Privacy
Your collection belongs to you. It’s private by default, exportable at any time, and public only if you choose.
Last updated October 24, 2026 · Questions? Contact us
Draft for legal review
This draft describes what The Collector does today in plain language. It hasn’t been reviewed by a lawyer yet, and the wording may change before it takes effect. Open questions for review:
- The legal entity that operates The Collector and a contact address for privacy requests.
- Minimum age, and how we handle accounts of younger collectors.
- Confirm the backup periods below (docs/backups.md) and the backup storage provider, and how long hosting logs keep data after it is deleted.
- The legal basis for each use of data and for transfers to the United States (GDPR, UK GDPR).
- Data processing agreements with each service listed below.
The short version
- We collect what you record about your comics, what you need to sign in, and what billing requires.
- Your collection, notes, photos, prices and storage locations are visible only to you.
- A public profile is off until you turn it on, and shows only what you choose.
- We don’t sell your data and we don’t show ads.
- You can download everything we hold about you, photos included, and delete your account, from Settings.
What we collect
Account. Your email address, which you sign in with through one-time codes. Our authentication provider records your signed-in sessions: the browser or app, the IP address and when each was last used. You can see and end them in Settings → Devices.
Bot checks. When you ask for a sign-in code, or send us a message while signed out, Cloudflare Turnstile checks that a person is using the page. It looks at your IP address and information about your browser and device.
Messages to us. When you use the help and contact form: the topic, subject and message, the email address we should reply to, your name if you give it, the page you came from and your browser’s user agent. A takedown request also records the work, where it appears and that you confirmed its statement. Messages sent while signed in are linked to your account.
Profile. A username, display name, photo, bio, location and website, all optional, plus your privacy and notification choices, including the day and time zone of your weekly digest if you turn it on, and which notifications are pushed and your quiet hours.
Devices that get push notifications. Only if you turn push on for a device: the address your browser’s push service gave that device and the keys that encrypt what we send it, a name for the device made from its browser and system (for example “Chrome on macOS”), and when it was added, last used and last reached. While a notification waits to be pushed (a few seconds to a few hours, longer in your quiet hours), we keep a copy of it with when it should go.
Your collection. The comics you own, want or sold and what you record about each copy: edition, grade, slab certificate, signatures, purchase and sale prices, value, storage location, tags, notes and photos. Also the series you follow and the conversations you have with the assistant.
Purchase history. If you import your purchase history from Whatnot: for each order, the order numbers, your and the seller’s usernames, the listing’s title, description and category, and the date, status and amounts, as the file had them. You download the file from Whatnot and upload it here; we never connect to your Whatnot account. We keep these purchases, not the file, and they’re visible only to you.
Labels. If you print QR labels, the code on each label, the copy or box it opens, when you printed or replaced it, your label templates and settings. A label’s code says nothing by itself: scanning it opens the copy only for you, and anyone else learns only that it belongs to someone’s collection (or sees the copy on your public showcase, if you turn that on and the copy is shown there). When you print on a NIIMBOT label printer, we keep the roll’s product barcode (a code for that kind of roll, the same on every one) with the label size you printed on, so every collector’s printer recognises that roll. Others see only which size most people used, never who. It’s deleted with your account.
Scans. Barcodes your camera reads and the cover photos you take while scanning. Photos are stored in private storage so a scan can be reviewed and identified, and are deleted 30 days after the scan is added to your collection or dismissed (see keeping and deleting data).
Cover contributions. If you send a photo of your copy’s cover for an issue the catalog has no cover for: the straightened photo (in private storage until it’s reviewed), the issue and copy it’s for, its size and image fingerprint, the checks it went through, whether you asked to be credited, and the moderator’s decision.
Billing. If you subscribe to a paid plan, your plan, its status and renewal date, and the customer id our payment processor gives us. Card details go straight to the payment processor; we never see or store them.
Stores. If you run a shop on The Collector: the business details you publish (name, address, phone, email, hours), your listings, holds and the messages exchanged with collectors, and the seller fees owed. If you request a hold as a collector, the shop sees your request and the messages you send it.
Usage of AI features. For each AI request we record which model ran, how many tokens it used and what it cost, to keep your plan’s monthly allowance.
Usage statistics. Unless you turn them off, which steps of getting started you’ve reached and which features you use, with the day (see cookies and analytics).
How we use it
- To run The Collector: show your collection, identify comics, keep your devices in sync and send sign-in codes.
- To read covers and answer the assistant. Cover photos and assistant messages, with the collection details the assistant needs, are sent to AI models (see below). Barcode scans and manual entry never use AI.
- To estimate values from anonymized prices, if you share them (see community value estimates).
