Privacy Policy

Controller: Brain Machine Hygeia (Company No. 16811156)

Address: 2 St. Marys Road, Tonbridge, TN9 2LB, England

Service: Clever Coach Club at clevercoachclub.com (the "Services")

Contact: support@brainmachineh.com

ICO: We pay/register the UK data protection fee where required. If you want our current registration details, email us.

1) What this policy covers

This Privacy Policy explains how we collect, use, share, store and protect personal data when you use the Services as a client or Trainer.

The Services provide AI-assisted fitness tools. They are not medical services. Please do not upload or enter clinical diagnoses, treatment details, medications, test results, or medical records.

2) Personal data we collect

A. Account and profile data

• Name
• Email address
• Authentication details (handled via Firebase Authentication)
• If you use Google sign-in: basic Google account info such as email and name (and, where provided by Google, profile image)
• Optional profile fields: phone number, date of birth (if you provide it), sex (if you provide it)

B. Fitness and training data

• Workout plans, workout histories, exercise completion and progress
• Exercise logs (e.g., sets/reps/weights/notes)
• Training preferences and survey answers (pre sign-up surveys and end-of-block surveys)
• Habit tracking and "general wellness" information you choose to enter
• Generic, non-medical injury labels (e.g., "knee injury", "shoulder pain")
• Optional body metrics: height, weight, BMI

C. Messages and AI Chat

In-app messages between clients and Trainers, messages to support, and chat transcripts with our AI assistants (e.g. Coach Clever).

Messaging is not end-to-end encrypted (it is protected in transit using TLS).

D. Form analysis media and outputs (optional)

If you upload a photo/video for form analysis:

• The raw photo/video (stored in Firebase Cloud Storage, typically under exercise-media/{userId}/)
• Analysis-related fields stored with exercise completion records in Cloud Firestore (e.g., an exerciseCompletions record containing fields such as mediaUrl, mediaType, and analysis outputs recorded in log fields like setLogs)

E. Technical, security and analytics data

• IP address, device/browser information, timestamps, logs needed for security and troubleshooting
• Cookies and similar technologies required for login/session and core functionality
• If you consent to analytics: pseudonymous identifiers, page views, route navigation, button/feature interactions, approximate location (country/region inferred from IP), referrer URLs, and UTM campaign parameters — collected via Google Analytics 4 (see Section 14)
• For signed-in users who have accepted analytics, your Firebase user ID may be associated with analytics events so we can understand product usage across sessions and devices

F. Payments (where enabled)

Web payments are processed by Stripe. We receive limited billing metadata (e.g., last 4 digits, card brand, payment status), not full card numbers.

On the iOS and Android apps, subscriptions may be purchased as in-app purchases through the Apple App Store or Google Play. Those purchases are processed by Apple or Google under their own terms, and we use RevenueCat to manage subscription entitlements — RevenueCat receives your app user identifier (your Firebase user ID) and purchase/subscription status, but no card details.

(At times, payment functions may be disabled in code; if payments are not active, we won't collect payment data.)

G. Wearable device data (not currently in use)

We do not currently collect any wearable or physiological data. This feature is not switched on in the Services. If we enable it in future, and you choose to connect a supported wearable device (such as WHOOP or Oura), we would collect and store:

• Physiological metrics retrieved from the device provider, such as readiness/recovery score, heart rate variability (HRV), resting heart rate, respiratory rate, sleep and related metrics, and trends and insights we derive from them
• Connection metadata (which provider you connected, connection status, and last-sync time)
• The secure access credentials (access/refresh tokens) needed to retrieve your data, which are stored encrypted and are accessible only to our backend systems — not to other users

Because metrics like HRV, resting heart rate and respiratory rate can relate to your health, we would treat this as sensitive information and process it only with your explicit consent (see Sections 4 and 6B). It is used to power fitness and wellness features and is not medical monitoring or diagnosis.

3) How we collect data

• From you (sign-up, profile, surveys, logs, messages, uploads)
• From Trainers (invites/referral codes, coaching activity within the platform)
• Automatically from your device (technical/security logs)
• From service providers (e.g., sign-in and payment status)

4) Why we use your data (purposes and legal bases)

Provide the Services (Contract)

• Create and manage accounts, authenticate users
• Generate, store and display workouts and progress
• Connect clients and Trainers
• Provide in-app messaging
• Provide form analysis when you use it

Wearable device data (not currently in use; Consent / Explicit consent if enabled)

We do not currently collect wearable data. If we enable this feature in future, the following would apply.

