For a word that determines so much of how social media and the creator economy operate, engagement can be pretty hard to pin down. So, we looked at the data.

This report documents how engagement works across social media in 2026. Not how we wish it worked or not how platforms market it — but what the data shows.

To understand what's actually happening across feeds right now, we dug into tens of millions of posts published through Buffer — looking at engagement baselines, reply behavior, posting frequency, and how different formats perform across platforms.

The short version: If you're spending more energy looking for the perfect time to post than you are replying to the people who showed up, the data suggests you might be overthinking it. Engagement means something different on every platform, and the most powerful thing a creator can do isn't about format or posting time — it's talking back to the people engaging with them. (As long as you're still posting consistently; that matters too.)

Across six platforms and nearly two million posts, accounts that reply to comments consistently outperform those that don't — by as much as 42% on Threads and 30% on LinkedIn. That doesn't mean replies cause engagement. But it's one of the strongest patterns we found, and — we think — one of the most untapped.

Beyond replies, things get messier. Typical engagement rates vary by more than 2x between the highest and lowest platforms. Year-over-year movement is split between platforms that are climbing and those that are dropping, and the reasons aren't always what you'd expect. Format performance varies a lot from platform to platform — what works on one network doesn't necessarily translate to another. (We learned that one the hard way across our own channels.)

We built this report to be a reference, not a rulebook. The baselines can help you understand what "normal" looks like, so you can set more realistic goals.

We had a lot of fun putting this together — and we're already applying what we learned across Buffer's channels and our own. We hope it's as useful for you as it's been for us.

Key findings at a glance

Each finding below gets its own section later in the report — we've included cross-references so you can skip straight to the parts that matter most to you.

How to use this report

We really wanted this report to be practical — something you could actually use in your social media strategy. With that in mind, here's where I'd start:

• Check baseline engagement rates first. "Typical" engagement looks pretty different from platform to platform — and the numbers aren't directly comparable across networks. Knowing what normal looks like makes everything else in this report more useful.
• Then dig into the platforms you care about. The format and approach breakdowns get specific. What works varies more than we expected.
• Save timing and frequency for last. They matter, but they're a secondary layer. Timing and frequency are worth optimizing once you know what content is landing, but not the place to start.

The baseline reality: 'engagement' isn’t just one thing

Before we get into what works and what doesn't, let's get clear on what "engagement" even means — because it's not the same on every platform.

Each network defines it differently — LinkedIn, for example, includes clicks in its engagement rate, while most other platforms don't — and some don't even provide the inputs for a comparable engagement rate.

These medians represent typical performance in Buffer's dataset, rather than universal benchmarks. (We wish universal benchmarks existed. They don't.)

2025 baseline for engagement across platforms

In Buffer’s cross-platform dataset, typical engagement rate is clustered into tiers:

• Higher median engagement: LinkedIn (~6.2%), Facebook (~5.6%), Instagram (~5.5%)
• Mid-tier: TikTok (~4.6%), Pinterest (~4.0%), Threads (~3.6%)
• Lower median engagement: X (~2.5%)

Engagement is uneven — and it’s shifting

Year over year (from 2024 → 2025), platforms moved in different directions:

• Up: X (~+44%), Pinterest (~+23%), Facebook (~+11%)
• Flat-ish: TikTok (+~3%)
• Down: LinkedIn (~-5%), Threads (~-18%), Instagram (~-26%)

A word of caution on these numbers: a drop in engagement rate doesn't mean a platform is in decline. It could reflect changes to the algorithm, a shift in who's posting or how often, or simply that the platform is growing and engagement hasn't caught up yet.

Instagram for example, has increasingly steered creators toward views as its primary success metric over the last year, which means the traditional engagement rate formula may be measuring less of what Instagram is actually optimizing for. Equally, a rise doesn't automatically mean a platform is thriving for everyone.

As Julian Winternheimer, Buffer's data lead, notes: "The dramatic changes in some metrics — particularly X's 44% increase, which led to a move from a lower baseline (1.96% to a 2.83% median engagement rate)likely reflect changes in the user base or metric definitions rather than genuine performance improvements."

Buffer's growing, evolving user base can also play a part in these shifts, he adds.

"The composition of accounts changes, which can have a bigger impact on medians than actual platform performance."

