Privacy Policy
What GrowthFabric collects, why, who helps us process it, how long we keep it, and your rights.
Draft for legal review. Not yet in force: our lawyer reviews every contract before it applies.
Last updated 2026-10-14
Items in brackets, like [company legal name: to be confirmed], are open facts we have to settle with our lawyer before this document takes effect.
The short version
| What | Why | Kept |
|---|---|---|
| Website visits: your IP address and browser details in our hosts' logs | To serve the site and keep it secure | Per our hosts' log retention [log retention: to be confirmed] |
| Your account: name, email, workspace, sign-in | To give you access | While your account exists |
| Platform access: the sign-in tokens Google or Meta give us | To read the accounts you connected | Until you disconnect or ask us to delete them |
| A small stored copy: account structure, daily account-level totals, findings with the minimal rows behind them | So agents can compare days and remember what they found | While the account is connected; deleted when you disconnect it or delete the workspace |
| What you ask agents and what they answer | To run the conversation and show you its history | While your workspace exists |
- We never sell personal data or use it for advertising.
- We do not train AI models on your data. Our AI provider's terms say it may not train on it either.
- The website uses no analytics or advertising cookies. It sets one session cookie that remembers your language.
1. Who we are
[company legal name: to be confirmed], registered at [registered address: to be confirmed], operates growthfabric.ai and the GrowthFabric app and tools ("GrowthFabric", "we").
- For our website, our customers' accounts and our own business contacts, we decide how personal data is used: we are the controller.
- For the data in the ad and analytics accounts our customers connect, we act for the customer, who is the controller. We are their processor, under the Data Processing Addendum. If your data reaches us that way, the customer that connected the account is your first point of contact.
Contact for privacy: dialog@growthfabric.ai. [EU and UK representative, if appointed: to be confirmed]
2. What we collect
When you visit the website
Our hosting providers (Vercel and Cloudflare) record standard request details, such as your IP address, browser type, the page asked for and the time, to deliver the site and protect it from abuse. The website runs no analytics, advertising or tracking scripts.
When you email us
Your email address, your name if you give it, and what you write.
When you use the app
- Account details: your name, email address and password. Passwords are handled by our sign-in provider (Supabase) and stored only in hashed form; we never see them.
- Workspace details: your workspace, its members and their roles, and settings such as where to send your daily brief (a Slack channel link or an email address).
- What you do in the app: questions you ask agents and their answers, findings you dismiss or act on, approvals, and an audit log of who did what.
When you connect an ad or analytics account
You sign in with Google or Meta and approve our access on their consent screen. We then receive:
- Access tokens. The platform issues them so we can read the account. We encrypt them before storing them and decrypt them only for the moment a read needs them.
- What the account contains, read when an agent needs it. Of that, we keep only a small stored copy:
- the account's structure: account and campaign names and IDs, status, currency and time zone;
- daily totals at account level, such as spend and conversions per day;
- findings, with the minimum evidence needed to show why each was raised.
Findings keep the minimal rows that show why each was raised, which can include a campaign, a keyword or a search term. We do not keep campaign, keyword or ad-level history, raw platform responses, audience lists, images or videos. Agents read those live when they need them and drop them after use.
Ad platforms give us figures about campaigns, not about the people who saw the ads. We do not receive or store data about individual members of the public.
When you use the free local server
The free local server runs on your own computer. It keeps its Google access token in a file on your computer and sends nothing to us. We receive no data from it.
3. Why we use it, and our legal bases
| Purpose | Legal basis (EU and UK GDPR) |
|---|---|
| Providing the website, the app and the agents' work | Contract with your business; our legitimate interest in running the service for the people who use it |
| Security, abuse prevention, debugging | Legitimate interest in keeping GrowthFabric and its users safe |
| Answering your emails and support requests | Legitimate interest; contract where you are a customer |
| Telling you about changes to the service or these documents | Contract; legitimate interest |
| Meeting legal obligations, such as tax records once billing exists | Legal obligation |
We do not use your data for advertising, and we do not send marketing email without your consent where the law requires it.
