Guide · 5 min read
Bookmarks vs likes on X: what the bookmark ratio tells you
Short answer
A like is public applause; a bookmark is a private note to come back later. When a post's bookmarks get close to or pass its likes, people found it useful enough to save but weren't moved to react in public, which is typical for guides, lists, threads and data. Divide bookmarks by likes to get the ratio, and compare it across your own posts rather than against a fixed 'good' number.
- Anyone can see a post's bookmark count, but nobody can see who bookmarked it.
- Bookmarks ÷ likes shows how much of a post's response was saving rather than applauding.
- Reference posts (how-tos, lists, threads, charts) usually have the highest bookmark ratios.
- There is no official benchmark. Our bands below are rules of thumb, and your own median is the best yardstick.
Some posts collect likes. Others quietly collect bookmarks. Since March 2023, the number of bookmarks has been displayed below a post alongside the likes, and the gap between the two gives us different information than the likes alone.
What's the difference between a bookmark and a like?
A like is a public reaction; a bookmark is a private save.
Liking a post notifies the post's author and shows your username in the list of likers. (Since June 12, 2024, other users can no longer see who liked a post; only the author can.) Bookmarking a post, however, notifies nobody at all. According to X's help center, everyone (author and reader) can see the number of bookmarks on a post, but the usernames of people who bookmarked it are never shown. The post simply goes into your own Bookmarks list.
That privacy changes behavior. People bookmark posts they want to refer back to: useful resources such as guides, tools, charts or long threads. They also bookmark posts they would rather not react to publicly, such as posts on sensitive topics like pay, health or job hunting. Liking a post shows agreement or appreciation. Bookmarking a post shows that you want to come back to it.
| Like | Bookmark | |
|---|---|---|
| Who sees that you did it | The post's author (private to others since June 2024) | Only you |
| Count shown on the post | Yes | Yes |
| Typical reason | Agreement, appreciation, support | Reference, to read later, private interest |
| Available in X's API | like_count (public) |
bookmark_count (public) |
How do you calculate the bookmark-to-like ratio?
Divide bookmarks by likes: 60 bookmarks and 200 likes gives 0.3, or 30%.
Both numbers are visible on the post itself, so you can work this out for anyone's posts, not just yours. On X's developer platform, bookmark_count is included in the public metrics, along with likes, reposts, quotes, replies and impressions. The total of all engagement types (engagements) is a private metric that only the post's author can see.
In this example we can see 3 posts from the same account:
| Post | Likes | Bookmarks | Bookmarks ÷ likes |
|---|---|---|---|
| Opinion on a product launch | 420 | 18 | 0.04 |
| Photo from a conference | 150 | 3 | 0.02 |
| Thread: 12 keyboard shortcuts | 260 | 310 | 1.19 |
Although the thread received fewer likes than the opinion post, it was saved 17 times as often. If this account wants people to come back to its posts, the thread is the type of content to repeat.
When do bookmarks outnumber likes?
Mostly on reference posts: guides, lists, step-by-step threads, templates and data.
If the ratio is greater than 1, more people bookmarked the post than liked it. This is common with content that stays useful. For example, people who see a list of 20 free tools may bookmark it to use later, but not stop to like it, because liking isn't what they came for.
The other case is sensitive topics. A well-written post about getting a pay rise, managing debt or a health question may receive many bookmarks and few likes, because the author can see who liked it but nobody can see who saved it. Few likes there do not necessarily mean the post was poorly received.
What do different bookmark ratios usually mean?
Below about 0.05 the post was mostly applause; above 0.5 it was mostly kept for later.