Year-over-year changes can point us in the right direction when it comes to understanding platforms, but they don't tell the whole story on their own.

Replying works on every platform

One behavior showed up consistently across very different networks: posts where creators or brands reply to comments tend to earn more engagement than posts where they don't. We expected this to be true on some platforms — we didn't expect it to hold up on all six.

Estimated engagement lift when replies are present:

• Threads: +42%
• LinkedIn: +30%
• Instagram: +21%
• Facebook: +9%
• X: +8%
• Bluesky: +5%

Now, we can't say with absolute certainty that replying causes higher engagement. It's possible that posts that perform well naturally attract more comments, and creators are then more likely to reply because there's more activity to respond to. But the analysis compares each account against its own baseline, not against other accounts. And the same pattern showed up across all six platforms, which honestly isn't something we see often in this kind of data.

However, that's where the consistency ends.

Formats don’t translate from platform to platform

Format performance varies a lot from platform to platform — what works on one network doesn't necessarily translate to another. And sometimes the answer changes depending on whether you're optimizing for reach or engagement on the same platform.

A few highlights:

• Instagram behaves like two platforms. Reels get 36% more reach than carousels — but carousels earn 12% more engagement. Part of this split comes down to how engagement rate is calculated: Reels are optimized for views and reach, which dilutes their per-impression engagement rate. Depending on your goals, those are two different strategies.
• LinkedIn is carousel-dominant for engagement. Carousels earned a median engagement rate of 21.77% — roughly three times that of video and images. Even a below-average carousel performs about as well as a typical video or image post.
• Threads rewards visuals more than its "text-first" positioning suggests. However, there's enough overlap across formats that any type of post can do well.
• Facebook's format gaps are tiny. Images, video, and text all land within one percentage point of each other. Format matters less here than almost anywhere else.
• X is increasingly tiered. Text posts lead in engagement, but the Premium divide matters more than format here. After January 2025, Premium and regular account engagement rates split sharply — and in the most recent months of the study, the median engagement rate for regular accounts hit 0%.

Timing and frequency are amplifiers of engagement

Top-performing accounts publish more often and more consistently than the median account. But there's no single "best time to post" or magic number of posts per week that works across platforms, niches, account sizes, or teams (though we can make some per-platform recommendations for when posts tend to perform well).

What we can say: going quiet has an impact.. In our frequency analysis of 4.8 million channel-week observations, accounts that didn't post in a given week consistently underperformed their own baseline growth rates. Any posting was better than not posting at all, and that held across platforms.

Posting more often gives you more chances to be seen. Posting at the right time improves those chances. But the biggest lever is still creating content people genuinely want to engage with.

Methodology

We know methodology sections aren't the reason anyone opens a report. But if you're the kind of person who wants to know how the sausage gets made — or if you're planning to cite any of these numbers — this is for you.

Every section of this report rests on the same dataset, the same metric definitions, and the same interpretation rules. When a specific section departs from these defaults, we note it.

Data sources and scope

Sources: Posts published through Buffer across the platforms included in this report.

Across the studies that inform this report, which includes tens of millions of posts — from 18.8 million X posts in the Premium analysis, to 15.7 million posts in the frequency and engagement study, to nearly 2 million posts across six platforms in the reply analysis.

What this represents: Buffer users and Buffer-posted content only. It's not a full-platform view of any network, and we don't treat it as one.

Time windows: Unless otherwise stated, cross-platform baselines use 2025 data with year-over-year comparisons to 2024. Our most recent data runs through December 3, 2025. Some platform deep-dives use different windows based on the underlying study (e.g., the Instagram format analysis that uses January 2022 – October 2024).

Eligible accounts (baseline and trend analyses): To reduce noise from dormant or one-off posting, accounts must meet minimum activity thresholds — posted at least 10 times in the past year, across at least 4 different weeks.

What 'engagement' means on each platform

"Engagement" is platform-defined. Where possible, we use engagement rate. Where that isn't available, we use the strongest proxy the platform provides.

Engagement components by platform (as used in this report):

• Instagram: likes + comments + shares (format analysis may also reference Instagram's broader definition, which includes saves, where available)
• Facebook: reactions + comments + shares
• X: likes + retweets/reposts + comments
• LinkedIn: total engagements
• Threads: likes + reposts + replies + quotes
• TikTok: engagement rate (engagements ÷ reach)
• Bluesky: likes + comments + reposts

Engagement rate definition (when available): engagements ÷ reach, where reach is the number of unique accounts that saw the post. This is closely aligned to the data we have for each platform in Buffer. However, it's worth noting that not every platform defines or reports reach the same way. Where reach isn't available, we use the closest equivalent (like impressions or views).

