Privacy policy

Draft for review · revised 4 November 2026.

This is a working draft written to match what the software actually stores. It has not yet been reviewed by a lawyer, and the date it takes effect has not been set.

Who is responsible

Loudloom is a product operated by Business Brain LLC, a Pennsylvania limited liability company. Business Brain LLC decides what Loudloom collects and why, and is who you ask about it: office@businessbrainllc.net.

What we collect

• If you use the publishing estimator: the initial estimate does not ask for contact details or create an enquiry. Its selected service options appear in the page URL and can appear in ordinary hosting logs. If you write to us through the form on this site: your name, email, company, website, the platforms you use, your budget band, what you are trying to get, how often you want to post, the extra work you ticked, how many businesses it covers, whether you have pages to link to, the estimate the form showed you, and your message. Our team is emailed a copy so a person can reply. The enquiry record does not include your IP address. Our hosting and security providers may process and retain request metadata, including IP addresses, request URLs and times, to operate and protect the service.
• If you become a client: your email for sign-in (there are no passwords), the browser and operating system you sign in from (for example “Chrome on Windows”, not your IP address), your passkey's public key if you add one, your brand profile, the content we draft and publish for you, files you upload, the messages you exchange with us, which portal pages you have opened since something changed on them, your email preferences, and the tracked-link clicks and platform metrics attached to your posts.
• Connected accounts: the access token each platform issues when you authorise us, or the scoped token or app-specific password you create and paste, plus the account's name or handle, and for the newsletter the list of recipient addresses you give us. Never your account password.
• Billing: your plan, its status, the current billing period and when you accepted the terms at checkout. Card details are held by Stripe, our payment processor; we never see them.
• Readers of published posts: when someone clicks a tracked link we record the time, the referring page the browser reports, and a browser family (desktop, mobile, automated or unknown) worked out from the browser's own description of itself, which we then discard. Our click record does not include the reader's IP address, and the tracked-link handler sets no cookies. Hosting providers may still process technical request metadata as described above. Automated fetches are recorded but not counted as readers.
• People who reply to or mention a client on a connected channel: their public handle, what they wrote, and the post they were answering, so the client can see it and we can suggest a reply the client may post themselves.

How we use it

To respond to enquiries and do the agreed work: plan campaigns, assist with drafting and review, publish within the agreed approvals or delegated authority, measure what happened and report it to you. We do not sell personal data, we do not train AI models on your data, and we use service data for delivering, supporting, securing and administering the service, including billing and legal obligations.

Who else sees it

Service providers we rely on, each under its own terms: Railway (hosting; our server, database and on-volume backups run in its Amsterdam region, as do files uploaded before we began storing uploads with Google), Google (Google Workspace; it stores the files you upload, as described below), Infisical (secret storage for connection tokens), Stripe (payments; it receives your billing email), Resend (email), an object-storage service holding copies of our database backups, and Anthropic, whose AI model supports drafting, planning and other requested AI-assisted functions. Depending on the function, Anthropic receives relevant brief, profile, content and message context, including: your brief, profile, goals and voice, the text of your pages, drafts being checked against those pages, and, for suggested replies, the reply or mention and the post it answers. It never receives your connection tokens or card. Anthropic states that by default it does not use inputs and outputs from its commercial API to train its models. This is not a promise of zero retention; provider retention and security exceptions apply. Our link previews and source checks fetch your public pages as any browser would.

Files you upload. They are stored with Google in a shared drive owned by Business Brain LLC, in a folder for your account only; Google receives each file and its name. Our server passes each upload to Google and each download back to you; it does not keep a separate permanent upload copy in the normal transfer flow, though transfer buffers and operational records may exist. You do not need a Google account, and you are never given access to our Drive. Files uploaded before we began storing uploads with Google remain on our Railway volume.

What connected platforms receive, and what we read from them

• Connecting. When you authorise us, the platform receives our app's identity and the permissions we ask for, and tells us which account authorised it (an account id and name or handle) along with the token. Some tokens expire and are refreshed, which is another request to that platform.
• Publishing. The text and supported link previews of each post published to that account. Attached client-uploaded media is not currently sent by our registered publishing adapters; a fetched page-preview thumbnail may still be uploaded where supported. On X a post goes out as the post and then a reply beneath it carrying the tracked link. A link preview is built by fetching your own page and uploading its preview image to the platform. The newsletter is sent through your own Resend account to the recipient list you give us.
• Reading. The counts a platform reports for posts we published (likes, reposts, replies and similar) for your report; a check that the connection still works; and, on Bluesky and X, replies and mentions addressed to your account and the posts they answer.
• Your profile. On Bluesky, when you ask us to, we read and change your account's display name, bio, banner and automated-account label.

Every request identifies the account and carries its token, and like any web service the platform sees technical details of the request, such as our server's address. What a platform does with what it receives is governed by its own terms and privacy policy.

Cookies

Our public marketing pages do not set advertising or analytics cookies. The portal uses necessary sign-in and security cookies, including a session cookie and a short-lived cookie that ties a sign-in link to the browser that asked for it. The operator console also uses a time-zone preference cookie. These application cookies are not used for advertising or cross-site tracking. Third-party services you visit have their own cookie policies.

International processing

Business Brain LLC is based in Pennsylvania, United States. Hosting runs in the Amsterdam region; our other providers may process information in the United States and other countries under their own service terms. We do not promise that all information stays in one country or region.

Security

Sign-in is by single-use email link, bound to the browser that asked for it, with passkeys or an authenticator app as a second factor. Connection tokens are kept in a dedicated secrets vault, separate from our database. Changes to a client's setup, connections, content and billing are written to an audit log that records whether a client, our team or the software made them.

How long we keep it

We have not yet adopted a comprehensive fixed retention schedule. Current operation: client data is kept for as long as you are a client, and afterwards until you ask us to delete it; enquiries from the form are kept until you ask us to delete them; the service schedules database backups daily, with a default of fourteen local backups on our hosting volume. A failed backup or cleanup can leave a different number.

Off-site database backups currently use a managed rolling 30-day retention setting. Cleanup is bounded and may fail or leave older or unmanaged copies, so this is not a guarantee that every backup disappears after 30 days. Send-decision records have a 400-day pruning threshold; records of publishing or uncertain outcomes can be retained longer.

On request we remove a client's content, files, messages and clicks from the active service and attempt to erase stored connection-token copies. Files in our shared drive are moved to its trash, subject to Google's trash and retention controls. Deletion requests may take time to complete, and unconfirmed token erasure remains incomplete. The audit lines that say what was done, and the invoices, are kept, and copies inside backups made before the deletion remain until those backups are removed.

Your choices and requests

You can disconnect any connected account from your portal; that closes its active connection and attempts to erase stored token copies. It does not guarantee immediate erasure of every copy or revoke the platform's own authorisation; you can revoke that access with the platform. A request already sent cannot be recalled. To see, correct, export or delete your data, write to office@businessbrainllc.net. Where applicable law provides them, you may have rights of access, correction, deletion, restriction, objection or portability, and a right to complain to a relevant supervisory authority. Contact us to make a request or ask what applies to your situation.

Contact

Business Brain LLC · office@businessbrainllc.net · +1 (717) 282-1680
Questions about the Loudloom service: hello@loudloom.dev