- To bill paid plans and seller fees, and to enforce plan limits.
- To keep the service secure and working: preventing abuse, fixing errors, and measuring page performance.
- To improve it: counting, in total, how many collectors get started and which features get used.
We don’t sell personal data, share it for advertising, or use your collection to train AI models.
What other people can see
Nothing in your collection is public by default. If you turn on your public profile in Settings → Privacy, anyone with the link can see your display name, username, photo, bio, location and website, the comics you own (cover, series and issue, grade, signed or key), your runs and key issues, and how many comics, series and keys you have. Values and your wishlist appear only if you switch them on as well. Search engines may index a public profile.
Never public: notes, storage locations, purchase prices, certificate numbers, photos of your copies, and any copy you hide from your showcase. Your profile photo is stored at an unlisted web address so it can be shown on a public profile; it isn’t linked from anywhere while your profile is private.
A cover photo you contribute becomes public once a moderator approves it: it’s the catalog cover everyone sees. If you asked to be credited, your display name (or @username) appears under it, linked to your profile only when your profile is public.
The catalog shows how many collectors own or want an issue, as a total count with no names. Shops see what you send them when you request a hold.
Community value estimates
Values in The Collector come from what collectors actually pay and sell for. When sharing is on, a price you record for a copy is added to an anonymous pool: the issue and edition, price, currency, grade, grading company, whether it’s signed, and the date. Your name, your account and the copy itself are never part of it, and a price band is only shown once at least three prices back it.
Sharing is on by default for new accounts. You can turn it off in Settings → Privacy; that also removes the prices you’ve already shared. When you delete your account, the prices you shared stay in the pool, still without anything that links them to you, unless you choose to remove them too.
Services that process your data
We use these services to run The Collector. Each processes data only to provide its part of the service.
- Supabase: our database, sign-in and file storage (including photos of your copies and scans), hosted in the United States (AWS, us-east-1). It creates sign-in codes and email-change links.
- Cloudflare: Turnstile, the bot check on sign-in and on the contact form (see above), and R2 storage for catalog cover images, served through our covers service at covers.collection.id. Cover requests carry your IP address and, when you’re signed in, a short-lived access token that includes your account id. R2 also holds our backups (see keeping and deleting data).
- GitHub: runs our nightly backup jobs. They copy the database, encrypted before it leaves the job, and stored files (such as photos of your copies) into our backup storage.
- Resend: delivers our email from hello@collection.id: sign-in codes and email-change links, notification emails and, if you turn it on, the weekly digest, both of which you control in Settings → Notifications, and store team invitations. It also delivers messages from the contact form to our inbox. It receives the recipient’s address and the content of each email. We don’t track opens or clicks.
- Push services of the browser you turned push on in: Google (Chrome, Edge on Android and other Chromium browsers), Mozilla (Firefox), Apple (Safari) or Microsoft (Edge on Windows). They receive each push for that device addressed to its subscription, encrypted so the push service can’t read it, with how urgent it is and how long to keep trying.
- Vercel: hosts the website and API, measures page loading performance (see cookies and analytics), and runs the AI Gateway we use to reach AI models.
- PageHum: page analytics on every page (see cookies and analytics).
- AI model providers, reached through the Vercel AI Gateway: currently Google (cover reading), Anthropic (cover reading and the assistant) and OpenAI (fallback and search). They receive the cover photos and assistant messages described above, with your account id so we can meter usage.
- Stripe: payment processing for paid plans and seller fees. Stripe receives your email address and the payment details you enter at checkout.
When you search the catalog or scan a book, our server may look up the title or barcode in the public comic databases of Metron, the Grand Comics Database and cataloggd. Nothing that identifies you is sent with those lookups.
When you record a slab’s certificate number, our server may look the certificate up with its grading company (PSA, and CGC once it allows it) to check the slab against the grader’s own record. Only the grading company and the certificate number are sent, never anything about you. The grader’s record is public and is kept once for everyone; whether your copy matches it is kept with your copy and shown only to you (and on a store listing, to buyers when it matches). When it matches, the key comments on the grader’s record (such as a first appearance) may be suggested to our catalog editors as notes for that issue, with the grader and certificate number but nothing about you, your copy or your store.
Cookies and analytics
We use cookies only to keep you signed in. A few display preferences stay in your browser’s local storage on that device and aren’t sent to us, like your theme, how your library, wishlist and runs are shown, and the scanner’s sound and camera settings. We don’t use advertising or cross-site tracking cookies.
Vercel Speed Insights records how fast pages load, without cookies: the page address, browser, device type and approximate country. Before anything is sent, pairing codes in phone pairing links and one-time sign-in tokens in sign-in links are removed from the address.
