Video Marketing Statistics Updated November 2026

Human Written & Fact Checked · Updated

100 Video Engagement Statistics for 2026

Your video can earn a click, lose a viewer and still answer that viewer's question. Treating all of those outcomes as one engagement score hides the decision you actually need to make: improve discovery, make the content more useful or change the next step.

This hub separates observed viewing behaviour from reported preferences. Wistia supplies platform benchmarks, TechSmith and Qualtrics supply a historical workplace-learning survey, and TwentyThree supplies platform research. Their populations, observation windows and definitions differ. Historical findings retain their original dates; the update stamp does not turn them into current measurements.

Use the figures as comparison points for a matching audience and format. Average percentage watched is not the percentage of viewers who finished. A form interaction is not a sale, and a preference survey cannot establish the length that will perform best on your website.

Headline figures

  • Wistia's 2025 platform data show a 24% play rate for videos on home pages. (Wistia, Apr 2026)
  • Wistia's 2025 platform data show a 2% play rate for videos on case-study pages. (Wistia, Apr 2026)
  • In TechSmith's May 2024 workplace-learning survey, 54% selected a description matching their learning need among their top reasons for clicking their most recent instructional or informational video. (TechSmith / Qualtrics, Sep 2024)
  • TechSmith's May 2024 workplace-learning respondents preferred 10–19 minutes for instructional videos at a rate of 26%. (TechSmith / Qualtrics, Sep 2024)
  • TechSmith's May 2024 workplace-learning respondents preferred 5–6 minutes for informational videos at a rate of 19%. (TechSmith / Qualtrics, Sep 2024)
  • Wistia's 2025 videos lasting under 1 minute had an average engagement rate of 52%, meaning average share of the video watched. (Wistia, Apr 2026)
  • Wistia's 2025 videos lasting over 60 minutes had an average engagement rate of 17%, meaning average share of the video watched. (Wistia, Apr 2026)
  • Videos longer than 45 minutes generated 26% of total engagement in TwentyThree's 2025 report. (TwentyThree, Dec 2025)
  • TwentyThree's 2025 research presentation reports average viewing time of 3 minutes 40 seconds for videos on organisations' own websites. (TwentyThree, Dec 2025)
  • TwentyThree's 2025 research presentation reports 25% conversion for forms placed before playback. (TwentyThree, Dec 2025)
  • In TechSmith's May 2024 survey, 57% selected easy-to-follow content as something that kept them interested in instructional or informational video. (TechSmith / Qualtrics, Sep 2024)
  • In TechSmith's May 2024 survey, 33% reported stopping a video because they had already obtained the information they needed. (TechSmith / Qualtrics, Sep 2024)

What makes viewers press play?

The decision to start watching deserves its own measurement. Placement and the promise made by a title can shape that decision before the opening frame has a chance to help.

Which website placements attract a play?

  • Wistia's 2025 platform data show a 5% play rate for videos on blog pages. (Wistia, Apr 2026)
  • Wistia's 2025 platform data show a 10% play rate for videos on contact pages. (Wistia, Apr 2026)
  • Wistia's 2025 platform data show a 2% play rate for videos on event pages. (Wistia, Apr 2026)
  • Wistia's 2025 platform data show a 6% play rate for videos on landing pages. (Wistia, Apr 2026)
  • Wistia's 2025 platform data show a 15% play rate for videos on product pages. (Wistia, Apr 2026)
  • Wistia's 2025 platform data show a 10% play rate for videos on thank-you pages. (Wistia, Apr 2026)
  • Wistia's 2025 platform data show a 20% play rate for videos on video-gallery pages. (Wistia, Apr 2026)

Play rates differ across Wistia's displayed page categories. These 2025 observations are contextual benchmarks, not evidence that moving an identical video will cause the same result. Product visitors and article readers may have different purposes. Match the page's job and audience before selecting a benchmark.

What do viewers consider before clicking?

  • In TechSmith's May 2024 workplace-learning survey, 35% selected an acceptable video length among their top reasons for clicking their most recent instructional or informational video. (TechSmith / Qualtrics, Sep 2024)
  • In TechSmith's May 2024 workplace-learning survey, 32% selected an interesting or intriguing title among their top reasons for clicking their most recent instructional or informational video. (TechSmith / Qualtrics, Sep 2024)

Relevance and expectations matter before playback begins. TechSmith asked eligible workplace learners about their most recent instructional or informational video, so these responses describe recalled reasons for choosing learning content. Respondents selected their leading reasons; the percentages are not a mutually exclusive distribution. The chart, instead of the conflicting surrounding prose, supplies the title finding below. TwentyThree's own-website viewing-time finding appears in the takeaways and measures a later stage of the journey, not the initial decision to click.

