Beckett should be clear before it asks for context.

During beta, Beckett can use Gmail, Google Calendar, Microsoft 365 Calendar, Outlook, Slack, and Chrome extension context only when you connect those tools and ask Beckett for coaching, or when you turn on an analysis setting yourself.

For Gmail, Beckett uses read-only access so it can understand email threads and help draft replies. For Slack, Beckett uses connected workspace context and may search relevant Slack history across authorized channels, DMs, group DMs, and private channels when you ask for coaching.

For Google Calendar, Beckett uses read-only access to list the calendars you choose and to read upcoming event titles, timing, and attendees from those calendars. This lets Beckett show your day and offer meeting context. Beckett does not create, edit, cancel, or respond to calendar events during beta.

For Microsoft 365, Beckett uses delegated read-only access to list the calendars you choose and read basic event titles, timing, and attendees. If you choose Analyze full thread in the Outlook add-in, Beckett also reads the messages in that one selected Outlook conversation. Beckett does not request Microsoft calendar write access, mail sending access, Teams access, or background change notifications.

The Beckett Outlook add-in reads a message or draft only after you choose it in Outlook and select Analyze message. Full-thread analysis is a separate action. Message content is sent for coaching only after you choose an analysis action. You may separately copy a response or insert it into an open draft, but Beckett never sends it.

Beckett is not meant to read your work communication in the background without your action.

Beckett stores account details, beta access status, onboarding answers, connection status, usage counts, timestamps, contacts you choose to add, coaching settings, and voluntary workday choices such as check-ins, reminder preferences, and a same-day focus when you save one. If you save or dismiss an earned learning suggestion, Beckett stores that explicit choice and the safe recommendation metadata shown to you.

This can include personal information such as your name and email address, authentication and connection information, user-provided communication preferences, and workplace communication context you choose to send for coaching.

Beckett does not store full Gmail or Slack message history by default, including raw Slack search results used for a coaching response. For product analytics and CRM, Beckett uses counts, timestamps, connection status, and safe event names, not raw message content.

Beckett does not store Google Calendar events. We retain a Google connection credential only while you keep that connection active, and remove it when you disconnect the service or delete your account.

Beckett does not store Microsoft Calendar events or selected Outlook message bodies by default. Microsoft OAuth credentials are encrypted while the connection is active and removed from Beckett when you disconnect Microsoft 365.

Beckett may store short summaries or metadata when needed to make the product work, debug beta issues, or remember your preferences.

Beckett uses user data to provide and improve its single purpose: workplace and workplace-adjacent communication coaching in Gmail, Slack, the Chrome extension, practice sessions, and skill modules.

That includes authenticating your account, enforcing beta access and usage limits, generating coaching responses, remembering your preferences, connecting Gmail, calendars, Outlook, or Slack when you ask, troubleshooting bugs, responding to support requests, and improving coaching quality.

Beckett does not use or transfer user data for purposes unrelated to workplace or workplace-adjacent communication coaching.

Google data is sent to an AI provider only when it is needed to provide a coaching feature you requested. Beckett does not use Google API data to train generalized AI or machine-learning models.

Beckett keeps your core profile, coaching preferences, voluntary workday settings, contacts, connected-service choices, and selected safety-resource region in one protected account system. When a Beckett feature is available on more than one device or surface, it is designed to follow those same choices rather than create a separate version of you.

Some actions always require a new confirmation at the moment you take them. These include connecting Gmail or Calendar, turning on notifications, any future calendar change, starting live meeting support, or saving meeting notes, transcripts, or audio. Beckett does not infer your location; it uses the country or region you choose for safety-resource information.

During beta, Beckett lets you choose a country or region for safety-resource routing. Beckett has reviewed resource sets for the United States, Canada, the United Kingdom, and Australia. If you select another region, it will say that clearly and show an international fallback rather than imply local accuracy. Beckett stores the selected region, not precise location data.

Beckett shares user data only with service providers and systems needed to run, secure, support, and improve Beckett.

These may include authentication and database providers such as Supabase, AI providers such as Anthropic for generating coaching responses, Google, Microsoft, and Slack APIs when you connect those services, hosting and infrastructure providers, analytics and debugging tools, email delivery tools, and beta/customer-support tools such as HubSpot and Loops.

Service providers receive only the information needed for their role. Beckett does not sell personal data or transfer user data to advertising platforms, data brokers, or other information resellers.

Beckett does not sell your personal data.

Beckett does not use Gmail or Slack content for advertising.

Beckett does not use or transfer user data to determine creditworthiness or for lending purposes.

Beckett does not collect payment card information through the Chrome extension during beta.

Beckett does not collect health information, precise location, or general web browsing history for its Chrome extension.

Beckett does not connect to LinkedIn, Zoom, or Google Meet during beta.

Beckett does not ask for or store your personal Anthropic API key.

The use of information received from Google APIs will adhere to the Google API Services User Data Policy, including the Limited Use requirements.

Beckett uses Google API data only to provide or improve user-facing workplace and workplace-adjacent communication coaching features requested by the user.

Beckett does not use Google API data for advertising, does not sell Google API data, and does not transfer Google API data except as needed to provide or improve Beckett, comply with applicable law, protect against abuse or security threats, or complete a merger, acquisition, or sale of assets with user consent where required.

Beckett uses HTTPS/TLS to transmit Google data and encrypts stored Google OAuth credentials using AES-256-GCM before they are written to its database. The encryption key is stored separately from the database in managed application secrets.

Connection credentials are available only to Beckett's server-side services that need them to provide the connected feature. They are not exposed to the browser, included in analytics, or written to application error logs.

Beckett does not permit routine human review of Google data. A team member may access specific data only with your explicit permission, when needed to investigate a security incident, or when required by law.

Beckett uses HTTPS/TLS to transmit Microsoft data and encrypts stored Microsoft OAuth credentials using AES-256-GCM with a dedicated production key stored separately from the database.

The Microsoft 365 connection uses delegated read-only access. It reads only basic event details from calendars you select, and reads mail only when you choose full-thread analysis for a selected Outlook conversation. Outlook message content is user-selected and is sent to Beckett only for the requested coaching action.

Microsoft credentials are available only to Beckett's server-side services that need them. They are not exposed to the browser, included in analytics, or written to application error logs.

If you submit beta feedback, that feedback may include the page, rating, your comment, and relevant debug context.

Extension feedback may include message context from the analysis you are reporting, because that helps us understand what went wrong. Only send feedback when you are comfortable sharing that context with the Beckett team.

We use beta feedback to fix bugs, improve coaching quality, and decide what needs to change before inviting more users.

To request help or report a privacy or security concern, email hello@meetbeckett.co.

During beta, Beckett acknowledges normal support requests within one business day. Active or suspected security incidents are reviewed as soon as they are discovered.

When reporting a problem, describe what happened without including private messages or coaching content unless the Beckett team specifically asks for a redacted example.

You can request account deletion from Settings. During beta, deletion is handled manually so we can remove data across Beckett, Supabase, HubSpot, email tools, and related systems.

Beckett currently targets completion within 30 days. If you need help, email hello@meetbeckett.co.