PageHum records page views without cookies too: the page’s path (never its query string), the site that referred you, whether you’re on a desktop, phone or tablet, how fast the page loaded, and short error messages when a page’s scripts fail. To count unique visitors it keeps a random visitor id in your browser’s local storage and a session marker in session storage; both go to PageHum with each page view, along with your IP address as part of the request. PageHum isn’t loaded on phone pairing links.
Our own usage statistics, kept in our database and never sent to an analytics or advertising company, count how many collectors get started and which features get used. When you sign up, finish onboarding, add your first copy, pair a phone, follow a series or start a pull list, place your first hold or upgrade, and when you add copies, scan, ask the assistant, open Insights or Releases, turn on the weekly digest or export, we record which of these it was, the day, your account id and at most one detail: how a copy was added (scan, search, import), the scanner mode, what you followed first, the plan, or the export format. Never which books, what you searched for, prices or anything you typed. We look at them only as totals, in which a count of fewer than five people is hidden, and without our own or test accounts. They’re deleted 180 days after the day they record, they’re in your data export, and you can turn them off in Settings → Privacy, which also deletes what was recorded.
Keeping and deleting data
We keep your data while your account exists, except for a few things. A scan’s cover photo is deleted 30 days after the scan is added to your collection or dismissed; the scan’s record (barcode, match, the copy it became) stays, and photos you add to a copy stay until you remove them. A cover photo you contribute is deleted from private storage once it’s reviewed or you withdraw it (an approved one lives on as the catalog cover). A full data export you request is kept for 7 days, then deleted from our storage. Usage statistics are deleted 180 days after the day they record.
A device’s push details are deleted when you turn push off on it or remove it in Settings → Notifications, when its push service tells us the subscription has ended (signing out on a device ends it there), after five failed deliveries in a row, or after 120 days without being used. Copies of notifications waiting to be pushed are deleted 7 days after they were sent or dropped, and after 2 days if they never went.
When you delete your account in Settings → Account, we cancel any paid plan and remove your payment details from Stripe, then delete your profile, collection, purchase history, photos, scans, assistant conversations, notifications, queued and sent notification emails, devices that get push and queued pushes, data exports, and settings.
What stays, without anything that identifies you:
- Anonymized prices you shared, unless you choose to remove them when you delete.
- Purchases you completed through a shop, in that shop’s sales records and seller-fee billing.
- Catalog corrections that were already applied to the shared catalog.
- Cover photos you contributed that were approved, as catalog covers, without your credit. Photos still waiting for review are deleted.
- A record that an account was deleted, kept in our audit log with an internal id only.
If email to your address bounced, was reported as spam, or you unsubscribed from store invitations, we keep the address on a do-not-email list so we don’t write to it again. Messages you sent us through the contact form stay as our record of the requests we handled, with the address you asked us to reply to, but are no longer linked to your account.
Stripe keeps records of past payments as the law requires. Backups may hold deleted data until they expire: Supabase’s daily database backups after 7 days; our own encrypted copies of the database after at most 6 months (daily copies after 14 days, weekly after 8 weeks, monthly after 6 months); and our copies of stored files, such as photos of your copies, 30 days after the file is deleted. Hosting logs may hold deleted data for a limited time before they expire.
Your choices and rights
- See and take your data: download everything in Settings → Your data, free on every plan: every copy and every field as CSV and JSON, the photos of your copies, your wishlist, price alerts, assistant conversations, holds and their messages, scans with their photos, imports, purchase history, and your stores’ records if you own one, with a README describing each file. It’s prepared in the background, we email you when it’s ready, and you can make one a day.
- Correct it: everything you recorded can be edited in the app.
- Limit it: keep your profile private, hide copies, stop sharing prices, turn off usage statistics, mute notification types, and choose which notifications are emailed, or turn notification email off. The weekly digest is off until you turn it on. Every notification email and every digest has a one-click unsubscribe link. Push notifications are off until you turn them on for a device, and you choose which ones are pushed, set quiet hours, and remove any device.
- Delete it: delete copies one by one, undo an import within 30 days, or delete your whole account.
Depending on where you live, you may have further rights, such as to object to some uses or to complain to a data protection authority. For anything the account tools above don’t cover, write to us through the contact form and choose “Privacy or my data”.
Security
Everything travels over HTTPS. Database rules restrict each collection to its owner, photos of your copies and scans sit in private storage that only your account can open, and phone pairing uses short-lived single-use codes that we store only as hashes. Staff with admin access can see account data to run the service, and every change they make is recorded in an audit log.
Changes and contact
If we change how we handle your data, we’ll update this page and its date, and tell you in the app before a significant change takes effect.
Questions about this page or your data: use the contact form. We reply by email.