How do you get more saves on a post?

Make the post worth returning to: a concrete how-to, a compact list, a framework, or a hard-won lesson stated plainly. People save reference material and depth. They rarely save a hot take or a one-liner.

guides · metrics

Saves vs likes on X

Name: Argus
Author: Argus

If you only track one number on X, track saves, not likes. Likes are the number everyone watches and the one that lies the most. Saves are quiet, easy to ignore, and the single best on-post predictor of whether reach turns into followers. Here is why the gap is so large, and how to write for the number that actually matters.

A like is a reflex. A save is intent.

A like costs nothing. It's a thumb twitch on the way down the feed, forgotten a second later. A save is different: someone decided your post was worth coming back to and filed it away. One is applause; the other is a small commitment. And commitment is what predicts the bigger commitment, the follow. That's the whole reason the two numbers diverge so hard.

a like

a reflex tap. it costs nothing and means almost nothing. it spikes in the first hour and then it’s gone.

a high like count feels like winning, but a room full of likers rarely turns into a single follow.

a save

someone bookmarked you to come back to. that’s intent, not applause. the strongest on-post sign they’ll follow.

the posts with the highest save rate are the ones that convert reach into audience. depth gets saved; hot takes don’t.

A like is a reflex that fades within the hour. A save is a bet that you're worth returning to. Only one of them reliably turns into a follow.

The algorithm agrees with you

This isn't just a nice theory about human behavior. X's ranking weights a save far more than a like, because a save is a strong signal that the post was worth someone's time. So save-worthy posts win twice: they earn more reach from the algorithm and more follows from the people who see them. A post engineered for likes gets neither of those compounding effects. It peaks and disappears.

Impressions and likes both lie

I've watched this play out in my own numbers more than once. A post can rack up tens of thousands of impressions and a pile of likes and convert almost nobody, because likes measure how agreeable a post was, not how useful. Meanwhile a plainer post that people actually saved brought in real follows off a fraction of the reach. When you rank your posts by saves instead of likes, a completely different set of posts rises to the top, and it's the set worth learning from.

Most tools stop at the top bar. Argus reads the whole funnel, so it can point at the drop from reach to follows: the leak, not the vanity number.

What a good save rate looks like

Don't chase a universal number. Save rate depends on your topic and audience, so the only benchmark that means anything is your own baseline. Divide saves by impressions for each post, find your typical rate, and then hunt for the posts that beat it by a wide margin. On my account one in-depth piece saved at around twenty times my normal rate. That multiple is the tell: it's not a slightly better post, it's a different category of post, and it's the one to make more of. You can do this math instantly with the engagement rate calculator, which shows your save rate alongside the usual number, or grade a single post with the tweet grader.

How to write posts people save

Saves reward substance. People bookmark things they want to use or refer back to, which points at a clear set of formats:

Concrete how-tos. A real method someone can act on, specific enough to follow.
Compact lists and frameworks. Something scannable they'll want when the situation comes up.
Hard-won lessons, stated plainly. The honest version of something you learned the expensive way.
Reference material. Numbers, comparisons, or a breakdown they can't easily reconstruct from memory.

The through-line is depth. A hot take earns likes and dies. A post that teaches something earns saves and keeps working, because every save is a person more likely to follow and a nudge to the algorithm to show it to more people.

Frequently asked questions

Are saves more important than likes on X?

Yes. A save signals intent to return, which correlates with following; a like is a low-effort reflex that rarely converts. The algorithm weights saves more heavily too, so save-worthy posts earn more reach as well as more follows.

What is a good save rate on X?

Compare against your own baseline rather than a fixed number. A post that saves at several times your average is a strong performer. One of mine saved at roughly twenty times normal, and that's the kind worth repeating.

How do you get more saves?

Make the post worth returning to: a concrete how-to, a compact list, a framework, or a real lesson stated plainly. People save reference material and depth, not one-liners.

Where Argus fits

Tracking saves by hand every week is exactly the kind of thing you stop doing after a fortnight. Argus reads your real analytics for you, surfaces which posts and formats actually earn saves and follows, and points you at the next thing to write, so you optimize for the number that compounds without doing the spreadsheet. You always press send. Read the 2026 algorithm guide for why the ranking rewards this, or the rest of the growth guides.