Privacy Policy

Last updated: October 17, 2026

Grow One works because you trust it with your email, your calendar, your voice, and your notes. This policy explains — plainly — what we collect, why, who processes it, how long we keep it, and how you delete it.

A Spanish translation is available at privacy-es.html. If the two ever differ, this English version controls.

Who we are

Grow One is made by Grow Forward MC LLC, a limited liability company based at 2950 Clay St #201, San Francisco, CA 94115, USA. Grow One is an AI personal assistant iOS app: it helps with email, calendar, tasks, meeting notes, thought capture, and topic digests, with an optional Social Agent for social publishing and insights.

For anything in this policy, contact us at ai@growfmc.com.

What we collect

Everything below is collected only after you grant it — each source is a separate, explicit connection or upload you make in the app.

Account and sign-in

You sign in with Google, with Sign in with Apple, or with your email address and a one-time code we email you (some longer-standing accounts use a password instead). From that sign-in we receive and store your email address, and your name when Google or Apple share it; Google also passes a link to your profile picture, which our authentication provider, Supabase Auth, keeps. When you later connect Gmail or Google Calendar, the userinfo.email and userinfo.profile scopes give us that same name and email. If you use Apple's "Hide My Email", we only ever see the relay address Apple gives us. During onboarding you answer a few plain questions about yourself or your business (your name, what you do, how you want the assistant to behave); those answers are stored with your account and are the basis of what the assistant knows about you.

Gmail (via Google OAuth)

If you connect Gmail, we request the gmail.modify scope. We use it to read your inbox so the assistant can classify and summarize your email, and to send replies — but only replies you approve. Grow One never auto-sends email unless you explicitly opt in to that behavior. AI-drafted replies are stored in our database, not in your Gmail drafts. Stored email follows the Email rules under "How long we keep things" below.

Google Calendar (via Google OAuth)

If you connect Google Calendar, we request calendar.events (read and write events) and calendar.calendarlist.readonly (to list your calendars). Events are created only when you confirm them.

Outlook and Microsoft 365 mail (via Microsoft sign-in)

If you connect an Outlook, Hotmail or Microsoft 365 mailbox, we request these Microsoft permissions: Mail.ReadWrite (read your mail so the assistant can classify and summarize it, and mark messages read), Mail.Send (send replies, only replies you approve), User.Read (your name and email address) and offline_access (stay connected without asking you to sign in again). We do not request access to your Outlook calendar, contacts or files. Grow One never auto-sends email unless you explicitly switch on Auto-Reply, and AI-drafted replies are stored in our database, not in your Outlook drafts. Data we receive from Microsoft is used only to provide these features to you: it is never used for advertising, never sold, never read by people except with your consent, for security, or as the law requires, and never used to train AI models. Disconnecting Outlook in the app deletes the stored connection; you can also remove Grow One from your Microsoft account under Privacy → Apps and services.

Voice and meeting audio

Voice recordings and meeting audio you capture are transcribed to power notes, tasks, and recall. Meeting audio is stored with its note — deleting the note also deletes its recording.

Voice recognition data (voiceprints)

If you use speaker-labeled meeting notes, Grow One stores short voice reference clips ("voiceprints") of speakers you name — including, optionally, your own voice via "This is me." Because they identify a person by their voice, we treat them as sensitive. They are used for one purpose only: to recognize and label who said what in your future meeting transcripts. They are created only when you name a speaker, stored in a private bucket readable only by our servers (never by any app client), never used for advertising or for identifying anyone outside your own meeting notes, never shared, and never sold. When speaker identification runs on a recording, the reference clips of the speakers you named are sent to OpenAI's transcription API for that recording only, under the same no-training API terms as everything else we send.

Deleting them: you can have the app forget any speaker (which deletes that speaker's voiceprint), and deleting your account permanently removes every voiceprint — the audio files and the database records. You can also email ai@growfmc.com to have them removed.

Photos

Photos you upload — for chat vision, schedule scanning, or the photo library your website pages and social agent share — are processed for the feature you used them in. Photos you add to the photo library or to a website page are stored resized to web size, never at full size.

Contacts

Contacts you create in the app, or that are derived from your email and meetings, are stored to build your personal contact history.

