Privacy.
Draft for review
This is a draft for review. It is not legal advice. It must be finalized by the founder with a qualified lawyer before launch. Placeholder tokens like <LEGAL-ENTITY-NAME> and <EFFECTIVE-DATE> are deliberate and will be replaced on incorporation.
Effective date: <EFFECTIVE-DATE> · Data controller: <LEGAL-ENTITY-NAME>, <REGISTERED-ADDRESS>.
Who we are
This service is operated by <LEGAL-ENTITY-NAME> (we, us, our), registered at <REGISTERED-ADDRESS>. For privacy questions, contact <CONTACT-EMAIL>. Our Data Protection contact is <DPO-CONTACT>.
Under the UK GDPR and the EU GDPR, we are the data controller for the personal data described below. The third-party AI providers named in this notice act as our data processors.
What we collect (website)
The waitlist form collects only your email address. We store it in our database (Supabase) and use it to email you about the product and the launch. You can unsubscribe from any email using the link in its footer.
What we collect (app)
When the app launches, we process: your account details (email, sign-in identifier), your practice-call audio and transcripts, the text of your lesson responses, your scores and progress, and basic device and usage analytics. We do not ask for, and do not want, real-world identities of anyone you describe in practice.
Third-party AI providers we share data with
The app uses named third-party AI services to deliver coaching. Before any feature sends your data to one of these providers, we ask for your explicit consent on a dedicated screen. You can withdraw that consent at any time in Settings. The providers are:
- Anthropic (Claude) for coaching analysis and feedback. We send the text of your lesson responses and call transcripts so Claude can generate a coach read. Anthropic processes this as our processor and, per its commercial terms, does not train its models on this data.
- ElevenLabs for voice synthesis and conversational voice. We send call audio and text. Voice may constitute biometric data under the GDPR (Article 9), so we treat it as a special category and process it only with your explicit consent. We do not permit your voice to be used for provider model training.
- OpenAI (Whisper), if used for speech-to-text transcription of your practice calls. Where this is used, we send call audio for transcription only, with training disabled.
These providers may process data on servers outside the UK and EEA. Where that happens, we rely on Standard Contractual Clauses or an equivalent transfer mechanism, and a Data Processing Agreement with each provider. We will name the exact transfer basis here once the provider contracts are signed (<LEGAL-ENTITY-NAME> to finalize).
Voice and transcripts
Practice-call transcripts are retained for 30 days, then deleted, unless you save a specific call to your history. Raw call audio is held only as long as needed to produce the transcript and coach read, then discarded. You can delete any saved call at any time in Settings.
AI training opt-out
We do not use your content to train any AI model by default, and we contractually require our AI providers not to train on it either. There is no setting that turns training on. If that ever changes, it will be strictly opt-in, off by default, and announced here first.
Your rights
Under the UK and EU GDPR you can: access a copy of your data, export it in a portable format, correct it, delete it, restrict or object to processing, and withdraw consent (including consent to AI data sharing) at any time. Access, export, and delete are self-serve in Settings. For anything else, write to <CONTACT-EMAIL>. You also have the right to complain to the UK Information Commissioner's Office (ICO) or your local supervisory authority.
Email and analytics
We send transactional and lifecycle email through Resend. We keep a minimal send log (recipient, template, status) so we do not email you twice and so we can honor unsubscribes. We use privacy-respecting product analytics to understand which lessons work; this is not sold or shared for advertising.
Full list of sub-processors
Beyond the AI providers above, we use the following processors to run the service. Each acts on our instructions under a data processing agreement, and where data leaves the UK and EEA we rely on Standard Contractual Clauses or an equivalent mechanism.
- Supabase , database, authentication, and file storage (account, lesson responses, scores, saved calls).
- Anthropic, OpenAI, ElevenLabs , the AI providers described above (coaching, transcription, voice).
- RevenueCat , subscription management (purchase and entitlement data).
- Resend , transactional and lifecycle email (email address, send metadata).
- PostHog , privacy-respecting product analytics (pseudonymous event data; no raw email).
- Sentry , error and performance monitoring (diagnostic data).
- Vercel , website hosting and cookieless web analytics.
- Cloudflare , DNS, anti-spam (Turnstile), and email routing.
We keep this list current as the service evolves. If we add a sub-processor that materially changes how your data is handled, we will update this notice.
Children
This product is for adults aged 17 and over (App Store 17+; marketed to men aged 18 and over). It is not directed at children and we do not knowingly collect their data.
Contact
Privacy questions: <CONTACT-EMAIL>. Data Protection contact: <DPO-CONTACT>. You can also reach us at privacy@talktowomen.ai.