The save button might be the most successful failed feature on the internet.
Successful, because everyone uses it. Billions of bookmarks, favorites, saved posts, and screenshots pile up every day. The gesture is universal: see something, feel a small spark of I'll want this later, tap. It costs nothing. It's the lightest possible way to take yourself seriously.
Failed, because almost none of it comes back. Ask anyone to open their saved posts and find one specific thing from two months ago. Watch what happens. The scroll, the squint, the giving up. The save worked perfectly as a gesture and completely failed as a promise.
I think about saves as messages. Saves aren't things you liked once. They're breadcrumbs your past self left for your future self. A saved restaurant is a message: take her somewhere good for the anniversary. A saved workout: get back into it. A saved listing, a saved essay, a saved crib assembly hack at 2am. Each one is a small, sincere note from a version of you who believed the future version would read it.
The note gets delivered to a mailbox no one checks.
Why the button does nothing
It's worth being precise about what a save actually is, technically. On most platforms, a save is a pointer. An entry in a list that says "this account flagged this post." That's the whole feature. The platform doesn't look at what's inside the post. It doesn't know the video is about a ramen shop in your city. It doesn't know the event in the caption is three weeks away. It stores the pointer and moves on.
Which means everything that made you save it — the place, the date, the idea, the reason — stays locked inside the content. The save remembers that you cared. It has no idea why. And without the why, it can never know when to bring the thing back.
So the entire burden of return falls on you. You have to remember that you saved it, remember roughly when, remember what the thumbnail looked like, and go digging. The save button outsources nothing. It's a to-do list where every item is written in invisible ink.
Why platforms leave it this way
Here's the thing: this isn't hard to fix, technically. Reading a caption, recognizing a place name, noticing a date. None of that is exotic anymore. So why does the save button still do nothing?
Because the save already did its job, from the platform's side. Feeds are optimized for time spent in the feed. A save is a signal that helps the recommendation engine and a tiny commitment that keeps you attached to the app. Both of those happen the moment you tap. Whether you ever return to the save doesn't move any number the platform watches.
And honestly, the ideal version of a save works against the feed. A save that resurfaces when you're near the restaurant pulls you out of the app and into your life. It ends a scroll instead of extending one. No feed has much appetite for building that. No villainy required. That's just what those products are for.
What a save should be
Imagine a save button that took you literally: "got it, I'll handle this." Then a save would need three properties.
It would understand what it holds: this is a restaurant, here's the address; this is an event, here's the date; this is a recipe, here are the steps. That means reading the caption, the text on screen, the words spoken out loud.
It would know what kind of moment it's waiting for. A restaurant waits for proximity or a free evening. An event waits for the on-sale date. A gift idea waits for the birthday. The trigger is different for each, and it's almost never "user scrolls a grid of thumbnails."
And it would come find you when the moment arrives. On the map when you're three blocks away. On the calendar before tickets go on sale. That's the actual promise you made yourself when you tapped: bring this back when it matters.
Bookmarks circa 1995 stored pointers because that's all a computer could do with a webpage. The save button still works that way out of inertia, not necessity.
The honest disclosure
We're building one attempt at this, an app called Stasht. I won't pitch it here beyond that; the product pages exist if you're curious, and there are other apps trying too, which genuinely makes me happy. The more teams treating saves as promises instead of pointers, the better.
But the essay stands without the product. Next time you tap save, notice what you're actually doing: writing a note to a future self, and dropping it into a box with no outbound mail. The spark of I'll want this later is real information about what you want your life to look like. It deserves better infrastructure than a grid of thumbnails.
Your saves were the signal all along.