How to clean up years of bookmarks

A bookmarks file rarely becomes unmanageable all at once. It happens across years, one saved page at a time, until the count reads four figures and you no longer trust it to hold anything you would actually go looking for. At that point most people do one of two things: ignore the file completely, or open it, feel the size of the job, and close it again.

Neither fixes anything. What works is a method that does not ask you to sort everything in one sitting. Five short passes, done over however many sessions it takes, get an old bookmarks file back under control without losing anything you might still need.

Why sorting everything in one go never works

The instinct with a large bookmarks file is to open it and start deciding, link by link, keep or delete. That works for the first fifty. By the three hundredth it stops working, because deciding is tiring in a way that clicking is not, and a task that tires you out gets abandoned partway through, usually worse off than when you started because half the folders are now empty and the other half are untouched.

Breaking the job into passes solves this. Each pass does one thing only, across the whole file, rather than trying to finish each bookmark completely before moving to the next. That is a smaller kind of decision, repeated, and it is the difference between a job that gets finished and one that gets abandoned in an afternoon.

How do you actually clean up years of bookmarks?

Five passes, in this order. Do not skip the first one, even if your plan is to delete almost everything that follows it.

1. Export a backup first. Every browser can write your bookmarks out as a single HTML file in under a minute, for free, before you touch anything. The exact steps for Chrome, Safari, Firefox and Edge take one menu each. Do this even if you plan to delete most of what is in there. It costs nothing, and it turns pruning from an irreversible decision into one you can undo.
2. Work in passes, not one long sitting. Go through the whole file once looking only for the obvious dead weight: duplicate saves, expired sale pages, anything from a job you left three years ago. Do not stop to puzzle over the harder cases yet. A later pass is where those get decided.
3. Delete anything you cannot justify in five seconds. If the title does not tell you why you kept it, and five seconds of looking does not either, delete it. A bookmark that needs research to justify its own existence was never going to get revisited anyway.
4. Flatten the folder tree. Anything nested more than two or three levels deep is a tree you will stop opening, because remembering the path costs more than remembering the site. Collapse subfolders into their parent wherever the split no longer earns its keep. There is a longer case for why the tree itself was never really the fix in the best way to organise bookmarks, if the folders keep coming back.
5. Check what is actually still there. Click through whatever survives the first four passes, or run the file through a link checker if the count is too large to click by hand. This last pass tells you how much of the work was worth doing, and it is usually the most sobering one.

How much of what you saved is already gone?

More than feels reasonable. The Pew Research Center tracked a large sample of webpages that existed at some point between 2013 and 2023 and found that a quarter of them were no longer accessible by October 2023. Pages from 2013 specifically fared worse: 38 percent had vanished a decade on. Those figures are for the open web in general, not for bookmarks files, but a personal bookmarks file tends to skew older and more neglected than the average webpage, so treat a quarter as a floor rather than a ceiling.

Anyone who has actually gone through an old bookmarks file recognises the shape of this without needing the citation. A large share of what comes up points at a page that has moved, been redesigned past recognition, or gone entirely. That is not a sign anything was done wrong. It is what happens to any list of bare URLs, given enough years.

What is worth keeping once a link is already dead

This is worth being honest about before it gets framed as a problem with an easy fix. Nothing keeps a page alive once the site behind it goes down, no bookmark manager included. What a plain bookmarks file leaves you with, once a link dies, is a title and nothing else, which is rarely enough to work out what you actually lost.

A saved link that also carries a preview image, a tag or two and a note about why you kept it survives a dead link better, not because the page comes back, but because you can still tell what it was. That is the idea Muse is built around: saving a link pulls the page's title, its site icon and its own preview image at the moment you save it, and you can add a tag or a note on the way in. If the page dies six months later, what is left looks like the thing it pointed at, which is usually enough to go looking for it elsewhere, through the Internet Archive or a search for the title, rather than staring at a blue link with no memory of what it used to say.

Where the cleaned-up list should go

The five passes above are worth doing regardless of what you use afterward. A smaller, honest bookmarks file is more useful than a large untrustworthy one even if it stays exactly where it is, inside the browser it has always lived in.

If it keeps growing back into the same mess within a year, that is less about willpower than about what a bookmark actually records: a URL and a title, nothing about why you kept it. Once this pass is done, a library built to hold the reason alongside the link is what stops the mess from rebuilding itself, rather than needing this same afternoon again in another two years.