Privacy Policy
Last updated: 13 October 2026
The short version
- Texterry stores message text and conversation details for your inbox, search, sync and enabled AI features.
- Telegram, WhatsApp and Discord attachments are not kept as a server media library. Delivery can involve temporary processing; your Mac can cache or save files.
- Messenger sessions are stored encrypted on our servers. Connections let Texterry receive and send messages on your behalf.
- You choose per chat whether the assistant is off, reads and digests, or replies. Other AI processing has limits explained below.
- We do not sell or resell your data, use it for advertising, or run ads. Providers help operate the service.
- Disconnect messengers or delete your account in Settings. Shared-team work, operational records and local files have separate retention rules.
- The website has no cookies, browser storage or analytics. The optional Windows waitlist stores only your email, platform and signup timestamp for one availability email.
Contents
- Who we are and what this policy covers
- Account, messenger, conversation and CRM information
- AI features and your controls
- Voice dictation and media handling
- Information on your Mac
- Device information, diagnostics and usage
- Payments and support
- How we use information and our legal bases
- Sharing and service providers
- No sale, resale or advertising use
- International transfers
- How long information stays
- Disconnecting and deleting an account
- Security
- Your choices and privacy rights
- California privacy notice
- Notes for EEA and UK users
- Other people's information and Business mode
- Third-party messengers and integrations
- Children
- Website visitors, Windows waitlist and storage technologies
- Do Not Track
- Changes to this policy
- Contact
1. Who we are and what this policy covers
Innolope LLC operates Texterry. “We”, “us” and “our” mean Innolope LLC. Our address and email appear in section 24.
Texterry is an Electron app for Mac combining Telegram, WhatsApp and Discord, with optional AI, dictation and a CRM and shared workspace called Business mode. Plans are Free, Plus and Pro; paid plans are coming soon. This policy covers the app, backend, website, Windows waitlist and support, including correspondents' and business contacts' information.
Our Terms govern the service. Their Delaware governing-law provision does not remove applicable privacy rights.
2. Account, messenger, conversation and CRM information
Information depends on your connections and features. Messages can contain almost any personal information, including sensitive facts you or a correspondent includes.
- Your Texterry account. The app starts with a Firebase guest identifier linked to your activity. Email is optional. We store the identifier, supplied email and display name, timestamps, memberships, current team, assistant preferences and plan status.
- Sign-in requests. Email-link sign-in uses your email, requesting account, language, status, expiry and a hash of the secret used to collect the result. A completed request can temporarily hold a one-time Firebase token until collected. You do not create a Texterry password.
- Messenger connections and credentials. Sessions, linked-device credentials or tokens authorize messenger access. Records can include labels, account identifiers, phone, username, name, connection status, sync preferences and ownership or team scope. Telegram profiles can include a small account picture. Sessions are stored encrypted on our servers.
- Messages and conversation data. We store conversation text, chat and message identifiers, sender names or identifiers, timestamps, direction, type and CRM-contact links on our servers. This supports inbox, search and sync across restarts and devices, and enabled AI replies, digests and summaries. Switching AI off does not remove inbox records.
- Attachment details. Records can contain type, file name, size, links and retrieval information, including WhatsApp media keys. These details are distinct from the attachment itself; handling and local copies are explained below.
- People in your chats and contacts. Messengers and your entries supply names, usernames, phone numbers and identifiers. CRM contacts may include email, company, address, channel identities, tags, custom fields, notes and history. Matching and merge records link people across channels and trace changes.
- CRM and teams. Business mode stores clients, deals, amounts, currencies, stages, tasks, assignments, notes, tags, templates, timelines and workflows. Teams have names, member identifiers, roles and former-member attribution. CRM connections include credentials, workspace and record identifiers, sync dates, webhook information and retry status.
- AI inputs and results. We handle prompts and instructions, relevant messages, names, timestamps, transcribed speech and generated results. Stored results may include digests, lead summaries, extracted events, transcription text and writing-style descriptions with example phrases. These are personal information when they relate to an identifiable person.
3. AI features and your controls
Texterry sends requests to Google's Gemini API for generation and transcription. Depending on the action, Google receives text, sender names, times and message identifiers, attachment descriptions, instructions, writing-style context, an existing lead summary, or audio. It returns text or structured results.