These components vary in intent — a save and a reply are very different behaviors. We get into that more in the caveats below.

Why we use medians (and when we don't)

Across the report, we default to median metrics.

Social performance distributions are heavily skewed — a small number of viral posts and very large accounts can pull averages far from what most people actually experience. Medians give a better picture of what "typical" actually looks like.

The exception: When the question is about relative change within the same account (e.g., does replying to comments correlate with better performance for this account?), we use within-account modeling — fixed-effects regression and z-score analysis — rather than aggregate medians.

This lets us compare each account to itself over time, which is a fairer test than comparing accounts of very different sizes to each other. Several of the studies below use this approach; we'll note the specifics (dataset size, platforms, validation) in each one rather than repeating the full explanation.

What you can (and can't) compare across platforms

We follow two rules throughout the report to keep comparisons fair:

• Compare like with like. Platforms where we have engagement rate data (LinkedIn, Facebook, Instagram, Threads, X, Pinterest, and sometimes TikTok, depending on available fields) can be compared to each other. Platforms where we use a proxy metric — Bluesky and Mastodon (total interactions) and YouTube (views) — shouldn't be ranked against engagement-rate platforms as if they're measuring the same thing.
• Treat each platform's metric as a within-platform benchmark. When we say a platform "leads," it means it leads within its own measurement lens — not that it's universally "better" than another platform using a different metric.

What to keep in mind when reading this report

These apply to every section unless we say otherwise.

• These are patterns rather than rules. Most findings are observational. We report patterns that are stable in the dataset without claiming they'll hold across every niche, account size, or team.
• Who's posting may have changed, too. Year-over-year movement can reflect platform changes and shifts in who's posting — adoption patterns, account mix, industry mix, and maturity. We can't always separate the two.
• Platforms change constantly. Features, ranking systems, and UI surfaces evolve all the time. Our findings describe how content performed in the window we measured, not how it will perform forever.
• Not all engagement is the same behavior. A like, a save, a repost, and a reply carry very different intent. They're counted as "engagement" where the platform defines them that way, but they're not interchangeable — and we try not to treat them as if they are.

How each study works

This report combines multiple analyses. Each one uses a method matched to the question it's trying to answer.

Cross-platform baseline and year-over-year comparisons

• Goal: Establish "typical" engagement by platform and how it's moved.
• Metric: Median engagement rate where available; otherwise, the strongest proxy metric (views, reach, or interactions).
• Output: Platform tiering, monthly trend lines, year-over-year deltas.

Reply effect analysis

• Goal: Measure whether replying to comments is associated with higher engagement within the same account.
• Method: Within-account modeling (as described above), comparing each account to itself over time, controlling for stable differences between accounts and relevant covariates (account size, niche, location, where available).
• Output: Estimated engagement lift when replies are present, by platform.

Content format performance by platform

• Goal: Identify which formats generate higher typical engagement within each platform.
• Metric: Median engagement metric per post by format — engagement rate for most platforms, engagement as a percentage of reach for Instagram, total interactions for Bluesky and Mastodon, and views for YouTube.
• Output: Ranked format comparisons and relative deltas.

Timing and frequency analysis

• Goal: Understand how publishing cadence and posting windows relate to performance.
• Frequency method: Compare median weekly posts for top performers vs. all users. Top performers are defined as the top 10% of total weekly engagement within each platform, each week.
• Timing method: Identify higher-performing time windows by platform, reported as windows rather than single "best time to post" slots.
• Output: Cadence and timing framed as amplifiers, not primary performance drivers.

Posting frequency and follower growth

• Goal: Measure whether posting frequency is associated with follower growth within the same account over time.
• Method: Within-account modeling across 4.8 million channel-week observations from approximately 161,000 profiles on Facebook, Instagram, and X.
• Validation: Z-score analysis measuring each channel's weekly growth relative to its own baseline.
• Output: Evidence of a positive frequency–growth relationship, including a measurable "no-post penalty" (accounts that skip a week tend to underperform their own baseline growth rate).