Thoughts and notes

Ideas and notes you capture are stored so the assistant can recall them when you ask.

What the assistant remembers

To answer you from your own information, the assistant keeps a memory of what you tell it and what you connect: facts you confirm about yourself and the people you mention (for example a partner's name, your kids' school, a client's timeline), summaries of your chats, and short excerpts of your notes, meeting transcripts, Notion pages, and email context. Much of this text is also stored as embeddings — numeric representations that let the assistant find the relevant passage when you ask a question. Memories and facts are visible in the app under Settings, where you can correct or delete them; deleting a note, page, or your account deletes its embeddings with it.

Notion

If you connect Notion, you choose in Notion's own screen which pages to share with Grow One. We read the text of those pages and store it, with embeddings, so the assistant can answer from them; the copy is refreshed about every 30 minutes and removed when you un-share a page, disconnect Notion, or delete your account. With your permission the app also writes to Notion: it saves your meeting notes and trip guides to a page you pick, and creates pages or database rows only when you ask and confirm on screen. It never deletes anything in Notion.

Slack

If you connect Slack, Slack's own screen shows what Grow One is asking for and you choose the channel for your alerts. We store the connection keys Slack gives us, your Slack workspace, your Slack member ID and display name (shown as "Connected as"), and the name of that channel. The alerts you switch on are sent to Slack: a new lead's name, email and message, or a payment's amount and payer; contacts you marked Personal are never posted. When you or a teammate ask Grow One a question in Slack, the question and the answer are saved in that project in Grow One, like any project chat, and the answer is posted in Slack where the people in that channel can see it. When you ask Grow One what is happening in a channel, it reads recent messages from channels you can see at that moment and does not store them; it never reads your direct messages. To write an answer, that text is processed by our AI providers, listed below; nothing from Slack is used to train AI models. Messages other people wrote in Slack are never saved to a project's memory. Disconnecting (or deleting your account) removes the keys from our servers and cancels Grow One's access at Slack. On a Business team, your manager can switch Slack off for you, which also disconnects it.

Stripe

If you connect Stripe, you create a restricted API key in your Stripe account and paste it into the app. That key lets us read your customers, invoices, payments, payouts, and disputes, and create customers and invoices when you ask and confirm; it cannot move money. We check for new Stripe events about every 15 minutes and keep a ledger of them so the assistant can tell you who paid, who owes you, and when a payment fails. Your customers' names and emails from Stripe may be used to recognize them in your email and contacts. Disconnecting Stripe deletes the key and the payment history stored with your account.

RevenueCat and App Store Connect (your own apps)

If you publish apps and connect RevenueCat, you create a read-only API key in your RevenueCat project and paste it into the app (or, where available, approve a read-only OAuth connection). That key lets us read your app's aggregate charts: revenue, subscribers, trials, churn, new customers, and breakdowns by store, country and plan. It cannot see your app's individual users and cannot change anything. We take one snapshot per day and keep a daily history so the assistant can answer questions, notice a week-over-week shift, and write a weekly read. Optionally you can also connect an App Store Connect API key with the Sales and Reports role; it lets us read your daily download counts, which app versions had activity, and your public App Store reviews — nothing else about your apps or your Apple account. You can add markers (a release, a price change, a campaign) to a timeline so the assistant can compare before and after. Disconnecting deletes the keys, the stored daily history, and the timeline for that app.

Social account data (via Zernio)

If you connect social accounts (Instagram, Facebook, LinkedIn, and others we support) through our publishing partner Zernio, we access your own posts, comments on your posts, and follower statistics, and we publish the posts you approve. We do not read the private messages of those social accounts.

WhatsApp Business inbox (via Zernio)

If you connect a WhatsApp Business number (Creator plan), customer conversations on that number — messages and any media — reach us through Zernio so the assistant can summarize threads and draft replies for you to send. Those conversations belong to you and your customers; we process them only to provide the inbox features, we never send a message you did not approve, and they are deleted when you delete your account. Your WhatsApp app itself is unaffected.