You choose, per chat, whether the assistant is off, reads and digests, or replies. The default assistant mode is off. Off prevents the assistant from capturing new messages for digests or automatically answering that chat; reading mode prepares digests without automatically replying. Reply modes can write as you or as a named assistant, depending on your settings. A stop control can stop automatic replies. Stopping replies and deleting stored information are separate actions.
Other AI actions have their own triggers. Asking for a draft or transcription sends the content needed for that action. Voice-message transcription can process incoming and outgoing spoken attachments as they arrive, separately from the assistant's per-chat mode. Business-mode automatic lead summaries follow the team's workflow settings. An assistant setting of off is therefore not a universal prohibition on every AI request associated with a conversation.
Gemini can describe your writing style using outgoing examples from the relevant chat or across recently active personal Telegram chats. Descriptions and example phrases are stored as context for later replies. This personalization is distinct from training a model.
Writing-style sampling is not restricted by each source chat's assistant setting. A chat marked off therefore should not be understood as an absolute prohibition on every AI use of its content. Turning off a chat assistant also does not disable the separate incoming transcription or Business-mode summary paths described above.
We do not train our own AI models on your messages. The application uses Gemini to generate responses and descriptions. Google's processing is governed by the Gemini API terms, which distinguish paid and unpaid services. Paid-service terms restrict product-improvement use but allow limited safety and abuse-prevention logging; unpaid-service terms can permit product improvement and human review. We do not make a blanket promise that Google never retains, reviews or uses submitted content. Those rules depend on the applicable service configuration and terms.
Digests appear in the app and can go to configured email or messenger destinations. They may contain names, summaries, actions, excerpts or original message lines. Choose appropriate destinations and review generated text before relying on it or sending it as your reply.
4. Voice dictation and media handling
Dictation starts when you use the microphone feature and allow microphone access on your Mac. The app records audio segments and sends them to our backend. The backend processes audio in memory, converts it to a suitable speech format and sends it to Gemini for transcription. The dictation flow does not save a recording in our application database or an audio-file archive. It returns text to the composer for review.
Dictation audio is not sent to correspondents. Sending the resulting text follows ordinary message rules. Google's terms govern its audio handling, including provider retention.
For Telegram, WhatsApp and Discord, photos, videos and files are not stored on our servers as a retained media library. Where media opening is supported, our servers may fetch a photo, video or file from the messenger and pass it to your app in transit without keeping a library copy. Telegram downloads use temporary memory buffers; WhatsApp media is fetched and streamed to the app. The current Discord integration records attachment descriptions and file names, rather than providing a general attachment download or upload service.
Sending an attachment through a supported feature transfers it to the messenger. Telegram uploads can use a temporary file on our server while delivery is attempted. The normal delivery path attempts to remove that file after completion or failure. An interrupted process, an early validation failure or failed cleanup can leave a temporary copy requiring removal; this is not a permanent media backup.
Transcribing a voice message or video requires the backend to receive audio, or extract the speech track, before sending audio to Gemini. This temporary processing is separate from keeping attachments as files. Your Mac can cache opened media and save files according to your saving choices and settings.
Transcription text, attachment metadata and small account profile pictures can be stored even though an attachment file is not. An expired, removed or protected attachment may no longer be retrievable from its messenger.
Channels that are not yet available may handle media differently; we will update this policy before they launch.
5. Information on your Mac
Electron's local app profile stores Firebase authentication state, inbox snapshots, cached message text, chat selections and scroll positions, notification preferences, storage settings and download progress to keep you signed in and show conversations quickly.
Viewed Telegram media can be cached in IndexedDB on your Mac without a separate manual save. The app also supports a local download folder and automatic saving rules. Current defaults enable saving new media from Saved Messages or self-chats, subject to the configured size limit. You can change the scope, limits and folder in Settings. The default folder is Texterry inside your Downloads folder.
Account deletion clears inbox snapshots but does not erase every cache, preference, export or downloaded file. Use storage controls and your Mac's file tools to remove local copies. Access to your unlocked Mac or its backups can expose local information.
