Video Marketing Statistics Updated November 2026

Human Written & Fact Checked · Updated

100 Social Video Statistics for 2026

Before you compare social video results, check what the percentage measures. A rate based on followers answers a different question from one based on views, and platform use does not mean someone watched your video.

This collection separates audience context from observed content performance. The original studies cover different populations and periods, so their figures remain attributed instead of being blended into an invented universal benchmark.

Headline figures

  • US adults ever using YouTube: 84%. (Pew Research Center, Oct 2026)
  • Instagram use, US adults aged 18–29: 79%. (Pew Research Center, Oct 2026)
  • Instagram Reels engagement, 10k–50k followers: 0.55%. (Socialinsider, Aug 2026)
  • Instagram Reels engagement, 45–60 seconds: 0.35%. (Socialinsider, Aug 2026)
  • TikTok engagement, 0–30 seconds: 6.00%. (Socialinsider, Aug 2026)
  • Facebook video average reach, 2025: 14,833.82. (Metricool, Dec 2025)
  • Facebook video average impressions, 2025: 15,794.95. (Metricool, Dec 2025)
  • Facebook video average interactions, 2025: 1,225.43. (Metricool, Dec 2025)
  • Facebook video average comments, 2025: 26.32. (Metricool, Dec 2025)
  • Instagram Reels average reach, 2025: 9,689.02. (Metricool, Dec 2025)
  • Instagram Reels average impressions, 2025: 15,492.65. (Metricool, Dec 2025)
  • YouTube average plays, 2025: 687.21. (Metricool, Dec 2025)

Which audiences use the platforms that carry social video?

Platform use establishes audience context, not exposure to a video. Keep the distinction before turning a population survey into a media forecast.

How widespread was platform use among US adults?

Pew's survey measures whether US adults ever use each service. It does not measure daily activity, time spent watching videos or the number reachable through an advertising product. The YouTube result appears in the takeaways; the remaining platform results are below. Respondents can use several services, so the percentages are not exclusive shares of a single audience. An advertiser should combine this broad population context with its own target definition and current campaign tools. The survey is useful for judging whether a channel is broadly established, but it cannot determine how many people will see a particular creative or which users are interested in a specific product.

How did platform use differ by age?

Pew's age categories show substantial differences in platform use. The percentages describe use within each age group. The youngest Instagram group is featured in the takeaways. The survey does not identify what respondents watch.

How did reported use differ between men and women?

Pew reports platform-use percentages separately for men and women. These describe the surveyed groups instead of the gender composition of the services' audiences. The comparison supplies context for audience planning, but it does not explain the reasons for a difference. Nor does it tell a marketer whether a specific topic, presenter or video format will connect. Avoid translating an overall demographic association into a claim about individual preferences. The original question concerns ever using the service; the results should retain that wording when quoted in a brief or report.

What do Instagram Reels benchmarks measure?

A Reel can gain views without earning a comparable response from followers. Check both the denominator and the observation period before comparing a benchmark with your dashboard.

How did Reels engagement vary with account size?

Socialinsider measures the engagement rates below against followers in its first-half observation window. These are brand-account comparisons, not a controlled experiment showing that gaining followers causes engagement to decline. Smaller and larger accounts can differ in content, audience composition and posting history. Compare an account with its relevant size tier before treating the rate as evidence of good or poor execution. Metricool's operational figures are discussed separately because its sample and measurement basis differ. An account can serve a small, interested community without achieving broad distribution. Conversely, a large following does not establish how many followers actually saw a particular Reel.

Which Reels lengths attracted engagement and views?

Socialinsider's duration groups show different patterns for engagement and average views. A high view count is not the same outcome as a high interaction rate against followers. The standout engagement result is in the takeaways. These comparisons are observational, so duration may coincide with topic, production quality or audience intent. Test length within a coherent content series before adopting a platform-wide rule. The longest category remains open-ended, unlike the bounded intervals.

What changed in Metricool's Reels sample?

  • Weekly reels, 2025: 1.88. (Metricool, Dec 2025)
  • Instagram Reels weekly reels, reported annual change: +35%. (Metricool, Dec 2025)
  • Instagram Reels reach, reported annual change: −35%. (Metricool, Dec 2025)
  • Instagram Reels impressions, reported annual change: −39%. (Metricool, Dec 2025)
  • Interactions, 2025: 624.41. (Metricool, Dec 2025)
  • Instagram Reels interactions, reported annual change: −37%. (Metricool, Dec 2025)
  • Engagement, 2025: 6.44%. (Metricool, Dec 2025)
  • Reported annual change: −3%. (Metricool, Dec 2025)
  • Likes, 2025: 475.93. (Metricool, Dec 2025)
  • Instagram Reels likes, reported annual change: −40%. (Metricool, Dec 2025)
  • Comments, 2025: 18.07. (Metricool, Dec 2025)
  • Instagram Reels comments, reported annual change: −9%. (Metricool, Dec 2025)
  • Saves, 2025: 38.90. (Metricool, Dec 2025)
  • Instagram Reels saves, reported annual change: −43%. (Metricool, Dec 2025)
  • Shares, 2025: 91.51. (Metricool, Dec 2025)

Metricool reports historical connected-account averages and relative annual changes. Percentage changes are not percentage-point movements. Its engagement measure and sample differ from Socialinsider's, so their raw rates are not interchangeable. Selected observations appear in the takeaways; the remainder follow here.