• Where you connect a WHOOP or Oura device, we collect, store, display and process your physiological metrics (such as readiness/recovery, HRV, resting heart rate, respiratory rate and sleep) on the basis of your consent, and your explicit consent insofar as this is treated as health data under UK/EU GDPR.
• Where you have a connected Trainer, connecting a device also relies on your explicit consent to share that data with your Trainer for coaching purposes (see Section 6B).
• Where you enable it, we use recent wearable data (typically up to the last two months) to personalise AI-generated workout programming (Coach Clever).
• You can withdraw consent at any time by disconnecting the device and/or turning off the relevant preference; this does not affect processing carried out before withdrawal.

Security and misuse prevention (Legitimate interests)

• Protect accounts and the platform
• Prevent fraud/abuse and enforce our Terms
• Investigate user reports and platform misuse (note: we do not run automated moderation today)

Improve and maintain the Services (Legitimate interests; Consent for analytics cookies)

• Debugging and reliability improvements based on error/technical logs
• Measuring how the Services are used (which features are used, where users drop off in signup/checkout, conversion rates), using Google Analytics 4. We only set analytics cookies after you consent via our cookie banner; if you decline, GA4 may still receive minimal cookieless pings for aggregated conversion modelling, but no identifiers are stored on your device.
• Investigating and fixing faults using AI-assisted engineering tools, which may process limited production data (see Section 7A)
• (We do not use Crashlytics or Remote Config. The mobile app declares push-notification capability so we can deliver transactional alerts in future; no push notifications are being sent today.)

Communications (Contract / Legitimate interests)

Transactional messages such as invitations, password resets, automated dashboard progress reports, account notices, and support responses.

Marketing (if introduced) (Consent where required)

We do not run ads trackers (no Google Ads pixel/Facebook Pixel) and we do not send SMS marketing. If we introduce marketing emails beyond essential service messages, we'll provide appropriate choices and obtain consent where required.

Model training

We do not use your content to train AI models, and we contract with our AI providers on terms under which they do not use it to train their models either (see Sections 7 and 7A). If we ever wish to do so, we will ask for your consent first, and you may decline without losing access to the Services.

5) Trainers, invitations, and what Trainers can see

Invitations and referral codes

Trainers can add clients by:

• entering a client email address, or
• sharing a referral code.

We may send transactional invitation/reminder emails to support sign-up.

Legacy Data Imports

Trainers may use our bulk import features (e.g., CSV upload) to transfer historic training records (such as past workouts and completion logs) from legacy platforms into the Services. Trainers are responsible for ensuring they have a lawful basis to transfer this data, and for giving any notice required to the individuals concerned. Imported files are parsed using the AI providers described in Section 7.

Trainer visibility (while connected)

When you connect to a Trainer, the Trainer can view, for coaching purposes (as applicable):

• Name and contact details (email; phone number if provided)
• Profile fields you provide (date of birth/age if provided; sex if provided)
• Height/weight/BMI if provided
• Workout plans, workout history, completion rates, and progress
• Training preferences and surveys (pre sign-up and end-of-block)
• Habit tracking and general wellness information you enter
• Wearable device data — not currently collected; if we enable the feature and you connect a WHOOP or Oura device while connected to that Trainer — your readiness/recovery score, HRV, resting heart rate, respiratory rate, sleep and related metrics, and the trends derived from them (see Section 6B)
• Non-medical injury labels (e.g., "knee injury")
• In-app messages between you and that Trainer
• Form analysis media you share and analysis outputs linked to that media

Disconnecting a Trainer

You can disconnect a Trainer in-app (using active/archived flags). This stops their ongoing access through the platform. However, while connected, a Trainer may be able to manually copy information they can view (for example, by taking notes or downloading what is displayed). We cannot guarantee that content viewed while connected cannot be retained by the Trainer outside the Services.

Roles with Trainers

We are the controller for operating the Services. Trainers may be independent controllers for personal data they process outside the platform for their own purposes (e.g., their own business records, offline sessions, or copies they retain).

6) Google Drive (importing your own videos)

Trainers may optionally import exercise demonstration videos from their own Google Drive. This is a one-way import into the Services: we read the specific files you choose and store a copy for use in your exercise library.

Access is limited to files you pick. We request the narrow drive.file permission, which lets us read only the files you actively select in the Google picker. We cannot browse, list or access anything else in your Drive.