6. Device information, diagnostics and usage
The app and backend use Sentry for crash and error reporting and performance tracing. Events can include device or operating-system information, app or environment details, stack traces, request context, URLs, timing information and technical events leading to an error. This is telemetry, used to operate and repair Texterry.
Specified Sentry reporting paths remove authorization and API-key headers. This scrubbing is limited: it does not guarantee removal of all identifiers, message content, URL parameters, request details or information included in an error. Avoid including secrets or unnecessary conversation content in support reports.
Production backend request logs record the remote network address, request URL and method, time, status, referrer and user-agent information. The address observed can depend on network proxies. Application logs also include operational identifiers, connection errors and, in some Telegram paths, phone numbers, authentication-related values and short message excerpts. These logs are separate from the message database.
Business mode includes team analytics derived from conversation timing, sent-message counts, handled conversations, assigned contacts, deals, completed tasks and pipeline changes. Owners and admins can see team performance; ordinary members receive their own performance row. These are workplace activity records. Sentry telemetry and Business-mode analytics constitute collection and use of information even though we do not use advertising analytics or cross-site tracking SDKs.
7. Payments and support
Paid Plus and Pro plans are coming soon. When you choose one, Creem handles payment as merchant of record. We receive billing references and subscription information, not your full payment-card details. Records can include your account and email, Creem customer and subscription identifiers, checkout references and URLs, plan, status, billing-period dates, cancellation details and team seat counts. A billing-event ledger records event identifiers and processing status to prevent duplicates.
Creem may collect payment, invoicing and tax information directly under its own notices. Account deletion is not a substitute for managing a subscription, and our deletion routine does not itself cancel a Creem subscription. Check your billing controls and contact support if necessary.
Support receives your email and submitted questions, screenshots, account details or diagnostics. Include only what is needed and remove other people's private content where possible.
8. How we use information and our legal bases
We use information to authenticate you, connect messengers, display and send messages, synchronize the inbox, run enabled AI actions, deliver digests, organize CRM records, operate team workflows and exports, administer plans, answer support requests, secure the service and diagnose failures.
Where the GDPR or UK GDPR applies, the appropriate legal basis depends on the purpose and our role:
- Contract: processing necessary to provide your requested service, including account administration, messenger connections, inbox operation and paid-plan administration.
- Legitimate interests: proportionate security, abuse prevention, troubleshooting, reliability and handling support or legal claims, subject to balancing those interests against your rights. You may object to processing on this ground.
- Consent: where applicable law requires it for optional processing or access to a device feature. You can withdraw consent and change relevant controls. Choosing an AI mode is a product instruction; the legal basis must also account for other people's information.
- Windows waitlist consent: using your submitted email for one notification when Texterry for Windows is available. You can withdraw this consent and request deletion by contacting support.
- Legal obligations: processing required for applicable tax, accounting, regulatory duties and lawful requests.
- Business customer instructions: when we process customer-controlled records on a business's behalf, the business establishes its legal basis. Our authority to process those records comes from its lawful instructions and applicable processing terms.
9. Sharing and service providers
We use service providers to run Texterry and share information needed for their function. Providers' roles and legal obligations differ; a payment merchant, messenger or CRM may act as a separate controller for some processing.
- Google Firebase: authentication, guest identities, token validation and email-link sign-in. It receives authentication identifiers, email when supplied, tokens and authentication-related technical information. The current implementation asks Firebase to send the sign-in email.
- Google Gemini API: generation and transcription. It receives relevant text, prompts, conversation or CRM context, style examples and audio as described above, and returns generated results.
- Sentry: crash, error and performance diagnostics. It receives technical event data, potentially including personal information in event context, subject to the limited scrubbing described above.
- Resend: transactional email delivery. The implemented Resend path sends configured digest emails, including destination address, subject and body. Our email services also support sign-in communications, currently delivered through Firebase.
- Cloudflare: website and network delivery, app downloads and waitlist hosting. It receives IP addresses and request information; its D1 database stores submitted waitlist email addresses, the platform value and signup timestamps.
- Creem: payments as merchant of record once paid plans launch. Checkout uses plan, customer and, where relevant, team and seat information; Creem returns subscription, checkout and customer references.