What video lengths do viewers prefer?

A duration preference is a creative constraint, not a universal editing target. Keep instructional and informational answers separate because a viewer seeking a procedure may want a different experience from someone seeking an explanation.

How long should instructional videos be?

Preferred instructional durations vary; the survey identifies no generally ideal running time. These are rounded responses from eligible workplace learners in May 2024. A displayed zero should be read as the report's rounded value, not proof that nobody ever wants that format. Use the distribution to frame an editorial test, then assess whether viewers can complete the task the video teaches.

How long should informational videos be?

Informational preferences also span several duration bands. TechSmith asked about this category separately; keep it distinct from instructional results. Do not add the categories together or interpret the answers as measured viewing time. These historical survey preferences can inform a hypothesis about format, but they do not replace current retention data for a particular audience, topic or distribution channel.

How much of a video do viewers watch?

Retention needs both a numerator and a definition. A longer video can deliver substantial viewing time while earning a lower average percentage watched, and that is a different outcome from whether a viewer reached the final frame.

What does Wistia report by duration?

  • Wistia's 2025 videos lasting 1–3 minutes had an average engagement rate of 50%, meaning average share of the video watched. (Wistia, Apr 2026)
  • Wistia's 2025 videos lasting 3–5 minutes had an average engagement rate of 49%, meaning average share of the video watched. (Wistia, Apr 2026)
  • Wistia's 2025 videos lasting 5–30 minutes had an average engagement rate of 40%, meaning average share of the video watched. (Wistia, Apr 2026)
  • Wistia's 2025 videos lasting 30–60 minutes had an average engagement rate of 29%, meaning average share of the video watched. (Wistia, Apr 2026)

Wistia defines engagement here as the average share of a video watched. It is not a completion rate or a measure of how many viewers stayed to the end. Its platform observations show why a duration-specific benchmark is more informative than a single target for every asset. Compare like-length videos with similar jobs before diagnosing a weak result, and examine where viewers leave instead of treating the aggregate as a full explanation.

What did the historical TwentyThree retention curve show?

The historical chart reports average play percentage for individual duration labels. It is useful as an archived benchmark, not as a current target or a controlled comparison with another platform. The report's research calendar identifies an August release; the downloaded PDF is a September revision. The exact observation window for this chart is not stated.

Why can longer video still attract substantial attention?

  • TwentyThree's 2025 research presentation reports that viewers watched 50% of a 15-minute video on average. (TwentyThree, Dec 2025)
  • Videos under two minutes accounted for 44% of production in TwentyThree's 2025 research presentation. (TwentyThree, Dec 2025)
  • Videos longer than 45 minutes accounted for 4% of production in TwentyThree's 2025 research presentation. (TwentyThree, Dec 2025)

TwentyThree's later research distinguishes the share of videos produced from the share of total engagement they generate. Those denominators answer different questions: how much content is made and how much attention that content receives. Neither distribution proves that length itself causes performance. The presentation's average-play example concerns a specific duration; it should not be substituted for an average across every long video. The takeaways retain the separate engagement-share finding without repeating its number here.

How do in-video forms convert?

Form performance depends on both placement and the audience that reaches it. These observational benchmarks support testing, but they cannot establish that a mandatory gate or a later placement caused an improvement.

What happens when forms appear in the first half?

  • under 1 minute, 1st playback quarter: 10% form CTR. (Wistia, Apr 2026)
  • 1–3 minutes, 1st playback quarter: 14% form CTR. (Wistia, Apr 2026)
  • 3–5 minutes, 1st playback quarter: 20% form CTR. (Wistia, Apr 2026)
  • 5–30 minutes, 1st playback quarter: 53% form CTR. (Wistia, Apr 2026)
  • 30–60 minutes, 1st playback quarter: 50% form CTR. (Wistia, Apr 2026)
  • over 60 minutes, 1st playback quarter: 49% form CTR. (Wistia, Apr 2026)
  • under 1 minute, 2nd playback quarter: 19% form CTR. (Wistia, Apr 2026)
  • 1–3 minutes, 2nd playback quarter: 6% form CTR. (Wistia, Apr 2026)
  • 3–5 minutes, 2nd playback quarter: 4% form CTR. (Wistia, Apr 2026)
  • 5–30 minutes, 2nd playback quarter: 55% form CTR. (Wistia, Apr 2026)
  • 30–60 minutes, 2nd playback quarter: 38% form CTR. (Wistia, Apr 2026)
  • over 60 minutes, 2nd playback quarter: 67% form CTR. (Wistia, Apr 2026)