Bulk import. If you use bulk import, you may need to set your own video files to "Anyone with the link" so we can fetch them. That setting applies to your own demonstration videos, and while it is in place anyone holding the link can open the file. You can change it back once the import is finished.

We do not export anything to Google Drive. The Services do not copy, upload or store client data, client media, or any other content into a Trainer's Google Drive or any other Drive account. There is no export path.

6A) Exercise demonstration videos (Bunny.net)

Trainers may upload exercise demonstration videos to the Services. These videos are hosted on Bunny.net ("Bunny CDN" / "Bunny Stream"), a third-party video hosting and content delivery platform that processes and delivers video data on our behalf.

What is stored: The video file itself is stored in Bunny Storage and made available for streaming via Bunny Stream. A reference URL pointing to the Bunny-hosted video is stored in Cloud Firestore alongside the relevant exercise data.

Access: Bunny-hosted video URLs may be accessible to anyone who obtains the URL. URLs are not publicly listed but are not individually access-controlled beyond URL knowledge.

No biometric identification: We do not use exercise demonstration videos for facial recognition or biometric identification.

Retention: Exercise demonstration videos are retained on Bunny.net while actively associated with exercises in the Services. If a Trainer removes or replaces a video, the previous version may be deleted from Bunny Storage. Videos are not automatically deleted upon account closure; deletion of Bunny-hosted content follows our standard retention schedule as described in Section 10.

6B) Wearable device integrations (not currently in use)

Not currently available. Wearable device connection is not switched on in the Services today. We do not currently collect, store or receive any wearable or physiological data, and there is nothing to connect. The description below explains how the feature will work if and when we enable it, and we will update this document before we do.

If enabled, you would be able to optionally connect a supported wearable device — currently WHOOP or Oura (Ōura) — to the Services from your Profile page. This is optional, and the Services work without it.

How the connection works: Connecting uses a secure authorisation flow (OAuth 2.0) operated by WHOOP or Oura. You log in to your device account and grant us permission to access defined categories of your data. We store the resulting access credentials encrypted on our backend, and use them to retrieve your metrics on a periodic and/or on-demand basis.

What we receive and store: readiness/recovery score, heart rate variability (HRV), resting heart rate, respiratory rate, sleep and related metrics, and trends and insights we derive from them. We store these in Cloud Firestore associated with your account and display them to you (for example on your dashboard and on the Physiology page).

Explicit consent to share with your Trainer: if you connect a device while you have a connected Trainer, you explicitly consent to share all data we receive from that device — including your readiness/recovery score, HRV, resting heart rate, respiratory rate, sleep and related metrics, and the derived trends and insights — with your connected Trainer(s) for coaching purposes. If you do not want a Trainer to see this data, do not connect a device while connected to that Trainer, or disconnect it.

Use for AI programming (optional): where you (or your Trainer, with your agreement) enable the relevant preference, we provide your recent wearable data (typically up to the last two months) to our AI assistant (Coach Clever, powered by our AI provider) to help personalise and adjust your workout programming. You can turn this off in the relevant preferences.

Disconnecting and withdrawing consent: you can disconnect a device at any time from your Profile page, which stops further retrieval of your data through the Services. You can also revoke our access directly from your WHOOP or Oura account. Data already retrieved before you disconnect is retained and deleted in line with Section 10.

Third-party providers: WHOOP and Oura are independent third parties with their own terms and privacy policies, which govern your device and your account with them. When we request your data from them, that request is processed by the provider (including, where applicable, in the United States — see Section 9).

No biometric identification: We do not use wearable data for facial recognition or biometric identification, and we do not use it to make decisions about you that produce legal or similarly significant effects without human involvement.

7) Generative AI processing (Anthropic and Google)

Some features of the Services are delivered using third-party AI providers. They process data on our behalf as our processors, under written contract. We currently use:

• Anthropic PBC and its group companies, including Anthropic Ireland, Limited (the Claude API) — our conversational assistants ("Coach Clever" and "How Do I"), workout plan generation and editing, and parsing of data files that Trainers import. Data sent: your chat prompts, relevant workout and training history, and the contents of imported files.
• Google LLC (the Gemini API) — AI form analysis of media you upload, coaching profile summaries ("Coach DNA"), and exercise search indexing. Data sent: your uploaded media, and your training history and submitted preferences.

Both providers process data in the United States (see Section 9). We can provide our current AI provider and sub-processor list on request (email support@brainmachineh.com).

Model training: we have contracted with both providers on terms under which the data we send them is not used to train their AI models. Neither we nor our providers use your identifiable content to train models.