One last note on how we write about all of this: we state the metric first in every section — engagement rate, median engagement per post, reach, views, or interactions — so you always know what's being measured. We label the unit of analysis (post-level, account-level, or week-level). And we default to conservative language — "associated with," "tends to," "in this dataset" — unless the claim is strictly definitional. If we say something stronger, we've earned it in the data.

The reality of engagement in 2026

Engagement is not the same across platforms.

The same account can publish similar content across platforms and see wildly different performance. That doesn't necessarily mean the content flopped — every network measures different actions, from different audiences, in very different feeds.

With that in mind, let's take a look at the basics. Here's what "typical" engagement looks like on each platform and how it's shifted from previous years. (If you want the full breakdown of how we define engagement by platform, that's in the methodology.)

Typical engagement in 2025

Platforms cluster into clear tiers based on median engagement rate:

• Higher median engagement: LinkedIn (~6.2%), Facebook (~5.6%), Instagram (~5.46%)
• Mid-tier: TikTok (~4.6%), Pinterest (~4.0%), Threads (~3.6%)
• Lower median engagement: X (~2.5%)

Most of the confusion around 'what's working' comes from ignoring these tiers. A post that generates a 4% engagement rate is underperforming on LinkedIn, but outperforming on X.

How the baseline shifted: 2024 → 2025

The only constant on social seems to be change. This applies to baseline engagement rates, too. Here's a look at how much these rates have shifted in just one year.

Up:

• X: ~+44% (from ~2.0% to ~2.8%)*
• Pinterest: ~+23% (from ~3.2% to ~3.9%)
• Facebook: ~+11% (from ~5.0% to ~5.6%)

Flat:

• TikTok: +~3% (from ~4.4% to ~4.5%)

Down:

• LinkedIn: ~-5% (from ~6.4% to ~6.1%)
• Threads: ~-18% (from ~4.4% to ~3.6%)
• Instagram: ~-26% (from ~7.3% to ~5.4%)

Important context: X's jump is the largest relative gain in the dataset, though X still sits at the bottom of the engagement-rate rankings. A big percentage jump from a low base.

A note on what's driving these shifts: Year-over-year deltas can reflect real platform changes — algorithm updates, feature launches, UI redesigns — but they can also reflect changes in who's posting.

In 2025, the number of posts we analyzed grew significantly across most platforms (often 2–3×). That strengthens our confidence in the 2025 medians, but it also means the underlying user base may have shifted.

A growing mix of newer, smaller, or differently-niched accounts can move medians even if the platform itself didn't change in any meaningful way.

We treat year-over-year movement as a directional signal, not a final verdict.

Where engagement rate doesn't apply

Not every platform in this report has a clean engagement rate. For some, we're working with a different primary metric entirely — which means they shouldn't be ranked against the engagement-rate platforms.

• YouTube Shorts: views. Median views more than tripled year over year (from ~86 in 2024 to ~268 in 2025). That sounds like a platform story, but it's likely at least partly a user-base story. As the mix of accounts publishing via Buffer shifts, typical view counts move even if the underlying distribution on YouTube is stable.
• Bluesky: interactions per post (likes + comments + reposts). The 2025 median sits around ~4 interactions per post, relatively stable month to month. Year over year, the median dipped slightly (from ~5 to ~4) while post volume nearly quadrupled — an expected pattern when usage broadens beyond early adopters.
• Mastodon: interactions per post (shares + favorites + comments). The median held steady at ~3 interactions per post through 2025, with no meaningful year-over-year change.

With all of the above in mind, you're probably seeing how tricky it is to rank platforms based on engagement rate. It's not quite as cut-and-dried as "LinkedIn has the highest engagement rate. Even when metrics are similar, you're comparing apples with oranges.

Views, interactions, and engagement rate are different metrics describing different things, and comparing them side by side is how you end up with misleading rankings.

What we can say for sure

Engagement isn't evenly distributed across platforms, and it doesn't mean the same thing everywhere.

So what does tell you whether someone actually cares about your content — not just scrolled past it or tapped a like out of habit?

That's where replies come in.

The effect of replies on engagement

We've spent a lot of this report explaining how different all the major platforms are, but in this one area, we saw similar results across the board.

The best part is, unlike many other factors on social, this is completely within the creator's control: replying to comments on your posts.