What do Facebook video benchmarks show?

Facebook video reach and engagement are separate questions. More distribution can coexist with a weaker rate, so a planning brief should preserve both measures.

How did Facebook Reels engagement vary by brand size?

Socialinsider reports Facebook Reels engagement against followers for each account-size category. The table describes the observed brands, not every Facebook publisher or every video format. Account size provides a more relevant comparison than a single platform-wide percentage. Even within a tier, however, a product demonstration, news clip and entertainment video can serve different objectives. The benchmark cannot explain the contribution of topic, paid support or existing audience loyalty. Use it to identify a question worth investigating in your own analytics, then examine the posts and audience behind the result.

How did Facebook Reels engagement vary by duration?

The Socialinsider duration comparison describes interaction rates against followers. Its pattern differs from Instagram's, so a length rule learned on one platform should not automatically become a production requirement elsewhere. These are grouped observations instead of randomized edits of the same video. Longer clips may address different topics or reach a more committed audience. Preserve the duration boundaries when quoting a result and compare several comparable posts before drawing a conclusion. The source's open-ended longest category does not establish an optimum runtime.

How did Facebook video activity and performance change?

  • Weekly videos, 2025: 2.06. (Metricool, Dec 2025)
  • Weekly videos, reported annual change: +19%. (Metricool, Dec 2025)
  • Reach, reported annual change: +44%. (Metricool, Dec 2025)
  • Impressions, reported annual change: +47%. (Metricool, Dec 2025)
  • Interactions, reported annual change: +22%. (Metricool, Dec 2025)
  • Comments, reported annual change: +42%. (Metricool, Dec 2025)
  • Shares, 2025: 28.34. (Metricool, Dec 2025)
  • Shares, reported annual change: +44%. (Metricool, Dec 2025)
  • Engagement, 2025: 8.26%. (Metricool, Dec 2025)
  • Engagement, reported annual change: −15%. (Metricool, Dec 2025)

Metricool reports format-level averages and annual changes for videos in its connected-account sample. The Facebook video category is broader than an interchangeable claim about all Reels. Reach, impressions and interactions count different things. An annual percentage change should retain its metric name instead of becoming a general statement that video performance improved. Historical sample composition and platform measurement can also affect comparisons.

How did TikTok engagement vary by size and duration?

TikTok engagement here is measured against views. Comparing its raw percentage with a follower-based Instagram or Facebook rate would create a false league table.

Did larger TikTok accounts have higher engagement rates?

Socialinsider's account-size groups do not show a simple rise in engagement as follower counts increase. The observed rates concern engagement against views, so follower size defines the comparison group instead of the rate's denominator. The sample describes brand accounts. It should not be used as a direct benchmark for every creator, paid campaign or country. A high interaction rate can coexist with limited distribution, while a widely distributed video can attract a lower share of interactions. Check absolute views and the audience's next action alongside the rate before judging performance. Account growth alone is not evidence that a video strategy has improved.

Did the shortest TikTok videos perform best?

The shortest duration category leads Socialinsider's observed engagement comparison, but that association does not prove shortening a video will improve it. The showed result appears in the takeaways. The remaining duration categories show why an isolated rule would lose useful information. An educational explanation and a brief visual joke have different demands on attention. Preserve the view-based definition when comparing clips, and assess whether the video delivered its intended message. The source's table and chart agree on the longest category; a conflicting prose description on the source page is excluded here.

What does the YouTube sample say about viewing and response?

YouTube averages describe a specific connected-account sample. Treat its unusually active publishing pattern as a property of that sample, not a recommended workload for a marketing team.

How frequently did accounts publish and attract plays?

  • Weekly posts, 2025: 20.83. (Metricool, Dec 2025)
  • Interactions, 2025: 16.09. (Metricool, Dec 2025)
  • Engagement, 2025: 2.34%. (Metricool, Dec 2025)

Metricool reports publishing activity, plays, interactions and engagement for its YouTube sample. These are distinct measures and should not be read as a funnel of unique people. The plays result appears in the takeaways. The report includes an inconsistent prior-year interaction value and growth label, which this article excludes. Current point estimates are retained exactly as printed. The publisher's press release also gives growth claims that differ from the PDF; those conflicting comparisons are not used. An account-level publishing average does not set the appropriate cadence for a team with a different audience, channel maturity or production budget.

Which kinds of YouTube interactions were recorded?

Metricool separates comments, dislikes, likes and shares in its YouTube reporting. These actions provide different signals and should not be treated as equivalent expressions of commercial intent. A comment can be a question or criticism, while sharing may reflect interest beyond the channel's existing audience. Neither establishes a purchase or qualified lead. Keep the original metric labels when adding these figures to a planning document, and investigate how the platform and analytics provider count each action. The observations concern the connected accounts in the study instead of the entire YouTube population. Use them as context while measuring the downstream outcome that counts to your own program.

Glossary

Methodology and sources

Every figure on this page comes from the organisation that produced it: regulators, statistics offices, public bodies, research institutes, and companies reporting their own data or surveys. Each one links to the original report or dataset and carries its release date. Survey results keep their population and fieldwork period, and financial figures keep the measure the producer used. Figures from different producers are not added together or ranked against each other. The page is reviewed monthly and updated when a producer releases newer data.