The number that counts is how far back the guide will scroll and still play, and it is usually shorter than the number on a sales page. Services generally keep somewhere between 24 hours and 7 days, decided per channel group rather than across the account. Measure it: scroll back until entries stop being selectable, note the timestamp, then repeat on two other groups. Write down both the longest and the shortest result, because the shortest is the one you will run into on the night you actually need it. A service that will not put a figure in days in writing has told you something useful.

Archives are created upstream, group by group. Where a source feed carries one, the group gets replay; where it does not, the guide scrolls back and every entry is dead. Nothing in your app repairs that and it is not a fault worth reporting. The buying consequence is that an average across a catalog means nothing — what matters is the four or five groups your household opens in a normal week. Test those specifically. A catalog of 54,000+ live channels and 219,577+ VOD titles is useful scale, but scale is not the same as reach into the handful of groups you use.

1. 1Check regional groups separately, they behave differently
2. 2Dead guide entries upstream are not an app fault
3. 3Average archive depth across a catalog is not a usable number

More than most people attempt. In one evening you can establish the window on your own groups, confirm that seeking inside an archived program works, confirm that a replay runs on a second device at the same time, and check that guide start times match the clock rather than rounded blocks. That last check matters because the archive request is calculated from those times, and rounded data produces replays that open late or cut off early. Activation takes about 5 minutes, nothing auto-renews afterward and no card is stored, so the test ends when you decide it ends.

Change players before you complain. If one account shows a week of archive in one app and nothing in another, the app is not reading the playlist attributes and the service is fine. Two other patterns are app-side. Replay that stopped working right after an update is a regression, not a change on the server, so rolling back is a reasonable move. And an oversized list destabilizes players above roughly 18,000 visible channels, which is why hiding groups you never open is a real repair rather than housekeeping. Setting buffer to none and using a public DNS resolver are the two settings that reliably help.

Price it against the whole subscription, not on its own. US cable and satellite typically run $70-130 a month and put recordings in a rented box, where they stop playing once service ends. A 12-month plan at $10 a month puts them on a server: no drive to manage, nothing to fill up, and equally nothing you keep. Neither model hands you a permanent copy. The figure worth interrogating instead is uptime behind the archive. A 99.99% target is roughly 53 minutes of downtime across a year, and anything not recorded during an outage is the one gap replay cannot fill.