- Our hosting providers: servers, databases and infrastructure. These systems handle account records, stored text, connection credentials, CRM records and operational data described here.
- CRMs you connect: HubSpot, Pipedrive, Attio, Zoho and Freshsales receive records covered by the connected sync. Sync can push changes out and pull updates back; retries can continue after an initial failure.
- Google Fonts: the website requests font styles and files from Google's servers, disclosing network and browser request information to Google.
HubSpot maps contact names, email, phone, company and lifecycle stage. Pipedrive maps person and organization details, an associated deal and stage or status. Attio, Zoho and Freshsales map names, email, phone and an associated deal and stage. We store external identifiers and sync status. These mappings do not export every message, note, task or deal field.
Telegram, WhatsApp and Discord receive messages you send through their connections. Team members receive information available within their authorized workspace. Configured digest destinations receive digest content. We may disclose information with your consent, to comply with lawful requests, protect rights or investigate abuse, or in connection with a merger, acquisition or transfer of the business, subject to applicable privacy duties.
10. No sale, resale or advertising use
We do not sell or resell your personal information or message content, use your data for advertising, run ads, or share information for cross-context behavioral advertising. Provider disclosures are explained above.
11. International transfers
Innolope LLC is a United States company. Providers may process information in the United States and other countries with different privacy laws; an endpoint does not establish every copy's location.
Where EEA or UK transfer rules apply, a transfer must have an appropriate legal mechanism. Depending on the recipient and circumstances, this may involve an adequacy decision, approved standard contractual clauses, a UK addendum or international data transfer agreement, and required additional protections. This policy does not certify any recipient's participation in a transfer framework. Contact us about locations and safeguards applicable to your data, including how to request a copy of relevant safeguards.
12. How long information stays
Retention depends on category, purpose, account or team relationship and legal requirements. Turning a feature off does not erase past results. Application retention differs from backup and provider retention:
- Account and connection records: generally while the account or connection is active, followed by the deletion process below. Disconnecting stops the connection; it does not necessarily delete all previously synchronized records.
- Message text, conversations and CRM records: retained to provide your inbox or workspace, subject to deletion requests, account deletion and shared-team rules. No universal automatic expiry period is stated for these records.
- Assistant digest inputs: the database is configured to expire buffered digest items 30 days after creation.
- Generated digest history: configured to expire 90 days after creation. Copies delivered to email or messengers follow those destinations' retention.
- Style descriptions, lead summaries and voice-message transcription text: retained as feature records or cached results; personal style and transcription rows are included in account deletion. There is no uniform expiry period for all these results.
- Dictation audio: processed as request audio in memory, without an application audio archive. Dictated text stays in the composer or, if sent, follows message retention. Provider retention is separate.
- Email sign-in requests: configured to expire one hour after creation. Database expiry runs asynchronously, so this is not an exact erasure deadline.
- Payment records: customer, checkout and subscription records can remain after account deletion for billing administration and applicable legal recordkeeping. The billing-event deduplication ledger is configured to expire events 90 days after receipt; this does not limit invoice or subscription retention.
- Diagnostics, logs and support: retained according to operational needs, provider settings and applicable obligations. A fixed period for these systems is not established by the app's deletion flow. Contact us about a particular record.
- Windows waitlist: kept for the requested availability notification unless you withdraw consent or request deletion. It is not repurposed for other marketing.
- Local copies: caches follow app storage controls and eviction; downloaded files and exports remain until removed on your Mac. Your device or cloud backups may retain copies.
- Backups and legal records: if backups contain your information, active-service deletion does not necessarily erase them immediately. Copies may remain until rotation or separate deletion. Information may also need to remain for a legal duty or claim. Ask about applicable arrangements; no fixed backup-erasure period is promised.
Configured expiry periods require the relevant database indexes and expiry mechanisms to operate. They describe eligibility for scheduled deletion, rather than a guaranteed deadline for every physical copy.
13. Disconnecting and deleting an account
You can disconnect a messenger at any time and delete your Texterry account in Settings. Deletion tears down your connections and removes your account and associated personal data, including personal assistant records. Workspaces you own without other members are removed with their conversations, messages, CRM records and related data. The process also removes the Firebase login when the authentication service is available.