Your website pages (landing pages)

On every plan you can build one-page websites in the app (up to 3 published pages on Essentials, 10 on Creator, 25 on Business shared by the team). A page is public on the internet the moment you publish it, at an address you choose under pages.growone.io, and anyone with the link can open it. We store the page itself (its text, design and address), the photos you add to it (in a public storage bucket, so the page can show them), an optional preview image and description for when the link is shared, daily visit totals (see "Visits to your pages" below), and your conversation with the website assistant. Every published page carries a small "Report" link so visitors can report abuse, a "Privacy" link to this policy (and, unless you turn it off, "Made with Grow One"). A report stores the page's address, the reason the visitor picked, what they wrote, their email if they chose to add one, and a scrambled (hashed) form of their internet address used only to stop repeated reports — we use it only to review the page.

What the website assistant knows, and does not know. Pages are written and edited by DeepSeek, or by Anthropic's Claude when DeepSeek is unavailable (see the provider table below). To build a page, the builder receives only a short brand guide we assemble from what you have given the app: your or your business's name, whether the account is a business, solo professional or personal account, the business type, brand colors and tone of voice you set (or, from your social strategy if you built one, its brand colors, fonts, visual style, tone of voice, business overview and products and services), the colors, fonts and visual style read from a website you asked us to match (our server fetches only the one page whose address you paste, and keeps a short description of its look), your phone number and website from your profile, the text of your own website's homepage, its logo and its real photos (from the website in your profile — the homepage and up to four of its own pages; the text is refreshed at most once a week, the photos are taken once, stock photos are skipped), confirmed facts about your business, your contact details, your work role, whether you have a logo, the design instructions we (Grow One) write for the builder, the page being edited (including its logo and photos), files you attach in the website assistant or keep with a page under Data (their text, for example a menu, a price list or your résumé), a description of a photo you attach as inspiration (written by Google Gemini), the address of a photo you ask to put on the page, the form answers you choose to turn into a newsletter (without email addresses or phone numbers), and the messages you type in the website assistant. That is the whole list. The builder never receives your email, calendar, meeting notes, contacts, thoughts, chat history with the main assistant, the assistant's memories about you, documents you upload elsewhere in the app (in Projects or the main chat), your location, or anything else about other people. You can see the exact brand guide the builder gets at any time under About your business / Website information inside the website workspace.

Visitors to your pages (people who fill in your forms)

If your page has a contact or sign-up form, a visitor's submission is stored for you: the name, email, phone and message they typed, any extra fields the form asked for, the page it came from, the visitor's browser type, and a one-way hash of their IP address that we keep only to stop spam. To filter spam we also check that the email's domain has a real mail server (a DNS lookup of the domain, not of the visitor) and refuse known throwaway-email domains. Submissions appear under Leads or Responses in the app. If you are a business and chose to, they also reach your Sales & Customer Hub — and any team members, Slack channel or push notification you set up for lead alerts — where the assistant may draft a reply for you; a reply to a form submission is never sent automatically. Visitors are not tracked across sites: pages contain no advertising or tracking scripts and no third-party analytics (how we count visits is explained just below). Submissions stay in Leads until you delete them or your account; deleting a page keeps its submissions, so you can still answer them. A visitor who wants their submission removed can ask the page owner, or write to us at the address below and we will remove it. We keep submissions for the page owner, use them only to provide the service to them, and follow their instructions about them. Under every form on a published page, a line names the business your details go to and links to this section. That business uses your submission to reply to you and handle your request; anything more is up to it and the laws that apply to it, so ask it directly if you have questions.

Visits to your pages (visit counts)

When someone opens a published page, we add one to that page's visit totals for the day: where the visit came from (for example Instagram, Google, a QR code, or typed directly — read from the page the visitor came from, a tag in the link, or the app they opened it in; we keep only the kind of source, never the address itself), whether it was a phone, tablet or computer (read from the browser's name), the city and country our hosting provider (Vercel) estimates from the connection, and the hour. We keep only those daily totals, in de-identified form, and will not try to re-identify any visitor. We set no cookies, store no identifier on the visitor's device, and keep no IP address or anything else that identifies a visitor, and we never follow visitors to other sites. Our hosting provider uses the visitor's internet address only to deliver the page and estimate the city, and holds its request logs briefly (see "How long we keep things"). Pages load their fonts from Google Fonts, so Google receives the visitor's internet address when a font loads (see "Service providers"). Known bots, link previews and the page owner's own visits from the app are not counted. If the visitor's browser sends a "Do Not Track" or Global Privacy Control signal, the visit is not counted at all. Only the page's owner, and the people in their business account who work on the website, see the totals in the app. When the owner asks the assistant about visits, or gets the Monday briefing, the totals for their pages are sent to our AI provider to write the answer (see "AI providers that process your data"). The totals are deleted with the page or the account.

