Do Hashtags Still Work in 2026?

Instagram capped hashtags at five per post in December 2025. If you are still following a guide that recommends twenty or thirty, or a three tier method that adds up to a dozen, you are following advice for a platform that no longer exists.

The cap is the smaller half of the story. The larger half is that Instagram's own head of product has been telling people for two years that hashtags do not do what most marketers think they do.

What Instagram has actually said

In December 2024, Adam Mosseri stated that the reach benefit of hashtags is virtually none. Instagram had already removed the ability to follow a hashtag.

In February 2025 he said more bluntly that hashtags do not work to increase reach. He clarified in May 2025 that they remain a reasonable way to signal what a post is about and connect it to related posts, but that they do not enhance visibility.

Then in December 2025 Instagram capped them at five, with this advice from its creators account:

Using fewer (up to 5) more targeted hashtags, rather than many generic ones, can improve both your content's performance and people's experience on Instagram.

There is one more piece of evidence, and it is the quiet kind. Instagram publishes an official explainer of how its ranking works, covering Feed, Stories, Explore and Reels. It describes activity signals, post information and poster history. It never mentions hashtags at all.

An absence is weaker evidence than a statement. But when a platform documents its own ranking signals in detail and hashtags do not appear anywhere in that document, that is worth noticing.

The measurement evidence contradicts itself

Here is where honesty is more useful than confidence. Three large studies have measured hashtag performance and they do not agree.

All three are vendor studies of their own customer base. All three are correlational. None isolates hashtags from content quality, which is the obvious confounder: accounts that stuff thirty generic tags onto a post may simply be making weaker posts.

Metricool's finding is the most current and the largest, and it points the opposite way from the conventional wisdom. It is also the easiest to misread. Posts using hashtags getting fewer views does not mean hashtags cause fewer views. It may mean that the kind of account which relies on hashtags is the kind of account that struggles for reach anyway.

Later, whose own study found reach rising with twenty to thirty tags, has since changed its public guidance to a five slot framework. The cap made the argument academic.

What to use on each platform

The counts that actually apply in 2026.

• Instagram: up to five, and Instagram says fewer and more targeted beats many and generic
• TikTok: three to five. TikTok's recommendation system leans on how people interact with the video rather than on tags
• YouTube: two to three. YouTube ignores all hashtags on a video if more than sixty are used
• Facebook: none to three. There is no evidence they help
• X: at most two. Elon Musk has publicly asked people to stop using them, saying the system does not need them any more
• LinkedIn: three to five. LinkedIn says it considers conversation topics and keywords to surface relevant content
• Threads: one topic tag per post, which is the platform limit

Note how narrow the useful range has become. Every platform has converged on a handful of tags or fewer, and two of them are actively discouraging the practice.

What hashtags are still for

They have not become useless. They have become a categorisation tool rather than a distribution tool, and those are different jobs.

A tag tells the platform, and anyone browsing that tag, what your post is about. If you run a bakery in Leeds, tagging a post so it can be found by people searching that specific niche is a reasonable thing to do. What has changed is the expectation that adding tags will push your post to people who were not already going to see it.

The practical implication is simple. Choose tags that describe the post accurately and specifically. Stop treating tag selection as a growth lever, because the evidence that it is one has fallen apart.

What to do with the time you get back

Hashtag research used to consume a real amount of effort. Here is where that effort produces more measurable returns.

Format. Buffer's analysis of 52 million posts found Instagram Reels get 36% more reach than carousels and 125% more than single images, while carousels win on engagement rate among people who already follow you. Choosing the right format for the goal moves numbers in a way five tags never will.

The opening. Instagram's ranking explainer weighs how quickly people like, comment, share and save a post. Everything that determines early engagement, which is the hook, the first line and the thumbnail, feeds signals the platform has actually confirmed it uses.

Frequency. Buffer found that posting three to five times a week rather than once or twice was associated with roughly 12% more reach per post. Consistency compounds in a way tag optimisation does not.

Specificity in the caption. The same job hashtags were doing, telling the platform what your post is about, is done better by writing clearly about the subject. Instagram's ranking systems read the content itself.

The short version

Use up to five relevant hashtags on Instagram, three to five on TikTok, and one or two elsewhere. Pick them to describe the post, not to reach new people.

Then stop thinking about them. Instagram's head of product has said the reach benefit is virtually none, Instagram's own ranking documentation never mentions them, and the largest recent study found posts using them underperform. Whatever value remains is small enough that format, frequency and a stronger opening will all outrank it.

If you are managing tags across several platforms, our free social tools include hashtag generators per network, and RepeatPost lets you save tag sets and reuse them across 16+ channels. Our guides to Instagram and TikTok cover what does move reach on each, and the multi channel strategy guide covers adapting a post per platform rather than copying it.