The short version
There is no account and no sign-up. Everything you record (which films you have seen, the shows you follow and the episodes you have watched, what you thought of them, when you watched them, and any notes) is stored on your device and never sent anywhere. There is no analytics, no advertising and no tracking. The only outside code the app ever runs is YouTube's video player, and only when you play a trailer.
What is stored, and where
Match Cut keeps your library in a database on your device. That includes:
- The films you have filed as Seen, Not Seen, Want to See, Not For Me or Maybe
- Ratings: Liked, Disliked, or Unsure
- The TV shows you follow, the episodes you have marked Seen, and what you thought of episodes, seasons and shows
- Shows you answered Not For Me or Maybe on the TV deck
- Watch dates and any notes you attach to them, for films and episodes
- Cast and crew for films you have marked Seen, which is how the app works out what you tend to watch
- Your settings, such as the deck's current mode and genre filter
None of this leaves your device. It is not backed up to any server of mine, and it is not synced between your devices. Two small pieces of the app involve servers I run, and neither handles your library: the relay that film information comes through, below, and the page a shared film opens: see Shared links.
What leaves your device
Film information comes from The Movie Database (TMDB). The app asks for it through api.matchcut.watch, a small relay I run on Cloudflare that passes each request on to TMDB and keeps TMDB's answer for a while, so that a question many people ask, such as what is popular today, reaches TMDB once rather than once per person. It keeps no record of who asked: nothing is logged, and what it stores is only the question and TMDB's answer. Posters and stills load from image.tmdb.org directly. Each request asks about one thing: a search term you typed, a page of films or shows for a deck, or a single film, show or season by its TMDB identifier. Some of those are in your library. Cast, crew and running time are fetched in the background for films you add, and if you turn on streaming notifications, the films on your Want to See list are checked from time to time for where they are streaming. The shows you follow are checked from time to time too, in the background, for new episodes and seasons, as are titles you put on your radar, to see whether TMDB has listed them yet, and when you open Pick for Me on TV, for their genres and where they stream. No request carries your ratings, your list as a whole, or any identifier for you, and because each request goes through the relay, TMDB sees it come from Cloudflare rather than from your IP address. Cloudflare, as any network does, sees your IP address when it carries a request, under its own privacy policy.
Importing from Letterboxd, or from a CSV file, is the one time titles are sent that you did not type. Neither file carries TMDB identifiers, so the app searches TMDB for each title, one request per title, through the same relay, as if you had typed each title yourself. Your ratings, dates and notes from the file are not sent.
TMDB is an independent service with its own privacy policy, and receives your device's IP address when posters and stills load from it, as any web request would. Streaming availability shown on a film's page is supplied to TMDB by JustWatch.
Discover shows what is popular and the latest trailers. Those lists are built from ordinary TMDB requests: what is in theaters in your region, what is coming soon, the most popular films and shows on each streaming service you chose and in each genre and year, and the trailers for them. They are fetched about once a day and say nothing about your library.
Other sources
A few things TMDB does not record come from elsewhere, each asked about one title at a time and never with anything about you.
- TVmaze, for the hour an episode airs. For a show you follow, the app asks
api.tvmaze.comfor the show by its TVDB or IMDb identifier, then for its episode times. TVmaze's data is used under CC BY-SA and credited in Settings. - Wikipedia and Wikidata, only when you ask for Recap a Season. The app asks
query.wikidata.orgwhich Wikipedia article belongs to the show's TMDB identifier, then reads that article's episode summaries fromen.wikipedia.org. - Plex, only if you sign in under Settings > Plex. You sign in on Plex's own page (
app.plex.tv), never in Match Cut, and Plex gives the app a token that stays in your iPhone's keychain. With it the app asksplex.tvwhich servers are shared with you, and asks the servers you choose which films they hold, a few times a day, to tell you when a film you want to see arrives. Your library is not sent to Plex. Sign Out forgets the token. Plex's own privacy policy applies to your account there. - Scary Meter, only for Trick or Treat in October. The app asks
scarymeter.comhow frightening each film in the draw is, by its TMDB identifier, and keeps the answers on your device for a month.
