PlayerSells
X Engagement Insights

Do hashtags work on X?

No hashtags runs +18.9% against the same accounts' other posts, across 4,781 X accounts measured in the 30 days to 23 August 2026. The only counts that differ measurably do so in the wrong direction: 2 hashtags at -8.7%, 1 hashtag at -9.3%, 3+ hashtags at -16.1%.

Key findings

  • Posts using no hashtags at all earned 18.9% above the engagement of the same accounts' other posts (95% interval +15.7% to +22.1%), measured across 4,065 X accounts and 138,641 of their posts in the 30 days to 23 August 2026. The effect held for 2,422 of those 4,065 accounts individually.
  • Posts using 2 hashtags earned 8.7% below the engagement of the same accounts' other posts (95% interval -12.3% to -5.9%), measured across 2,765 X accounts and 24,757 of their posts in the 30 days to 23 August 2026. The effect held for 1,548 of those 2,765 accounts individually.
  • Posts using 1 hashtag earned 9.3% below the engagement of the same accounts' other posts (95% interval -11.5% to -7.0%), measured across 4,781 X accounts and 59,545 of their posts in the 30 days to 23 August 2026. The effect held for 2,650 of those 4,781 accounts individually.
  • Posts using 3+ hashtags earned 16.1% below the engagement of the same accounts' other posts (95% interval -20.8% to -12.3%), measured across 2,250 X accounts and 26,916 of their posts in the 30 days to 23 August 2026. The effect held for 1,331 of those 2,250 accounts individually.
  • Hashtag use is already rare at this scale: 55.5% of the 249,859 posts measured carry no hashtag at all, so the no-hashtag bucket is the norm rather than a minority behaviour.

Engagement by number of hashtags

All four buckets, in ascending order. The zero bucket is not a control group set aside for comparison, it is the overwhelming majority of the sample, which is what makes the flat result meaningful rather than under-powered.

Median engagement by hashtag count, 249,859 original X posts
Hashtags in the postEffect (95% interval)Accounts comparedPosts
No hashtags+18.9%+15.7% to +22.1%4,0652,422 agree138,641
1 hashtag-9.3%-11.5% to -7.0%4,7812,650 agree59,545
2 hashtags-8.7%-12.3% to -5.9%2,7651,548 agree24,757
3+ hashtags-16.1%-20.8% to -12.3%2,2501,331 agree26,916

A row is only reported as a finding when its 95% interval stays on one side of zero and at least 30 independent accounts contributed a comparison. Measured over the 30 days to 23 August 2026.

What a null result is and is not worth

A null result on a small sample means nothing. A null result on a six-figure sample means that if an effect exists it is smaller than the things you are already deciding about. That is the claim here, and it is a narrower claim than "hashtags do not work": hashtags may still put a post in front of a community that follows a tag, which is a different outcome from raising engagement on the post itself, and one this dataset cannot observe.

If you are looking for a lever that does move, the measured order of magnitude runs outbound links first, then whether the post carries media, then the day of the week, with hashtags at the bottom.

Frequently asked questions

Do hashtags work on X (Twitter)?
In this dataset no hashtags runs +18.9% against the same accounts' other posts, measured across 4,781 X accounts in the 30 days to 23 August 2026.
How many hashtags should I use on X?
The counts that measurably differ are no hashtags (+18.9%), 2 hashtags (-8.7%), 1 hashtag (-9.3%), 3+ hashtags (-16.1%), measured across 4,781 X accounts in the 30 days to 23 August 2026.
Why does this contradict the usual advice about hashtags?
Most hashtag advice is inherited from Instagram, where discovery works differently, or from X as it worked before its timeline became recommendation-driven. It is also usually asserted without a sample. What is measured here is narrow but specific: for accounts with over 1.6 million followers, in a 30-day window, hashtag count does not predict engagement. A smaller account relying on tag-based discovery may well see something different, and this dataset contains no accounts of that size to check.
Could a small effect be hidden in the noise?
Yes, and this page says so rather than claiming a proof of zero. A flat result here means the measured 95% interval around the effect spans zero, not that the true effect is exactly nothing. What the interval does rule out is an effect large enough to matter next to the format decisions measured on the same posts, because a large hashtag effect would have pushed the interval clear of zero at this many accounts.
How many posts is this based on?
249,859 original X posts over a trailing 30-day window, excluding replies and reposts, of which 138,641 carry no hashtag.

How this was measured

Method
Public posts are collected per account and replies and reposts are discarded. Within each account, the median engagement of the posts in a bucket is compared against the median engagement of that same account's posts outside it, producing one ratio per account. The reported effect is the median of those per-account ratios, with a distribution-free 95% interval around it. Account size therefore cancels out instead of being re-measured, and the unit of evidence is the account rather than the post, because posts inside one account are not independent observations of how the platform behaves.
What an effect percentage means
An effect of +10% means the typical measured account earned 10% more engagement on posts in that bucket than on its own posts outside it. The interval beside it is a 95% confidence interval on that figure; where the interval spans zero the page reports no measurable difference rather than a small effect. It is not a click-through rate and not a guarantee for any individual account.
Sample and window
13,508 X accounts carry an engagement aggregate, built from 675,955 collected posts. Every aggregate covers a trailing 30-day window, last recomputed on 23 August 2026. Interaction counts are read at collection time and reflect that moment, not a live figure.
How uncertainty is handled
Every effect carries a distribution-free 95% confidence interval and the number of independent accounts behind it. A result is only stated as a finding when that interval stays entirely on one side of zero; where it spans zero the page says no measurable difference rather than printing a small number that would get quoted without its caveat. Accounts are the unit of evidence, not posts: several thousand posts from one account are one observation of how the platform behaves, not several thousand.
What we do not claim
These are observational differences, not a controlled experiment. A format that correlates with higher engagement here has not been shown to cause it: posts that carry a link, or an image, or a hashtag also differ in what they are about, and this data cannot separate the format from the content. A flat result means the interval spans zero, not that the true effect is exactly nothing.
How hashtags are counted
The count comes from the hashtag entities the platform reports on the post, so a hash character inside a word or a URL is not counted. Cashtags are not hashtags and do not enter these buckets. Posts with three or more hashtags are pooled into one bucket because individual counts above three are too rare at this scale to report separately.

Cite this page

The aggregate findings on this page are published under CC BY 4.0. You may reuse them anywhere, including in AI-generated answers, provided the source is named. The licence covers the derived statistics only, not the underlying post text, which is not redistributed.

PlayerSells. "Do hashtags work on X?." PlayerSells X Engagement Insights. Data measured 23 August 2026. https://playersells.com/insights/hashtags

Machine-readable findings (JSON)Methodology

The full dataset index lives at https://playersells.com/insights.