Your Grow One subscription (Apple and RevenueCat)

You subscribe to Grow One through Apple's App Store. Apple processes the payment; we never see or store your card. We use RevenueCat to manage the subscription: when you start a trial, buy, renew, change or cancel a plan, Apple tells RevenueCat and RevenueCat tells us which plan is active, so the app can unlock it. RevenueCat receives your Grow One account ID (a random identifier, not your name or email) and the purchase details Apple provides (plan, price, dates and the country of your App Store account). We store only your current plan and its status with your account. You manage or cancel the subscription in your iPhone settings.

Business plans are billed on the web by Stripe, our payment processor. Stripe processes the card; we never see or store it. Stripe receives the subscriber's email, the business's and the subscriber's Grow One account identifiers, the number of places on the plan and, where tax applies, a billing address, and tells us the subscription's status so the app can unlock the plan. We store that status, the billing period, renewal date and number of places, the Stripe customer and subscription identifiers, and a log of Stripe billing events. Receipts, card changes and cancellation are handled on Stripe's billing page, opened from the Team panel.

Device and technical data

We keep this deliberately minimal. Grow One contains no third-party analytics, advertising, or tracking SDKs — no ad identifiers, no cross-app tracking, no behavioral profiling. What we do process: a push-notification token so your iPhone can receive the alerts you asked for, and standard server logs (IP address, timestamps, request paths) kept for security and reliability. When a screen of the app fails, the app sends us a short error report: what broke and where in our code, which screen, the app version, whether it was the iPhone app or the web, and a random identifier for that app install (not tied to your account). It is designed not to contain message text, anything you wrote or account details — emails, long numbers, ids and quoted text are removed before it is sent — is not linked to your account, and is deleted after 30 days. iPhone crash reports also reach us through Apple if you've chosen to share analytics with app developers in your iOS settings — not through any SDK we added. The one library we did add is RevenueCat's purchases library, included only to process your Grow One subscription through the App Store; it does not track you across apps, show ads, or build a profile of you.

Location (iPhone only)

The first time you open the iPhone app after this feature ships, iOS asks once whether Grow One may use your location. If you allow it, City mode turns on; if you decline, nothing is read or stored and iOS will not ask again. You can change this any time in Settings → Privacy & Data, in one of two modes. City: when you open the app, your iPhone converts its position to a city name using Apple's own geocoder and we store only the city, region and country — never coordinates. Precise: the same, plus your exact coordinates are sent with a chat or search request so "near me" means where you are; those coordinates are used for that one request and are not stored. Turning location off deletes the stored city. Nothing is read in the background, and web browsers never have access to location.

Google API Services — Limited Use disclosure

AI providers that process your data

To generate features, relevant content is sent over encrypted connections to the AI providers below, which process it solely to deliver the feature you used. We use their commercial APIs. Per their API terms, Google, OpenAI and Anthropic each state that data submitted through their APIs is not used to train their models; we rely on those contractual commitments and do not grant any provider permission to train on your data. DeepSeek is the exception: under its own terms it may store data on servers in China and use it to improve its services, which is why it only ever receives what the website builder needs — the short brand guide, your website instructions and anything you choose to attach — never your email, calendar, contacts or memories. Attach only files you are comfortable sending to it. Stability AI (the photo fixes in Studio) is the other one to know about: its terms let it use what it receives to improve its models unless the account opts out — we have opted out for our account, and it only ever receives the photo you are fixing and the words you typed for that fix, never anything else. Their processing is governed by their own terms and privacy policies, linked below.

Meeting recordings and transcripts are never used to train AI models — not by us, and not by our providers. Audio is sent for transcription and the resulting transcript is sent for summarizing, both on paid commercial API tiers whose terms exclude training on submitted content. We do not use your recordings, transcripts, notes, or email to train any model of our own, and we have no model of our own to train.