Provider-side retention: under our commercial agreements, our AI providers delete the content of a request within 30 days. If a request is flagged as breaching a provider's usage policy, that provider may retain it for longer — in Anthropic's case up to 2 years, and safety classification scores for up to 7 years. This is separate from our own retention, described in Section 10.

No biometric identification: We do not use AI features for facial recognition or biometric identification, and no AI feature makes a decision about you that produces a legal or similarly significant effect without human involvement.

Imported data: where a Trainer imports historic records (see Section 5), the file may contain the names and email addresses of people who do not yet have an account. That content is processed only to structure the import, and is not retained by the provider beyond the period described above. Trainers are responsible for having a lawful basis for the import.

Your responsibility: Only upload media you have the right to share, including permission from anyone who appears in it.

7A) AI-assisted engineering and support

To build, maintain, debug and secure the Services, we use an AI-assisted development tool — currently Anthropic's Claude (including Claude Code). When we investigate a fault or a support request, these tools may process live production data, including records relating to your account. Only our two company directors have this access; we do not give production access to contractors.

We do this on the basis of our legitimate interests in keeping the Services working, secure and accurate. We limit access to what is needed to resolve the issue in hand.

Anthropic acts as our processor under written contract, on a business account held by us and configured so that the data is not used to train their models. We contract with Anthropic PBC and/or its group companies (including Anthropic Ireland, Limited, as applicable). Processing may take place in the United States, under the safeguards described in Section 9.

Session records: unlike the product features in Section 7, a record of an engineering session is kept in our workspace with the provider rather than being deleted after 30 days. Our accounts are configured so that these records are automatically deleted after one year.

You have the right to object to processing we carry out on the basis of legitimate interests. Email support@brainmachineh.com and we will consider your request, though we may not be able to stop this processing while you continue to use the Services, because we cannot maintain and repair the Services without being able to inspect the data they hold.

8) Who we share data with

Our processors/service providers

We use service providers that process personal data on our behalf under contract, including:

• Google Firebase (Authentication, Hosting, Cloud Firestore, Cloud Storage, Cloud Functions)
• Anthropic PBC (Claude API — AI processing for our chat assistants, workout plan generation and editing, and parsing of imported data files)
• Google Gemini API (AI processing for form analysis, coaching profile summaries and exercise search indexing)
• Google Analytics 4 (product usage analytics — only if you consent via the cookie banner)
• Bunny.net (Bunny CDN / Bunny Stream) for exercise demonstration video hosting, streaming, and content delivery
• Resend (transactional emails)
• Stripe (web payments, where enabled)
• RevenueCat (subscription entitlement management for in-app purchases on iOS/Android — receives your app user identifier and purchase/subscription status)
• Apple App Store and Google Play (processing of in-app purchases on their platforms, under their own terms)
• Cloudflare (Turnstile bot protection on certain forms — processes technical signals such as IP address and device/browser information to distinguish humans from bots)
• WHOOP and Oura (Ōura) — not currently in use. If we enable wearable connections in future, these providers would supply physiological data to us via their APIs, in line with the permissions you grant

We can provide a current sub-processor list on request (email support@brainmachineh.com).

Other disclosures

We may share data:

• With your connected Trainer (as described above)
• With professional advisers (legal/accounting) where needed
• With authorities where required by law

We do not sell your personal data.

9) International transfers

Where our infrastructure runs:

• Cloud Functions (our backend compute) are deployed in europe-west2 (London, UK).
• Cloud Storage (uploaded media such as profile photos and form-analysis videos) is hosted in europe-west2 (London, UK).
• Cloud Firestore (our primary database, holding account, profile, training, survey and message data) is hosted in Google's United States multi-region (nam5). This means that data is stored on Google infrastructure in the United States, under Google's data processing terms and the transfer safeguards described below.

US processing also occurs for:

• Anthropic (Claude API) requests, which are processed on US-based infrastructure,
• Gemini API requests (which route to US-based endpoints),
• Anthropic, where our engineering tooling processes production data during support and debugging (Section 7A), which may be processed on US-based infrastructure,
• Google Analytics 4 (event data is processed via Google's global infrastructure, which includes US-based servers; we have enabled IP anonymisation and have not enabled advertising features),
• Resend (US-hosted email delivery infrastructure),
• RevenueCat (US-hosted subscription entitlement infrastructure),
• Bunny.net (video content is delivered via Bunny's global CDN edge network; storage may be located in EU or other regions depending on configuration), and
• WHOOP and Oura — not currently in use; if enabled in future, our requests to retrieve your data would be processed by these providers, which may process data on US-based infrastructure.