Posts where the account replies to comments tend to earn more engagement than posts where they don't.

These findings were similar across the six networks where we have reply data. Here, we used the fixed-effects approach to compare each account to its own performance over time, not to other accounts.

The headline numbers

Across nearly 2 million posts from 220,000+ accounts on Threads, LinkedIn, Instagram, Facebook, X, and Bluesky, posts with replied-to comments consistently outperformed those without.

Threads: +42% engagement

The largest lift in the dataset, and the Buffer team wasn't surprised to see Threads right at the top of the list. Threads gives replies unusual weight in both its UI and its ranking. At the profile level, about two-thirds of accounts performed better on posts where they replied.

LinkedIn: +30% engagement

Within the same account, replying correlates with meaningfully stronger post-performance. LinkedIn also gives comments more in-feed weight than most other platforms, and now even has impression metrics for comments on posts. About 83% of profiles performed better when they replied — the highest rate of any platform in the dataset.

Instagram: +21% engagement

Even after controlling for whether posts had comments at all, replying correlates with higher engagement relative to the account's own baseline. About 63% of profiles performed better when they replied — a smaller share than LinkedIn, but notable on a platform where the feed is built around images and video, rather than conversation.

Facebook +9.5% engagement

On Facebook, we measured reactions — likes, loves, hahas — to see the effect of replies on engagement, rather than total engagement. That means the lift doesn't come only from the replies themselves, which add to the comment count. But when an account replies to comments, the post gets more reactions from other people (possibly because it is surfaced more by the algorithm). The conversation seems to drive a separate, independent response from the wider audience.

About 54% of pages performed better when they replied. On a platform this big and this mature, even a modest lift adds up to real volume.

X: +8% engagement

This is the least certain result in the set — with smaller reply samples and X's tiered visibility mechanics, the data doesn't fully rule out noise. However, it's still statistically significant and directionally consistent with the other five platforms.

Bluesky: +5% engagement

This is smallest lift in the set, from smaller samples on a newer platform. That said, it's still statistically significant and worth watching as the platform matures and reply behavior becomes more established.

A little more context

The platforms built around conversation, where replies get real visibility in the UI and the algorithm, are the ones where replying correlates most strongly with performance.

Threads and LinkedIn are both designed for discussion, and their interfaces actually surface replies in ways most platforms don't. The lift from replies is still meaningful on Instagram and Facebook, just smaller. And it's weakest on X and Bluesky, where reply samples are smaller, and distribution is more unpredictable.

It's also worth noting that the causal arrow could point in either direction. Strong posts attract more comments, which creates more opportunities to reply. And replying to comments drives engagement up, and that engagement drives replies, or so on.

The differences in engagement by content type: A platform-by-platform breakdown

We've touched on this already, but it bears repeating here: what works on one platform might not on another.

The platform-by-platform data that follows is where that gets specific.

LinkedIn

LinkedIn has the highest median engagement rate of any platform in our dataset at ~6.1% in 2025. It dipped slightly from ~6.5% in 2024, but it's still comfortably on top.

It's also a platform in the middle of an identity shift. LinkedIn has been courting creators, experimenting with a dedicated video feed and improved analytics. But it's carousels (document/PDF posts) that earn the most engagement on LinkedIn.

• Carousels earned a median engagement rate of 21.77%.
• Video came in at 7.35%.
• Images were close behind at 6.52%.
• Link posts at 3.81%.
• Text posts at 3.18%.

There's a lot of variation within carousels, though. Among stronger-performing carousel posts, engagement was above 41%. Among weaker ones, it was around 5.4% — which is pretty close to the median rate for video and images. So even a below-average carousel is doing about as well as a typical video or image post.

In an episode of Buffer Chat, LinkedIn's Head of Scaled Programs, Callie Schweitzer, emphasized video as a key priority for creators in 2025. Our theory is that LinkedIn might be headed down a similar behavior path as Instagram, where videos mean reach, but carousels mean engagement. More on this below.

Threads

Threads' median engagement rate came in at ~3.6% in 2025, down from ~4.4% in 2024 — an 18% decline that puts it closer to X (~2.5%) than to the higher-engagement platforms.

Threads is positioned as a conversation-first space (or Instagram's text-forward sibling). But the formats need a bit more nuance than simply ranking them against each other.