Recap a Season is written by Apple Intelligence on your device. The summaries go to the model on your iPhone and nowhere else.
Get Tickets, on a film that is in theaters, opens the ticket seller you chose in Settings, or a web search for showtimes near you, in Safari. Nothing is sent until you tap it, and from then on that site's own privacy policy applies.
Trailers
Trailers are YouTube videos, so playing one loads YouTube's embedded player in privacy-enhanced mode, from www.youtube.com and www.youtube-nocookie.com. That only happens when you tap play on a film's or show's page or choose one from the trailer rows on Discover; simply opening a page does not contact YouTube. YouTube then sees your IP address and which trailer you are watching, and its own privacy policy applies to that. The app keeps nothing the player stores: no cookies or site data survive once the trailer is closed. If a trailer can't be played in the app, a Watch on YouTube button opens it in the YouTube app or Safari instead.
Apart from those above, and YouTube when you play a trailer, the relay and TMDB's image host are the only ones the app contacts.
Permissions
Match Cut asks for two permissions, each only when something first needs it: notifications, and adding a picture to your photo library.
Notifications
Used to tell you about what you follow and what you saved: a film you want to see opening in theaters or arriving on a service you pay for; a show you follow returning, renewed, ending, or with a new episode out; a paused show's next season getting close, or one you're behind on going untouched for a month; something you scheduled for yesterday that was never marked seen, asking whether you watched it; your monthly and yearly reports; and something on your radar turning up on TMDB. Each can be switched off on its own in Settings, under Notifications. All of them are scheduled on your device by iOS. Nothing is sent through a server, and no push tokens are collected. Decline it and the rest of the app is unaffected: films you save still land on your Want to See list, and you can log a watch by hand.
Photos
Add-only, and only if you save an In Review report card as an image. The app can put that one picture in your photo library and cannot see anything already there.
The app does not request access to your location, your camera, your microphone or your contacts.
Exports, backups and imports
You can export your library, films and shows, to a file. The file is written wherever you choose to put it (iCloud Drive, another app, or somewhere local) and that choice is yours, not the app's. Once written, it is an ordinary file outside the app's control.
Importing from a Trakt, Letterboxd, Sofa or Marathon export or a CSV file, or restoring one of your own backups, reads files you select. Nothing is uploaded in either direction, beyond the TMDB lookups described above.
Shared links
Sharing a film, a show or an episode from its page produces a link on link.matchcut.watch. That address is served by a small program I run on Cloudflare, and it exists for one thing: when someone receives the link, their messaging app asks it what the link is for, so the message can show the poster and title instead of a blank card.
What that program receives is a film's or show's TMDB identifier, and for an episode its season and number, along with what any web request carries: an IP address and the name of the browser or app asking. It does not receive your library, your ratings, or anything saying who shared the link. It asks TMDB about the film or show and answers. It writes nothing down. There is no database behind it and no request log, though Cloudflare counts requests at its edge as any host would.
It is only ever involved once a link has been shared and someone else opens or previews it. Using the app does not contact it, and neither does sharing. Only receiving does.
Deleting your data
Settings has reset options for individual lists, and a Reset All Data option that removes everything the app has stored. Deleting the app from your device also removes its database. Because nothing is stored off your device, there is nothing left elsewhere for me to delete or for you to request.
Children
Match Cut is not directed at children and does not knowingly collect information from anyone. Film data comes from TMDB and may describe films intended for adult audiences.
Changes
If this policy changes, the revised version will be posted here with a new date. If a change ever meant the app started collecting something, that would be described plainly rather than buried.
Contact
Questions about this policy, or about anything the app does with your data: feedback@matchcut.watch.