Some features have alternate providers wired in that are not enabled today (Groq for speech-to-text, Deepgram for read-aloud, Perplexity and Tavily for web search, Moonshot's Kimi as an alternate chat model). If we ever switch one on, we add it to this table before it processes any of your data.

The current list, with a link to each provider's own privacy policy, is also shown inside the app under Settings.

Data about other people

Some of what you connect or create contains information about people who are not Grow One users: the people you email, the people in a meeting you record, your customers on WhatsApp or in Stripe, people you mention in notes and memories, and the visitors who fill in a form on your page. For that data you decide what to connect, record and keep; Grow One processes it only on your instructions to run the features you use, never for advertising, profiling or sale, and it is deleted with your account. If you are one of those people and want something removed, tell the Grow One user you dealt with, or write to us at the address below and we will help.

Service providers (infrastructure)

We rely on a small set of infrastructure providers to run the service:

Each provider is engaged only to perform its role for us. Zernio processes your social account connections and your public social content to publish and report on your behalf; its processing is governed by its own terms and privacy policy (at zernio.com), and we do not grant it permission to use your data for anything beyond providing that service.

What we never do

• We never sell your personal data.
• We never show third-party advertising.
• We never share your data with data brokers.
• We never use your data to train AI models, and we don't grant our AI providers permission to either.
• We never make automated decisions about you with legal or similarly significant effects. The assistant sorts, drafts, suggests and — only in the modes you switch on, such as Auto-Reply for email — acts on routine tasks, and you can turn every mode off. Social posts are never published without your approval.

How we protect it

Data moves over encrypted connections (TLS) and is stored encrypted at rest by our infrastructure providers. Database access is scoped to your account with row-level security, and the most sensitive stores — voiceprints and captured thoughts — are additionally restricted so only our servers can read them, never any app client directly. Provider credentials and API keys are held server-side only. No system is perfectly secure, which is why we also make the commitment below.

If there's a data breach

If a security breach affects your personal information, we will notify you promptly — without undue delay, and within the timelines applicable law requires — and tell you what happened, what data was involved, and what we're doing about it.

How long we keep things

• Email: the full text of email alerts — and replies sent through Grow One — is deleted after 90 days. Subjects, summaries, and other stored email content (including replies you drafted) are deleted after 1 year; a travel booking keeps its subject until the trip has passed. We keep who you emailed, when, and whether you replied, so the assistant still remembers who you've talked to.
• Assistant cards and notifications (nudges, task alerts): deleted after 90 days — except the few that only remember a choice you made, like "don't ask again".
• Page reports: deleted 180 days after they are made.
• Photo library photos: a photo the social agent published is removed from our storage 10 days after publishing, unless one of your website pages shows it — then it stays until it is off the page (or the page is deleted).
• Meeting audio: kept with its note; deleting the note deletes the recording.
• Recording consent records: kept for 3 years, then deleted automatically. These record that you were shown our recording notice, and what you told us about each recording (whether it was just you, or a conversation you confirmed you had informed). They never contain any audio, transcript, note content, or the names of anyone in the conversation.
• Voiceprints: kept until you have the app forget the speaker, or until you delete your account.
• Chats, memories, and facts: kept until you delete them in the app or delete your account.
• Notion index: refreshed on each sync; a page's copy is removed when you un-share it, disconnect Notion, or delete your account.
• Subscription: your current plan and its status are kept with your account; RevenueCat and Apple keep purchase records under their own policies for billing and tax. For a Business plan, its status and Stripe billing log are kept with the business, and Stripe keeps payment records under its own policies.
• Outlook: the connection is deleted when you disconnect Outlook or delete your account; stored email follows the Email rules above.
• Stripe: the key and the payment ledger are deleted when you disconnect Stripe or delete your account.
• RevenueCat and App Store Connect: the keys, the daily metrics history and the timeline are deleted when you disconnect that app or delete your account.
• WhatsApp conversations: kept while the number is connected; deleted when you delete your account.
• Website pages and photos: kept until you delete the page or your account. Deleting a page removes the photos uploaded for it (a photo still used by another of your pages stays). Form submissions: kept until you delete them or your account; deleting a page keeps them. Daily visit totals: deleted with the page or your account. A deleted or unpublished page stops being served within about a minute. The website assistant's working copies of a page are kept with your account and deleted with it.
• Server and security logs: our hosting provider's request logs are held briefly for reliability; our own security log (IP address, browser type, request path, event) is kept for 90 days and then deleted automatically.
• App error reports: deleted automatically after 30 days.
• Connections: you can disconnect Google, Notion, Stripe, WhatsApp, or social accounts at any time; disconnecting removes the stored credential. The authorization itself also stays listed in your Google, Notion or Stripe account until you remove it there.
• Account deletion: immediate. Deleting your account in the app removes your data, storage files, and social account connections through a verified deletion pipeline. One narrow exception: the recording consent records described above are kept for the remainder of their 3-year window, because they are the record of what you told us about recordings you made, and we may need them to establish or defend a legal claim. Both the GDPR (Art. 17(3)(e)) and the CCPA (§ 1798.105(d)) permit retaining data for that purpose. They hold no conversation content — only that a notice was shown and what you asserted — and they are deleted automatically when the window closes. Two things that hold no personal data also remain: an anonymous monthly total of AI usage that cannot be linked back to you, and a bare deletion record (an internal id and a timestamp). Our social publishing partner Zernio is told to disconnect your accounts; it keeps its own records under its policy.