• Video led with a median engagement rate of 5.55%.
• Images weren't far behind at 4.55%.
• Text posts came in at 2.79%.
• Link posts sat at 2.34%.

There's a lot of variation within each format, though. A good text post can easily outperform a mediocre video. There's enough overlap across formats that any type of post can do well on Threads.

Threads is still young and still refining its algorithms. We wouldn't be surprised to see shifts that change these numbers. But for now, mix in visuals with your Threads posts to give your posts a boost.

Instagram

Instagram's median engagement rate fell from ~7.3% in 2024 (the highest in the dataset that year) to ~5.4% in 2025 — a 26% decline that moved it from first place to third, behind LinkedIn and Facebook.

When we look at engagement rate as a percentage of reach, carousels come out on top:

• Carousels led with a median engagement rate of 6.90%.
• Single images came in at 4.44%.
• Reels followed at 3.31%.

Carousels earn roughly 109% more engagement per person reached than reels, and single images earn about 34% more than reels. Even static images comfortably outperform video when it comes to engagement on Instagram. However, as always, there’s nuance here.

Reels, carousels, and single images serve different purposes.

And there’s one thing worth noting: we're measuring engagement rate here — likes, comments, saves, and shares as a percentage of reach. But reels are often optimized for views rather than these kinds of interactions, so a lower engagement rate doesn't necessarily mean Reels aren't working. It may just mean people are consuming them differently.

In addition, the format breakdown above doesn't capture the full picture, because reach and engagement point in different directions on Instagram.

A separate analysis of 4M+ posts published via Buffer between January 2022 and October 2024 showed us that:

Reels tend to get the most reach

• Reels vs carousels: 1.36× the reach (+36%)
• Reels vs single-image posts: 2.25× the reach (+125%)

Instagram has a dedicated reels discovery tab, so reels have a built-in advantage for reaching people who don't already follow you — feed-native formats don't get that same boost.

Carousels tend to get the most engagement

• Carousels vs reels: 2.09× the engagement rate (+109%)
• Carousels vs single-image posts: 1.55× the engagement rate (+55%)
• Single images vs reels: 1.34× the engagement rate (+34%)

Carousels keep people on the post longer, meaning more chances to save, share, and comment, and potentially multiple chances to reappear in-feed.

It's a bit like Instagram is two different platforms in one, depending on where you post your content. And which 'platform' you choose depends on the goal of your content. Here's a helpful way to look at it:

• Discovery mode (reaching new people): Reels are more likely to reach people who don't follow you.
• Relationship mode (engaging your existing audience): Carousels drive deeper interactions from people who already do.

The "best format on Instagram" has no single answer as it depends on your goals.

Facebook

Facebook's median engagement rate rose to ~5.6% in 2025 (up from ~5.0% in 2024, a +11% gain), making it the second-highest engagement platform behind LinkedIn and one of only three where engagement moved meaningfully upward year over year.

But on Facebook, the gaps between formats are small.

• Images led with a median engagement rate of 5.20%.
• Video at 4.84%.
• Text posts at 4.76%.
• Link posts at 4.43%.

That's less than one percentage point separating images from text. On Facebook, format choice matters less than almost anywhere else in this dataset. Images have a slight edge, and link posts are slightly behind (which is consistent with the broader trend of platforms keeping users on-platform). But the differences are small enough that what you post about probably matters more than whether it's a photo or a video.

X/Twitter

X's median engagement rate jumped from ~2.0% in 2024 to ~2.8% in 2025 — a +44% increase, the largest relative gain in the dataset. But X still sits at the bottom of the engagement-rate platforms, and the bigger story is structural.

X introduced Premium accounts in March 2023, promising several new features for paid users, with better content performance among them. In our research into the effect of X Premium on reach and engagement, we started seeing that happen around January 2025.

Before that, Premium and regular accounts moved in similar directions on engagement rate. After January 2025, they split — Premium engagement rates rose while regular account engagement rates fell.

But Premium divide aside, text still wins on X by a wide margin:

• Text posts led with a median engagement rate of 3.56%.
• Images at 3.40%.
• Video at 2.96%.
• Link posts at 2.25%.

Text and images are close enough that both work well. Video can work on X, but it doesn't carry the same default advantage here as on other platforms.

