we keep it, and the rights you have under the General Data Protection Regulation (GDPR).
Who we are
The data controller responsible for the personal data described in this policy is:
- Hyped, Lda
- NIPC 514727926
- Rua António Correia Calhatro, nº 11, 4700-126 Braga, Portugal
- Email: crwds@goinghyped.com
Data Protection Officer
Hyped, Lda has assessed Article 37(1) GDPR and concluded that appointment of a Data Protection Officer is not mandatory at current scale and processing model. Our core activities involve aggregation of publicly available streaming platform data (viewer counts, follower counts, stream metadata via Twitch/YouTube/TikTok public APIs) combined with minimal private data (operator emails, encrypted payout IBAN). We do not engage in large-scale systematic monitoring of data subjects or processing of special categories of personal data under Article 9. We will reassess this position annually or when material scale or processing-model changes occur.
Privacy enquiries: crwds@goinghyped.com.
What we collect
- Streamer identity — the OAuth-verified platform handle and immutable platform user id (Twitch, X, Instagram, TikTok, YouTube) you authorise CRWDS to link to your CRWDS account.
- Follower / subscriber counts — fetched from each linked platform's public API at link time and refreshed on a daily schedule.
- OAuth access and refresh tokens — encrypted at rest with platform-isolated symmetric keys. The plaintext exists only in CRWDS's process memory at request time and is never logged.
- Profile fields you provide voluntarily — legal name, phone number, tax identification number (NIF), residential address, postal code, and country of residence. The identity-document fields are encrypted at rest.
- Payout information — IBAN (encrypted at rest), account holder name and tax country.
- Campaign delivery data — overlay impressions, viewer-hour aggregates, and per-campaign statistics generated while a campaign is active on your stream.
- Operational metadata — login timestamps, session activity, and audit-trail rows for sensitive actions (e.g. IBAN changes).
- Chat keyword mentions — on channels where a streamer has opted a campaign's chat bot in, a pseudonymous identifier and a short message excerpt are recorded when a viewer's public chat message contains a configured brand keyword. Nothing is recorded for messages that do not match. See CRWDS chat bot.
Why we collect it
- Campaign delivery — rendering brand overlays on your stream and tracking the resulting impressions.
- Analytics — surfacing per-campaign performance to both the streamer and the brand customer.
- Payouts — manual SEPA payouts to the IBAN you provide on a monthly reconciliation cycle.
- Identity verification — confirming that the streamer linking each platform account owns the underlying handle.
- Tax and accounting compliance — under Portuguese commercial law (see retention below).
Third-party services
CRWDS uses the following platform APIs and complies with each platform's developer terms:
- YouTube — CRWDS's use and transfer of information received from Google APIs adheres to the Google API Services User Data Policy, including the Limited Use requirements. The
youtube.readonly scope is used solely to verify the streamer's channel identity and to read public subscriber counts.
- Instagram — CRWDS adheres to the Meta Platform Terms and the Instagram Platform Policy. The
instagram_business_basic scope is used solely for identity verification and reading the public follower count.
- TikTok — CRWDS adheres to the TikTok Developer Terms of Service. The
user.info.basic, user.info.profile and user.info.stats scopes are used for identity verification and reading public follower counts. We store only the streamer's handle, opaque platform user id, and follower count. The API request queries additional public fields (following count, likes count, video count) which are not retained — only the follower count is stored as the REACH metric.
- X (Twitter) — the streamer manually enters their @handle and follower count via the streamer portal. CRWDS does not access the X API, does not request OAuth, and does not store any tokens. Only the manually-entered handle and self-reported follower count are stored.
- Twitch — CRWDS adheres to the Twitch Developer Services Agreement. The
user:read:email scope is used for streamer-portal login and identity verification; the moderator:read:followers scope is used solely to read the total follower count for the REACH metric. No individual follower data is fetched or stored — only the aggregate count. Separately, CRWDS operates a chat bot on Twitch under its own account — see CRWDS chat bot below.
With the single exception of the CRWDS chat bot described below, CRWDS does not post, message, follow, or take any action on the streamer's behalf on any of these platforms. The streamer-portal OAuth scopes listed above are read-only.
CRWDS chat bot
CRWDS operates a chat bot on Twitch under the account @CRWDSat. It is operated by Hyped, Lda — the same controller as the rest of the CRWDS platform — and it is active only in the channels of streamers who have accepted a campaign assignment with the bot enabled. It is never present in any other channel.
What the bot does
- Posts campaign messages. While an opted-in streamer is live, the bot posts a campaign-configured message prefixed
[Ad] at a cadence set per campaign between 5 and 60 minutes. It uses the user:write:chat and channel:bot scopes granted for its own bot account.
- Listens for brand keywords. On opted-in channels, the bot subscribes to chat messages to count mentions of the brand keywords configured for that campaign.
- Watches the stream category. The bot is notified when an opted-in channel changes its Twitch category, so that it can stop posting automatically if the stream moves to a category the brand has excluded.
What we process about viewers
When a viewer's chat message contains a configured brand keyword, and only then, CRWDS records:
- A pseudonymous identifier: the viewer's Twitch account name combined with a per-streamer value, hashed with SHA-256, of which only the first 8 hexadecimal characters are kept. Because the hash is salted per streamer, the same viewer produces a different identifier on every channel — they cannot be tracked across streamers.