Your rights (including California)

You can exercise these directly, self-serve, in the app:

• Access: your data is visible to you inside the app — your email summaries, notes, contacts, and settings.
• Deletion: delete your account in the app and everything is removed, immediately and completely — except the narrow recording-consent exception and the short-lived server logs described above.

California residents have these rights under the CCPA/CPRA, including the right to know, delete, correct, and not be discriminated against for exercising them. We do not sell or share personal information as those terms are defined in the CCPA. For any request you can't complete in the app, email ai@growfmc.com.

How to exercise your rights

• In the app: your data is visible on each screen, every connection (Gmail, Calendar, social) can be disconnected in Settings, and account deletion in Settings removes everything immediately.
• By email: send your request to ai@growfmc.com from the email address on your account — that's how we verify it's really you. We respond within the timelines applicable law sets (generally 45 days under the CCPA and one month under the GDPR). California residents may use an authorized agent with written proof of authorization.

California "Shine the Light"

We do not disclose personal information to third parties for their own direct marketing purposes (Cal. Civ. Code § 1798.83), so there is nothing to opt out of — but you're welcome to ask at ai@growfmc.com.

Do Not Track and Global Privacy Control

We don't track you across other sites or apps, show ads, or sell or share personal information — so there is nothing for a "Do Not Track" or Global Privacy Control signal to turn off. This website sets no tracking cookies (only a theme preference stored on your own device). Published pages do listen for both signals: a visit from a browser that sends either one is left out of the page's visit totals (see "Visits to your pages").

International users and data transfers

Grow One is operated from the United States, and your data is processed and stored in the US. If you use Grow One from outside the US, you understand that your data is transferred to and processed in the US, where privacy laws may differ from those of your country. For users in the EEA, UK, and Switzerland, where the law requires a transfer mechanism we rely on recognized safeguards such as the European Commission's Standard Contractual Clauses and, where a provider participates, the EU–US Data Privacy Framework.

Legal bases (EEA/UK users)

Where GDPR-style laws apply, we process your data on these legal bases: performance of a contract (running the assistant you signed up for), consent (every data source — Gmail, Calendar, voice, photos, social — is a separate opt-in connection you can withdraw at any time), legitimate interests (securing and improving the service; counting visits to published pages, as daily totals only, so their owners can see how their pages are doing; and storing form submissions for the page owner who asked for them), and legal obligation (where law requires retention or disclosure). You also have the rights to access, rectification, erasure, restriction, objection, and portability, and may lodge a complaint with your local supervisory authority.

Children

Grow One is not directed at children under 13, and we do not knowingly collect personal information from children under 13.

Changes to this policy

If we change this policy, we'll announce it in the app and on this site — and for material changes, we'll tell you before they take effect and refresh your consent where the law requires it. The "Last updated" date at the top always reflects the current version.

Contact

Questions about privacy at Grow One: ai@growfmc.com
Grow Forward MC LLC · 2950 Clay St #201, San Francisco, CA 94115, USA

See also our Terms of Service.