TikTok

TikTok's median engagement rate came in at ~4.5% in 2025, roughly flat from ~4.4% in 2024. It sits in the middle of the pack — behind LinkedIn, Facebook, and Instagram, but ahead of Pinterest, Threads, and X.

The format finding here probably won't surprise anyone: on a video-first platform, video performs best.

• Video led with a median engagement rate of 3.39%.
• Images at 1.92%.

What is interesting is how competitive images have become. TikTok started as a pure video platform, but with the introduction of carousels and photo posts, images are proving more viable than you might expect.

Bluesky

Bluesky uses total interactions (likes + comments + reposts) rather than engagement rate, so it's not one-to-one with the other platforms in this section.

• Video earned a median of 5 interactions per post.
• Images at 4.
• Links at 3.
• Text at 3.

The median dipped slightly year over year (from ~5 to ~4) while post volume nearly quadrupled. This is to be expected as a platform grows beyond its early adopters and the user base broadens toward smaller and newer accounts.

A brief look at some other platforms

Not every platform in our dataset got its own deep dive in the web report.

For these platforms, the data we have is solid enough to share what we have, but not enough for the full treatment we gave those above.

Here’s where things stand on Pinterest, YouTube, and Mastodon.

Pinterest

Pinterest's median engagement rate rose to ~3.9% in 2025, up from ~3.2% in 2024 — a +23% gain that makes it one of only three platforms where engagement moved meaningfully upward, alongside X and Facebook.

Video is the clear winner on Pinterest.

• Video led with a median engagement rate of 5.75%.
• Images at 3.15%.

That's nearly double the engagement for video — one of the largest format gaps in the dataset. Pinterest has been investing in video features, and the data suggests that investment is paying off. If you're still treating Pinterest as an image-only platform, consider adding videos to your strategy.

YouTube

For YouTube, we measure median views rather than engagement rate, which makes it difficult to compare directly with rate-based platforms. (See The Baseline section for full context on why.)

The median YouTube video published through Buffer earned a median of 433 views (52 views on the lower end and 1,224 on the higher end).

However, this data more likely reflects shifts in who's publishing via Buffer at least as much as YouTube's underlying distribution — as the mix of accounts changes, typical view counts shift even if the platform itself is stable.

The main thing to note here is: views are the first gate to pass on YouTube. Likes, comments, subscriptions, and shares are often sparse relative to view volume. A strong median view count can still come with very low interactions.

Mastodon

Mastodon uses total interactions (shares + favorites + comments) instead of engagement rate, and is the most stable platform in the dataset.

• Images and video both earned a median of 3 interactions per post.
• Links and text both at 2 interactions.

Timing and frequency can boost engagement –– but not drive it

“How often should I post?” and “When should I post?” are two of the most common questions creators and teams ask us at Buffer.

The honest answer is there isn't a single universal number for either one. But there are clear patterns in the data.

THere’s how to think about both: timing and frequency are amplifiers. They increase your chances of success and concentrate it into higher-probability windows. But they don't create engagement on their own: they boost what's already working.

On platforms where virality plays a bigger role in whose posts get seen — TikTok is the best example — there's another factor to consider: posting more also increases the odds of any single post breaking out. In that context, frequency isn't just an amplifier; it's also a numbers game.

One thing worth noting up front: frequency and performance tend to travel together, and there are a few possible reasons:

• Resources. Successful accounts can afford more output thanks to bigger teams, better workflows, and more assets.
• Momentum. Higher engagement motivates more posting. The causal arrow runs in both directions.
• Platform fit. Some platforms may reward frequent publishing more than others, but the strength of that effect varies by audience and format.

We can't fully separate these in observational data. But what we can do is show you what the patterns look like.

Frequency: top performers post more, more consistently

We compared weekly posting frequency between two groups on each platform: the median account and the top 10% by total weekly engagement. To qualify, accounts needed at least 10 posts in the past year, in at least 4 different weeks.

Across platforms, top-performing accounts post more frequently than the median user — and they do it consistently across the year, not just during spikes.

The gap is widest on text-forward platforms: X, LinkedIn, and Threads. These are feed-dense environments where it takes less production effort to post, so top performers pull ahead more clearly by posting more often.

The gap is closer on visual-heavy platforms, especially Instagram and TikTok. Top performers still tend to post more, but the difference is less consistent — probably because these formats take more effort to create, so it's harder to maintain a high volume.

