Journal
Essays5 min read

Why your read-it-later list keeps growing

Saving takes two seconds. Reading takes forty minutes. Every app in the category is built for the two seconds — which is exactly why your backlog never shrinks.

My Minute Mix

The bookmark bar is a graveyard of good intentions.

Open your saved list. Not the app you use for work — the other one. The one with the articles.

Somewhere in there is a piece you saved eighteen months ago that you still describe, to other people, as something you are going to read. You are not going to read it. You have known this for a while. But deleting it would be an admission, so it stays, and every few weeks you scroll past it and feel a very small, very specific kind of bad.

This is not a discipline problem. It is a design problem, and it has a precise shape.

The asymmetry nobody designs around

Saving something takes about two seconds. One tap, one keyboard shortcut, one share sheet. It costs nothing, it happens in a moment of genuine interest, and it produces a small hit of relief — the thing is handled, you can close the tab.

Reading it takes forty minutes, requires sustained attention, and has to compete with everything else in your life for a block of time you will probably never deliberately schedule.

So the input rate exceeds the output rate, permanently, by a factor of roughly a thousand. There is no version of this where the list gets shorter on its own.

The two actions, to scale
Saving an article2 sec
Reading it40 min
Drawn at true proportions — 2 seconds against 2,400. The saving bar is not a rendering error; that is what a 1,200:1 ratio looks like.

Every read-it-later app is optimised for the two seconds. Almost none of them are optimised for the forty minutes.

Look at where the effort has gone in this category. Browser extensions, share sheets, keyboard shortcuts, mobile widgets, one-click capture, automatic tagging. All of it makes an already effortless action slightly more effortless.

Meanwhile the reading side of these apps has barely changed in fifteen years. It is still a reverse-chronological list of everything you have ever saved, sorted by the least useful possible signal: when you happened to click the button.

What a year of saving looks like
SavedFinished12 months
Illustrative, not measured — the shape is the point. Two lines that never reconverge, and a gap that only ever widens.

A long list is not neutral

The standard defence is that an unread list costs nothing — storage is free, so let it grow.

But it does cost something, in three ways:

  • It taxes every visit. Opening a list of four hundred items means making a decision among four hundred options before you can read a single word. Most people resolve that by closing the app.
  • It decays. A news analysis piece has a shelf life of about a week. Half of a typical backlog is material that has quietly expired, and it is mixed in indistinguishably with the half that has not.
  • It accumulates guilt. This is the real one. The list stops being a resource and becomes a ledger of things you have failed to do. People abandon these apps not because they stop finding good articles, but because opening the app started to feel bad.

What actually changes the behaviour

The fix is not more organisation. Folders, tags, and nested collections add work at save time and change nothing at read time — you still face the same undifferentiated pile, now with extra admin.

What changes the behaviour is removing the decision. Not by picking for you at random, but by narrowing the choice using the one constraint that is actually binding in the moment you open the app.

Not what you are interested in. You are interested in all of it — that is why you saved it.

How much time you have.

Six free minutes is a fundamentally different situation from a free hour, and it calls for a completely different set of items. Yet no conventional saved list knows the difference, so it shows you the same four hundred things either way, and you read none of them.

Invert it. Start from the time — five minutes, twenty, an hour — and let the list assemble itself to fit: a short read and a three-minute video for the gap before a meeting, a long feature and a podcast episode for the commute. Suddenly there is no decision left to make, and the backlog starts moving for the first time in years.

Five minutes, answered
How much time do you have?
5 MIN15 MIN30 MIN1 HR
Your mix
ThreadWhy shipping slowly compounds
3 min
ArticleThe case for boring software
2 min
Total5 of 5 min
The same library that shows four hundred undifferentiated items in a conventional list.

That inversion is the whole idea behind My Minute Mix. Everything you have saved, in one place, assembled around the minutes you actually have.

Try it

Your saved internet, mixed for the time you have.

Articles, videos, podcasts and threads in one place — assembled to fit the minutes you actually have.

Get started for free →