An owner cannot delete their account while they own a team with other members. The app asks them to resolve those memberships first. If you belong to another person's team, deletion removes your membership and hands over relevant open work, while shared-team records remain. A former-member record can retain your identifier, name, email and handover information to keep past work attributable.
Deletion does not erase messages from original messengers, other people's accounts, connected CRMs or delivered digest destinations. It does not erase downloads from your Mac or cancel an external subscription. Billing rows, temporary sign-in requests, diagnostics, logs and applicable legal records may outlast deletion, as described above.
If deletion fails or is interrupted, do not assume it completed. Retry through the app or contact support; no universal completion or provider-cleanup deadline is promised.
14. Security
Messenger sessions are encrypted at rest, traffic between the app and our servers is encrypted in transit, and the Mac app is signed and notarized by Apple.
Authentication tokens and ownership or team-access checks limit access to account and workspace functions. Administration and exports have additional role checks. The backend must use connection credentials and process message text to provide requested features; encrypted session storage does not mean that text is inaccessible to our service.
No system is perfectly secure. Contact us about suspected compromise. We will notify affected people and authorities of personal-data breaches as required by law.
15. Your choices and privacy rights
Settings control assistant modes, replies, digests, local storage, connections and account deletion. Depending on location, legal grounds and our role, your rights may include:
- Access: obtain information about processing and a copy of personal information we hold about you.
- Correction: ask us to correct inaccurate or incomplete information.
- Deletion: request erasure, subject to applicable exceptions and shared-record responsibilities.
- Export or portability: receive eligible information in a usable format and, where applicable, have it transferred to another controller.
- Objection: object to processing based on legitimate interests and to direct marketing if applicable.
- Restriction: ask us to limit processing in circumstances provided by law.
- Withdrawal of consent: withdraw consent for processing that relies on it, without affecting processing that was lawful before withdrawal.
- Complaint: complain to a supervisory authority, including where you live or work, or the UK Information Commissioner's Office where appropriate.
Write to support@texterry.com with your request and enough context to locate the information. Do not send passwords, session tokens or card details. We may proportionately verify identity or an authorized agent's authority before releasing or changing information.
Where the GDPR or UK GDPR applies, the ordinary response period is one month. A lawful extension of up to two further months may apply for complexity or multiple requests, with notice within the original period. Where the CCPA applies, requests to know, delete or correct generally require a response within 45 calendar days, with a permitted extension of another 45 days and notice. Identity verification and other timing rules apply as provided by law. These are statutory response periods, not guarantees that every technical erasure finishes within them. See the ICO's timing guidance.
Owners and admins can export clients, deals and tasks as CSV, or CRM records and conversations as JSON with optional stored messages. This does not export every account, log or provider record. Contact us for personal-data requests. We do not discriminate for exercising applicable rights. Lawful limits can protect other people or required records; we will explain an applicable refusal.
16. California privacy notice
CCPA rights apply where its coverage requirements are met. During the preceding 12 months, the following categories may have been collected or disclosed for business purposes through features used during that period; paid-plan collection applies when available.
- Identifiers and contact details: account IDs, names, email, phone, usernames and network addresses; disclosed as needed to authentication, hosting, delivery and diagnostic providers, connected messengers or CRMs, and authorized members.
- Customer and commercial records: account, team, plan, checkout and subscription details, CRM contacts and deals; disclosed to hosting providers, Creem for billing, connected CRMs and authorized members as applicable.
- Electronic activity: request details, device and app context, errors, performance events and service or team activity; disclosed to infrastructure and diagnostic providers and through authorized team analytics.
- Audio and electronic content: dictated audio, audio for transcription, text, conversation details and support content; disclosed to Gemini for the processing described above, relevant hosting or delivery providers, messengers and configured recipients.
- Professional information and inferences: company and contact information, work assignments and metrics, generated summaries and style descriptions; disclosed through relevant AI, CRM, hosting and team features.
- Sensitive personal information: account-access credentials and private communications, and sensitive facts contained in them; handled to operate your connections and requested functions, including the separate AI processing in section 3.
Sources are you, connected services, teammates, device interactions and provider responses. Purposes, recipients and retention appear above. Transcription processes speech without creating an identification voiceprint.