The no-post penalty

This is the finding that surprised us most in the frequency data.

In a separate analysis of 4.8 million channel-week observations from ~161,000 profiles on Facebook, Instagram, and X, we measured how follower growth changes when the same account posts at different frequencies across weeks (see Methodology section for details).

The pattern was clear: accounts that didn't post in a given week consistently underperformed their own baseline growth rates. We call this the "no-post penalty." Even posting just 1–2 times per week produced a meaningful improvement over weeks with no posts at all.

And the benefits continued to scale. Accounts posting 10+ times per week saw the largest gains, averaging 32 additional followers per week compared to silent weeks. But the most important threshold is the first one: any posting is substantially better than no posting. Consistency matters more than volume.

There's a tension here, though: while posting more is associated with higher total engagement and follower growth, our engagement rate analysis of 15.7M posts suggests that reach per post tends to decline at higher frequencies.

Posting more helps you grow in aggregate, but each individual post may reach a smaller share of your audience. The best approach is a cadence you can sustain while protecting quality — not maximum volume at the expense of everything else.

What we can and can’t claim

One thing we can say clearly from this data: top performers publish more often than the median account, across platforms.

What we can't say is that a single "optimal" frequency exists across niches, account sizes, or teams — or that posting more causes higher engagement.

How timing fits (and why it’s not the 'secret sauce')

From our timing analysis, two things are consistently true:

• There’s no universal “best time to post” across platforms. Each network has its own usage rhythms.
• The “best time” is usually a window, not a single slot. High-performing posts tend to cluster in certain parts of the day and week, but the difference between top time blocks is often smaller than people expect.

Timing is a distribution advantage or an amplifier. It can help a good post get its first push but it can’t turn an average post into a high performer.

The windows below are where higher-performing posts clustered in our data. Use them as starting points for testing, not rules:

• Facebook: 8–11 a.m. weekdays, peaking Thursday at 9 a.m.
• Instagram: 6–9 p.m. weekdays, peaking Thursday at 9 a.m.
• LinkedIn: 3 p.m.–8 p.m. weekdays, peaking on Wednesday at 4 p.m.
• TikTok: 8 a.m.–11 a.m. weekends, peaking Sunday at 9 a.m.
• X: 6–11 a.m. weekdays, peaking Tuesday at 9 a.m.
• Threads: 6–11 a.m. weekdays, peaking Thursday at 9 a.m.
• Bluesky: 6-9 p.m. weekends, peaking Sunday at 5 p.m.

The data suggests a pretty clear pecking order: what you post matters most, how often you post matters a lot, and when you post matters least.

That's not to say timing is irrelevant — but the biggest gap in this data isn't between "good timing" and "bad timing." It's between posting and not posting. So experiment with timing to find what works for your audience, but don't let the search for a perfect schedule keep you from hitting publish.

What this means

We set out to document how engagement is actually functioning across platforms — not to tell readers what to do. But after analyzing tens of millions of posts, a few things stand out.

We kept looking for a sophisticated answer to engagement in 2026, but the data kept giving us the simple one.

The strongest signal in this entire dataset wasn't a format trick, a timing hack, or an algorithm exploit. It was replies.

On every platform we studied, creators who reply to comments do better than creators who don't. It's maybe the simplest possible version of what social media was supposed to be: people talking to the people who talk to them.

The next thing the data kept saying: show up. The biggest gap in the frequency data isn't between good timing and bad timing. It's between posting and not posting. The no-post penalty was real and consistent across all platforms. So show up first, optimize second.

And the third takeaway: fragmentation is real, but it's not bad news. Every platform defines engagement differently, measures it differently, and rewards different behaviors. There’s no single playbook to copy — which means there's no single algorithm to lose to, either. Growth can happen anywhere, on any platform, as long as the work is good and you're showing up.

It's also worth noting that the platforms where reply effects were strongest — Threads and Bluesky — are also the newest. They were built in an era where the value of conversation is understood differently than when Facebook and X first launched. We can't prove social media is shifting toward conversation over reach. But the platforms being built right now are designed as if it is — and the engagement data from those platforms looks like that bet is paying off.

Whether or not that's a trend, the practical takeaway is the same: reply to the people who engage with you, post consistently and make good content.

View the full article