These bands are our own rules of thumb, not official information from X. They are useful when comparing posts, and the edges are soft.
| Bookmarks ÷ likes | Usual reading |
|---|---|
| under 0.05 | Reaction post: jokes, news, opinions, personal updates |
| 0.05 to 0.2 | Mixed: some readers saved it, most just reacted |
| 0.2 to 0.5 | Useful post: a real share of readers want to keep it |
| 0.5 to 1 | Reference post: saving is nearly as common as liking |
| above 1 | Saved more than liked: guides, lists, threads, or a topic people keep private |
Small numbers swing wildly. 4 bookmarks and 3 likes give a ratio of 1.33, which tells us very little about the post. As a very rough guideline, if there are fewer than 50 likes and bookmarks combined, don’t read too much into the ratio.
Is a high bookmark ratio better than a high like count?
Neither is better on its own; they measure different things, and the right one depends on what the post was for.
If you make humorous or commentary content, a low ratio is expected, and chasing bookmarks would only make your posts worse. If you make tutorial content, a low ratio could suggest people enjoyed the post but did not find it valuable enough to keep. Perhaps it was too generic.
Be careful with explanations about the algorithm. X has not published how much a bookmark counts for when it ranks posts, so a claim that bookmarks are worth 10 likes has no public source behind it. What we can be confident about is that the ratio shows how readers used the post.
How should you use bookmarks in your own analysis?
Track bookmarks per post next to likes and replies, and compare each post with your own median.
- Take your last 20 posts and note likes, bookmarks, replies and views.
- Work out bookmarks ÷ likes for each, and find the median.
- Sort by the ratio. Look at the top five: what format were they, how long, what topic?
- Run the totals through the engagement rate calculator, which has a bookmarks field, to see whether saves are a big or small part of your overall engagement.
- Repeat in a month and see whether the format you leaned into moved the median.
Bookmarks also help when you are reading someone else's account. If a creator regularly gets bookmarks on their posts, people are finding them useful. If a large account has many likes but very few bookmarks, its content is likely to be mostly entertainment. That is fine, but useful to know when you are judging it as a source.
Can bookmark counts be faked?
Yes, like any public count, so treat a ratio that looks wildly out of line with views and replies with suspicion.
Bookmarks can be bought the same way likes can. Buying any engagement breaks X's platform manipulation rules and tells you nothing about real readers. Unusual patterns stand out: thousands of bookmarks on a post with only a few hundred views, or every post from an account landing at the same ratio. Our guide to bot-like ratio signals covers the wider pattern.
For the reply side of the same picture, the Ratioed? reply checker compares replies with likes, reposts and quotes.
Sources
- X Help Center: About Bookmark counts · Counts are visible to everyone; who bookmarked stays private
- X Developer Platform: Metrics · bookmark_count is a public post metric; engagements is non-public
- NPR: X now hides your likes from other users (June 2024)
- X Help Center: Platform manipulation and spam policy
- Engadget: Twitter's bookmark counter (March 2023)
Questions
Can the author see who bookmarked their post?
No. X shows the total number of bookmarks on a post, but bookmarks stay private to the person who saved them. The author sees the same count as everyone else, without names.
Do bookmarks count as engagement on X?
They're an interaction with the post, and tools that build an engagement rate from public counts usually include them. X's own analytics report an engagements total for the author, and X doesn't publish a detailed breakdown of every action inside it, so use whichever total your analytics give you and stay consistent.
Does a high bookmark ratio help a post's reach?
X doesn't publish how much weight a bookmark carries in ranking. Treat the ratio as feedback on what people find worth keeping, not as a lever for the algorithm.
Why does my post have bookmarks but almost no likes?
Usually because it's reference material people want to reread, or because the topic is one people don't want to be seen engaging with in public. Both are normal. A handful of bookmarks on a low-view post can also be noise, so look at several posts before drawing conclusions.
More guides
- What does 'ratio' mean on X (Twitter)?
- Where 'the ratio' came from: a short history of the term
- What is a healthy follower ratio on X?
- Why follow/unfollow tactics hurt: X's rules on follow churn
- Bot-like ratio signals: what makes an X account look automated
- Why can't I follow more people on X? Follow limits explained
- How to see who doesn't follow you back on X
- Why so few of your followers see your posts on X