Where data is transferred outside the UK/EEA, we rely on appropriate safeguards (such as the UK IDTA / UK Addendum to the EU Standard Contractual Clauses and equivalent contractual protections used by our providers, including Google's Data Processing Addendum), plus additional measures where appropriate.

10) Retention

We keep personal data only as long as needed for the purposes in this policy:

• Account/profile, training data, surveys, habits, body metrics, and messages: generally retained while your account is active.
• After account closure: we aim to delete or anonymise most account-related data within 90 days, unless we need to keep some records longer for legal/accounting reasons (typically up to 6 years for business records).
• Form analysis media and outputs: you may be able to delete media in-app (where available). Otherwise, we generally retain it while your account is active and aim to delete/anonymise within 90 days after closure.
• Wearable device data: not currently collected. If enabled in future, connection credentials would be retained until you disconnect the device or close your account, after which they are revoked/deleted. Physiological metrics and derived trends are retained while your account is active and we aim to delete/anonymise them within 90 days after closure. Disconnecting a device stops further collection.
• Data sent to AI providers: retained by the provider for a short security and abuse-monitoring period (typically up to 30 days), then deleted from their systems. It is not used to train models.
• Backups: deletion may not be immediate across backups; backups may persist for up to around 90 days.

11) Security

We use reasonable technical and organisational measures designed to protect data, including encryption in transit (TLS), access controls, and secure cloud infrastructure. In-app messaging is not end-to-end encrypted. We do not currently implement automated moderation; we may review content where needed for support, security, or in response to reports.

No system is 100% secure; please keep your credentials confidential and use secure devices.

12) Your rights

Under UK GDPR (and, where applicable, EU GDPR), you may have rights to:

• access your data
• correct inaccurate data
• request deletion (you can permanently delete your account at any time via the Profile Settings page in the app, or by contacting support)
• restrict processing
• object to certain processing based on legitimate interests
• data portability (where applicable)
• withdraw consent where processing is based on consent (if we introduce consent-based features)

To exercise rights, email support@brainmachineh.com. We respond within one month (subject to lawful extensions) and may ask for verification.

You can also complain to the ICO (www.ico.org.uk).

13) Children and age

The Services are intended for people aged 16 or over. You must confirm your age at sign-up, and where we collect a date of birth we check it against that minimum and refuse the registration if it is not met. We do not otherwise independently verify age. If we become aware or reasonably suspect that a user is under 16, we may request confirmation and may suspend/terminate the account.

14) Cookies and similar technologies

We use cookies and similar technologies (such as browser localStorage and IndexedDB) for two purposes: keeping you signed in, and — only with your consent — understanding how the Services are used.

Strictly necessary (no consent required)

These are required to operate the Services. You cannot opt out and still use the site.

• Firebase Authentication session storage (stored in browser IndexedDB and localStorage) — keeps you signed in across pages and reloads.
• App preferences (stored in browser localStorage, e.g. ccc_consent, ccc_app_version, pending trainer intent flags) — remember UI state and your consent choice between visits.
• Stripe session cookies on the checkout page — required for payment fraud prevention.

Analytics (consent required)

When you accept analytics via our cookie banner, we use Google Analytics 4 to understand product usage. GA4 sets the following first-party cookies on clevercoachclub.com:

• _ga — distinguishes unique visitors. Expires after 2 years.
• _ga_G-6JJVMP5KGS — persists session state. Expires after 2 years.

We have configured Google Consent Mode v2 so that, until you accept analytics, these cookies are not set. We have not enabled Google Analytics advertising features (no Google Signals for ad personalisation, no remarketing audiences, no Google Ads link). We do not run advertising trackers (no Google Ads pixel, no Facebook Pixel, no TikTok pixel).

Changing your choice

You can change your cookie choice at any time using the Cookie preferences link in the website footer, or by clearing your browser's site data for clevercoachclub.com (which will cause the consent banner to reappear on your next visit).

You can also opt out of Google Analytics across all sites by installing the Google Analytics Opt-out Browser Add-on, or by enabling "Do Not Track" / Global Privacy Control in your browser.

15) Changes to this policy

We may update this Privacy Policy from time to time. We will post the updated version with a new "Last updated" date and provide notice of material changes where appropriate.

16) Contact

Brain Machine Hygeia

2 St. Marys Road, Tonbridge, TN9 2LB, England

Email: support@brainmachineh.com