Placement quarters describe playback progress, not calendar time. Wistia's 2025 matrix reports form click-through, not sales. Each pair is a separate observation; undisclosed cell sizes prevent establishing statistically reliable winners.

What happens when forms appear in the second half?

  • under 1 minute, 3rd playback quarter: 1% form CTR. (Wistia, Apr 2026)
  • 1–3 minutes, 3rd playback quarter: 9% form CTR. (Wistia, Apr 2026)
  • 3–5 minutes, 3rd playback quarter: 16% form CTR. (Wistia, Apr 2026)
  • 5–30 minutes, 3rd playback quarter: 27% form CTR. (Wistia, Apr 2026)
  • 30–60 minutes, 3rd playback quarter: 24% form CTR. (Wistia, Apr 2026)
  • over 60 minutes, 3rd playback quarter: 38% form CTR. (Wistia, Apr 2026)
  • under 1 minute, 4th playback quarter: 40% form CTR. (Wistia, Apr 2026)
  • 1–3 minutes, 4th playback quarter: 18% form CTR. (Wistia, Apr 2026)
  • 3–5 minutes, 4th playback quarter: 35% form CTR. (Wistia, Apr 2026)
  • 5–30 minutes, 4th playback quarter: 54% form CTR. (Wistia, Apr 2026)
  • 30–60 minutes, 4th playback quarter: 51% form CTR. (Wistia, Apr 2026)
  • over 60 minutes, 4th playback quarter: 75% form CTR. (Wistia, Apr 2026)

Later placements reach viewers who chose to continue watching, complicating causal comparisons. These remaining cells preserve Wistia's 2025 duration bands. Compare similar audiences and check whether the requested action fits viewer intent.

How does TwentyThree describe form placement and gating?

  • TwentyThree's 2025 research presentation reports 11% conversion for forms placed during playback. (TwentyThree, Dec 2025)
  • TwentyThree's 2025 research presentation reports 2% conversion for forms placed after playback. (TwentyThree, Dec 2025)
  • TwentyThree's 2025 research presentation reports 13% conversion for forms placed during playback when non-dismissible. (TwentyThree, Dec 2025)

TwentyThree reports form conversion in its research presentation, using a different platform and terminology from Wistia. The figures should therefore stay in a separate comparison. The during-playback result and the non-dismissible result are overlapping views of form behaviour, not categories that should sum to a whole. The report does not establish an experimentally isolated benefit from removing a skip option. Evaluate the lead quality and viewing experience alongside the submitted-form rate before selecting a gate.

What keeps viewers watching or makes them leave?

Useful content can succeed even when a viewer leaves early. Separate task completion from frustration before asking an editor to fix every dip in the retention curve.

Which qualities help hold a learner's interest?

TechSmith's May 2024 workplace learners reported qualities that held their interest. These are preferences, not causal effects of editing techniques. Multiple choices can describe the same video, so the results are not mutually exclusive. Improve task relevance.

Why do viewers stop before the end?

  • In TechSmith's May 2024 survey, 25% reported stopping a video because expected information was missing. (TechSmith / Qualtrics, Sep 2024)
  • In TechSmith's May 2024 survey, 22% reported stopping a video because the video was boring or uninteresting. (TechSmith / Qualtrics, Sep 2024)
  • In TechSmith's May 2024 survey, 9% reported stopping a video because video quality was poor. (TechSmith / Qualtrics, Sep 2024)

Stopping early can reflect either success or a failed expectation. TechSmith's survey distinguishes viewers who had already obtained the information they needed from those who found the content unhelpful or difficult to watch. These recalled explanations cannot be assigned to every drop in an analytics curve. Pair a retention review with feedback about whether the viewer completed the intended task, then decide whether the opening promise, the explanation or the production quality needs attention.

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.