We have not sold personal information or shared it for cross-context behavioral advertising during that period and do not do so now. We do not knowingly sell or share children's information or use sensitive information for ads. California residents can exercise applicable access, correction, deletion, sensitive-information and authorized-agent rights through the contact route above, without discrimination. You do not need to create an account just to contact us. See the California Attorney General's CCPA guidance.
17. Notes for EEA and UK users
For our account administration, website, waitlist, support and service operations, Innolope LLC is the controller deciding the purposes and means of processing. For business-controlled customer records, our role can instead be processor. The bases in section 8 and transfer considerations in section 11 apply where the GDPR or UK GDPR covers processing.
Messages may contain health, political, religious or other special-category information. An ordinary legal basis may not be enough to authorize processing it. Business customers must consider any additional condition required by law before placing it in Texterry or enabling AI.
Any duty to appoint an EEA or UK representative depends on applicable law and processing circumstances. Contact us about the appropriate local contact. You may complain to your supervisory authority independently of our support process.
18. Other people's information and Business mode
Correspondents may not have Texterry accounts. Consider their expectations before enabling AI, shared access, exports or external digest delivery. Permission to read a message does not automatically authorize every onward use.
A business managing customers generally acts as controller for those records; we act as processor where we handle them on its behalf under its instructions. The business is responsible for notices, a lawful basis, member access and customer requests. We separately control our account, billing and security processing. If your use requires a data processing agreement, contact us to discuss required terms before connecting customer data.
Team-scoped connections and CRM records are accessible according to membership and roles; a private connection is not shared merely because you belong to a team. Owners manage membership, and owners and authorized admins can export records and see team performance. Turning Business mode off hides its interface rather than deleting the workspace or revoking membership. Leaving a team does not delete its history. Correspondents and employees should contact the business controlling the record as well as us where appropriate.
19. Third-party messengers and integrations
Telegram, WhatsApp and Discord have their own policies, retention and account controls. Texterry sends through your connected account or authorized bot connection. Replies configured as an assistant can identify themselves as such.
Connected CRMs, payment checkout, email recipients and linked websites also have their own policies. Disconnecting Texterry does not erase information they already received. Review connection permissions and use providers' controls for requests about their copies.
Read the Telegram privacy policy, WhatsApp privacy policy and Discord privacy policy for their practices.
20. Children
Texterry is not intended for children under 16, and we do not knowingly collect their information as users. Contact support if a child supplied information directly. Businesses must consider age-related obligations for contacts; a child's mention in a conversation differs from using Texterry.
21. Website visitors, Windows waitlist and storage technologies
Our website consists of static files served by Cloudflare. It sets no cookies, stores nothing in your browser, including no localStorage, and has no analytics or advertising trackers. Cloudflare receives IP addresses and ordinary request information to deliver the site and keeps normal server logs. The website loads fonts from Google Fonts, so your browser makes requests to Google that disclose network information such as your IP address.
The optional “Texterry for Windows” waitlist asks for your email address. If you submit it, we store only that email, the platform value “windows” and a timestamp in a Cloudflare D1 database. We use it to email you once when Texterry for Windows is available. It is not used to market anything else and is not shared with other businesses for their own purposes. Cloudflare hosts the database for us.
Submitting the form is your consent to that specific notification. Joining the list does not create an app account or subscribe you to unrelated messages. You can withdraw consent and request deletion at any time by emailing support@texterry.com.
The app uses local storage, IndexedDB and authentication state for functionality as described in section 5. These are similar storage technologies, although they are not advertising cookies. External checkout, authentication or CRM pages may have their own cookie practices. Clearing local storage can remove preferences or sign you out without deleting server records.
22. Do Not Track
Texterry does not change its processing in response to browser Do Not Track signals. Our no-sale and no-advertising-sharing commitments apply regardless, including for California opt-out preference signals. Contact us about privacy rights.
23. Changes to this policy
We will update this date when the policy changes and notify you in the app or by email about meaningful changes. An updated policy does not replace any legally required new consent or notice.
24. Contact
For privacy questions, requests, concerns or information about business processing terms, contact us:
Innolope LLC, 8 The Green, Suite A, Dover, DE 19901, United States
Email: support@texterry.com