How do you increase your Twitter engagement rate?
To increase your Twitter engagement rate, post fewer but sharper posts, open with a clear first line, ask questions people can answer, and reply to every good comment within the first hour. Rival IQ's 2026 benchmark put the median X engagement rate at 0.03% per post per follower, so small gains matter.
Here's a moment most people know. You post something you are proud of, check back an hour later, and see 400 views and two likes. The views came. The reactions didn't. That gap is your engagement rate, and the good news is that it responds to a few simple habits.
Engagement rate is the share of people who did something with your post after seeing it. That "something" can be a like, reply, repost, bookmark or click. It matters on X (formerly Twitter) because engagement is a signal that a post is worth showing to more people. A higher rate also tells you which topics your audience actually cares about.
What is a good engagement rate on X?
A good engagement rate on X depends on how you measure it. Per follower, the typical brand rate is tiny: Rival IQ found a median of 0.03% per post in 2025. Per impression, rates are higher because the denominator is smaller.
Here's the thing. There are two common formulas, and people mix them up all the time:
- Per impression: engagements divided by impressions, times 100. This tells you how well a post landed with the people who saw it.
- Per follower: engagements divided by your follower count, times 100. This is what the Rival IQ 2026 Social Media Industry Benchmark Report uses.
That report, now published by Quid, sampled 150 companies per industry. It says the X median doubled from 0.015% in 2024 to 0.03% in 2025. Those are brand accounts, so treat them as a floor, not a target. A personal account that talks with its audience can often do far better.
If you want your own number, the free engagement rate calculator does both formulas in seconds. For a step-by-step walkthrough of the maths, see how to calculate Twitter engagement rate.
Which engagement types count the most?
Replies count far more than likes in X's ranking code. As of October 2026, X's current open-source weight file gives a reply 5.0 and a like 0.66. The older 2023 release pointed the same way, with 13.5 for a reply and 0.5 for a like.
The current numbers come from param.rs in X's xai-org/x-algorithm repo, first published in January 2026. The repo says these defaults track production values, and the file changes almost daily, so treat them as a snapshot. X also explains that each weight multiplies the predicted chance of an action, not a raw count. So "one reply equals so many likes" is the wrong way to read them.
| Signal | Weight as of October 2026 | 2023 weight (history) |
|---|---|---|
| Like | 0.66 | 0.5 |
| Repost | 1.0 | 1.0 |
| Reply | 5.0 | 13.5 |
| Quote post | 5.0 | not listed |
| Share via copy link | 20.0 | not listed |
| Reply the author engaged with | not in the 2026 file | 75.0 |
| Mute author | -58.8 | -74.0 (all negative feedback) |
| Report | -234.0 | -369.0 |
The 2023 column comes from the 2023 ranking README and is history only. What this means for you is simple. A post that starts a real back-and-forth is worth more than one that collects quiet likes. That is why the steps below lean so hard on replies.
Steps to increase your Twitter engagement rate
The fastest way to increase your Twitter engagement rate is to raise the quality of each post and lower the number of weak ones. Every post you publish adds to the denominator, so a filler post can pull your average down.
1. Write a first line that earns the second
Most people decide in one glance. Lead with the point, the result or the surprise. Skip warm-ups like "So I've been thinking." Our guide on how to write engaging Twitter posts has examples of strong and weak openers.
2. Be specific
Vague advice gets scrolled past. A real number, a named tool or a short story from your own work gives people something to react to. Specific posts also invite specific replies, which are the ones that count.
3. End with an easy, real question
A good question is one your reader can answer in a sentence from their own experience. "What's the one tool you'd keep if you could only keep one?" works. "Thoughts?" rarely does. There's a full list of questions that start conversations on X if you need ideas.
4. Reply to your replies, fast
This part is actually really easy, and it's the one most people skip. When someone replies, answer them while the post is still fresh. Each answer extends the conversation. A 2025 Buffer study of about 30,000 X posts found that posts where the creator replied saw about 8% higher engagement, though the result was only marginally significant. More on this in how to get more replies on X.
5. Cut the bait
"Like if you agree" and "Drop your link below" can spike numbers once. They also invite mutes, and as of October 2026, muting an author carries -58.8 in X's weight file. Ask questions you genuinely want answered instead.
6. Review your best posts weekly
Once a week, sort your recent posts by engagement and look at the top three. What did they share? A topic, a format, a time of day? Do more of that. The guide on finding your best performing tweets shows how.
I built a free extension, XVision, that rates each of your posts against your usual views, so it's easier to spot which ones beat your normal level without opening a spreadsheet.
Why does engagement rate drop as you grow?
Engagement rate often drops as an account grows because new followers are less invested than your first ones. A post can reach more people and still get a lower percentage. That is normal and not a sign that something broke.
Based on what I've seen, early followers often know you personally or found you through a reply. Later followers may have followed after one viral post and care less about your usual topics. So judge your rate against your own past weeks, not against someone with ten times your audience. If you want to understand the link between the two numbers, read Twitter impressions vs engagement explained.
What I'd do this week
You can absolutely move this number. Pick one change, not six. My suggestion: for the next seven days, post once a day with a specific first line and one honest question, then reply to every reply within an hour. At the end of the week, run the numbers through the calculator and compare.
Small, steady wins add up. A better engagement rate is not a trick. It's what happens when your posts give people a reason to respond, and you show up when they do.
Frequently Asked Questions
How do you calculate engagement rate on X?
Engagement rate on X is usually calculated by dividing total engagements by impressions and multiplying by 100. Some benchmark reports, such as Rival IQ's, divide by follower count instead. Both are valid, so always check which formula a number uses before comparing your account to someone else's result.
Do likes or replies matter more for engagement rate?
Replies matter more for reach on X. As of October 2026, X's open-source ranking file weights a reply at 5.0 and a like at 0.66, and the 2023 release showed the same pattern. Both count toward a basic engagement rate, but replies signal real conversation and tend to bring people back to a post.
Does posting more often lower engagement rate?
Posting more often can lower engagement rate if the extra posts are weaker, because each post adds impressions without matching engagements. Posting more can still grow total reach. A sensible test is to compare two weeks: one with fewer, stronger posts and one with your usual volume.
Why is my engagement rate high but my reach low?
A high engagement rate with low reach usually means a small but loyal audience is responding to every post. That is a healthy base. To widen reach, reply to larger accounts in your niche, post when your followers are active, and share posts that new people can understand without context.
Is a 1% engagement rate good on X?
It depends on the formula. Per follower, a 1% engagement rate on X would be far above typical: Rival IQ's 2026 report put the brand median at only 0.03%. Per impression, no official benchmark defines "good," so compare a 1% result with your own past posts rather than with other people's numbers.