The archive is a backlog of practice, not a feed to scroll through.
The archive has two failure states. The first is never opening it, which wastes a large library of reusable practice. The second is scrolling it the way you scroll a feed, which spoils dozens of songs in five minutes and teaches nothing. The useful middle is to treat it like a reading list.
A reading list has a shape. You choose a few items, you read them properly, you stop. Applied to old rounds, that means picking three dates, playing those rounds as if they were new, and reading the titles only after committing to a guess. Three dates, not thirty.
The first time I tried this I was surprised by how much slower the old rounds felt. There was no social pressure, no daily reset, no reason to rush. That slowness is the advantage: the archive is the only part of the game where you can take as long as you like on a round that counts for nothing.
For the habit to survive, keep the list small. Three rounds is about twelve minutes. If you have twenty minutes, do five. If you have five, do one. The number is not the point; the honesty is.
You can start the habit on the free daily round by picking three dates from last month and playing them before you read anything. Then browse past songs only for the titles you actually missed, and only after the three rounds are done. That order is the whole method. A reading list also has a stopping point, which is the part feeds lack. When the three rounds are done, close the tab. The archive will still be there next Sunday, and the next three dates will be waiting. That rhythm is what turns a backlog into a library you actually use, and a library is easier to return to than an obligation.