- A short excerpt of the matching message — at most 60 characters, being roughly 20 characters either side of the keyword.
- The keyword matched, the campaign, the streamer, and the time.
Raw chat content, usernames, and unhashed identifiers are never stored. Messages that do not contain a configured keyword are discarded in memory and never written down. The bot does not read private messages, and it has no access to any channel it has not been invited to.
Legal basis
- Streamers — Article 6(1)(b), performance of a contract: bot operation is part of the campaign participation the streamer accepted.
- Viewers — Article 6(1)(f), legitimate interests: measuring campaign effectiveness in a pseudonymised, non-identifying form. Public chat in a channel running a disclosed
[Ad] campaign is the narrow scope; the balancing test rests on the pseudonymisation, the 60-character excerpt limit, the per-streamer salt, and the 30-day deletion below.
Retention of chat-derived data
- Keyword mention records are deleted after 30 days. This is enforced automatically by the database itself, with the 30-day cut-off fixed server-side rather than left to application code.
- Aggregate counts derived from them (totals per campaign, per streamer, per keyword) carry no identifier and are kept indefinitely.
- A record of each message the bot posted — its text, the channel, and the time — is kept as an audit trail for as long as the campaign record itself is retained. It contains no viewer data.
- Click counts on campaign links are aggregate figures only, and identify no individual.
Viewer rights
The rights listed below apply to this processing. In practice, because the stored identifier is a salted hash and the excerpt is short and carries no username, CRWDS usually cannot locate a specific viewer's records without that viewer telling us their Twitch account name and the channel concerned — we then derive the same hash and search for it. If you would rather not have your chat messages counted at all, contact us at crwds@goinghyped.com and we will exclude you.
Where this data goes
Chat events reach CRWDS from Twitch. Nothing derived from chat is shared with any third party. Brands see aggregate mention counts and the anonymised excerpts described above; they never receive identifiers. The sub-processors in Annex A are unchanged by the bot.
Your GDPR rights
You can exercise the following rights at any time by emailing crwds@goinghyped.com:
- Access — receive a copy of the personal data CRWDS holds about you.
- Rectification — correct any inaccurate or outdated personal data.
- Deletion — request deletion of your account and personal data. See the Data Deletion page for the process and timelines.
- Portability — receive your personal data in a structured, machine-readable format.
- Restriction and objection — restrict or object to specific processing activities.
- Withdrawal of consent — disconnect any linked OAuth account at any time from the streamer portal; doing so revokes the stored tokens.
- Complaint — lodge a complaint with the Portuguese data-protection authority (CNPD) or the supervisory authority in your country of residence.
Retention
Personal data is retained for as long as your CRWDS account is active. After account deletion or termination:
- Account-linked PII (profile fields, OAuth tokens, social-link records) is deleted within 30 days.
- Campaign performance data and payout records (including IBAN) are retained for 10 years from the end of the relevant fiscal year, in alignment with Portuguese commercial law (Código Comercial Art.40 + RGCT) and tax record-keeping obligations.
- Internal audit-trail entries (e.g. IBAN changes, OAuth-token rotation) are retained for 5 years in non-identifying form.
- Error-tracking provider and log-aggregation provider records are retained for 90 days by default.
- Anonymised aggregate statistics (no PII) may be retained indefinitely for operational reporting.
Security
Sensitive fields (IBAN, OAuth access and refresh tokens, tax identification number, residential address lines) are encrypted at rest by our database provider with symmetric keys. Plaintext values exist only in CRWDS's process memory at request time and are never written to records hosted at our error-tracking provider or log-aggregation provider. Access to production data is restricted to the operator and audited.
Children
The CRWDS service is not directed to children under 16 years of age (GDPR Article 8 default). We do not knowingly collect personal data from children under this age. If we become aware that we have collected personal data from a child under 16 without verifiable parental or guardian consent, we will terminate the account and delete the data.
Changes to this Policy
We may update this Privacy Policy from time to time. Legally significant changes are communicated by email to affected data subjects. The effective date at the top of this page reflects the current revision.
Annex A — Sub-processors
The following sub-processors process personal data on behalf of Hyped, Lda under written contracts complying with GDPR Article 28. Cross-border transfers are governed by Standard Contractual Clauses (SCCs, EU 2021/914) where applicable.
- Supabase — database, hosted on AWS in Frankfurt (eu-central-1). Replaced Neon as our database provider in 2026; Neon no longer holds CRWDS data.
- Vercel — hosting for admin and brand portals (EU where available; SCCs apply where applicable).
- Cloudflare R2 — object storage (EU where available; SCCs apply where applicable).
- Better Stack — log aggregation (EU, eu-fsn-3).
- Google — OAuth identity provider (global; SCCs apply).
- Microsoft Entra ID — OAuth identity provider (global; SCCs apply).
- Twitch — public-API metadata source and streamer-portal sign-in (global; public-API consumer).
- YouTube — public-API metadata source (global; public-API consumer).
- TikTok — public-API metadata source (global; public-API consumer).
- X (Twitter) — public-API metadata source (global; public-API consumer).
- Instagram (Meta) — public-API metadata source (global; public-API consumer).
Contact
Questions about this policy or about how CRWDS handles your data: crwds@goinghyped.com.