4. How the AI agents use your data
Our agents run on Anthropic's Claude models through Anthropic's commercial API. When an agent works on your account, it sends Anthropic the part of your data that task needs and receives an answer. That can include account names and ids, findings and their evidence rows (for example search terms behind a waste finding), your Playbook and journal notes, open proposals, a bounded set of campaign, ad set and ad tables for Meta, and what you ask in chat. We do not strip names or other text from what we send. The Security page lists it in full.
- No training. We do not train or fine-tune any model on your data. Anthropic's commercial terms say it may not train its models on customer content.
- Retention at Anthropic. Anthropic says it deletes API inputs and outputs within 30 days by default, and keeps some longer in cases such as content flagged for policy violations. These are Anthropic's statements; see Anthropic's privacy information. We have no zero-retention agreement.
- Figures come from code. Every number an agent shows is computed by our code from your data; the model only writes the words around it.
- People decide. Agents recommend; they make no decision with legal or similarly significant effects on anyone. Today they cannot change your platforms at all.
5. Google user data
Our use and transfer of information received from Google APIs follows the Google API Services User Data Policy, including its Limited Use requirements. The permissions we ask for, and what each allows, are listed on the Security page.
6. Meta data
We ask Meta only for permission to read ads (ads_read). We use one advertiser's Meta data only for that advertiser, keep each advertiser's data separate, and never sell it, combine it to serve someone else, or train models on it. Connecting Meta is not open to customers yet. How to delete Meta data is on the data deletion page.
7. Who we share it with
- Our sub-processors, the providers that host and run GrowthFabric for us, listed with what each does on the Sub-processors page. Each is bound by a contract to use the data only for us.
- Destinations you choose, such as the Slack channel or email address you set for your daily brief.
- The platforms themselves, when we read your accounts through their APIs.
- Authorities, only where the law requires it.
- A buyer or successor, if our business is sold or merged, bound to this policy, and with notice to you.
We do not sell or "share" personal data as the California Consumer Privacy Act defines those words.
8. International transfers
Our database is hosted in the European Union (Frankfurt, Germany). Our server and website run on Cloudflare's and Vercel's global networks. Some providers, including Anthropic, process data in the United States, and our company is in the United Arab Emirates. When personal data from the EU, UK or Switzerland moves to a country without an adequacy decision, we rely on the European Commission's Standard Contractual Clauses and the UK Addendum to them, as set out in the Data Processing Addendum. You can ask us for a copy.
9. How long we keep it
- Account and workspace details: while your account exists. They are deleted when the workspace is deleted.
- Platform tokens: until you disconnect the account or delete the workspace. Disconnecting asks Google to withdraw them.
- The stored copy and findings: while the account is connected. Disconnecting an account deletes its copy; deleting the workspace deletes all of it.
- Conversations with agents, settings, the Playbook and journal: while your workspace exists.
- Billing records: while your workspace exists. Our payment provider, once paid plans open, keeps its own records as merchant of record.
- Audit and change records: kept after deletion with who, where, the accounts and the before and after values removed; what remains is the kind of action, its time and internal reference ids.
- The record of a deletion: the confirmation code, the dates and the number of rows each step removed.
- Deletion requests: completed within 30 days of the request. Today you ask by email and we carry it out. See the data deletion page.
- Backups: overwritten on our database provider's schedule [backup retention: to be confirmed].
- At Anthropic: per Anthropic's terms, 30 days by default (section 4).
- Website logs: per our hosting providers' log retention [log retention: to be confirmed].
10. Security
The measures we take, and what we do not have yet, are on the Security page.
11. Your rights
Depending on where you live, you may have the right to:
- see the personal data we hold about you and get a copy;
- correct it;
- have it deleted;
- restrict or object to how we use it;
- receive it in a portable format;
- withdraw consent where we rely on it;
- complain to your data protection authority.
These rights come from the EU and UK GDPR, the California Consumer Privacy Act, the UAE Personal Data Protection Law and similar laws. To use them, email dialog@growthfabric.ai. We answer within one month and will not treat you differently for asking. If your data reached us through a customer's connected account, we will pass your request to that customer and help them answer it.
12. Children
GrowthFabric is for businesses and is not meant for anyone under 18. We do not knowingly collect data about children.
13. Changes
We will post changes on this page with a new date, and tell customers by email or in the app before a material change takes effect.
14. Contact
dialog@growthfabric.ai · [registered address: to be confirmed]