What Does Sound On vs. Sound Off Mean for Short-Form Video?

A Verizon Media and Publicis Media study of 5,616 U.S. adults aged 18 to 54, reported by Streaming Media (streamingmedia.com), found that 80 percent of people who use captions are not hearing impaired. They use captions because they watch on mute. The same study found 69 percent of respondents watch video with the sound off in public places. Helen Lin, Chief Digital Officer at Publicis Media, drew the operator conclusion in the report, verbatim, "Leveraging captions isn't just about addressing sound-off environments, it's understanding the consumer mindset and creating greater advertiser opportunity than one would think," per Lin. That is the entire premise of designing for both modes: a large, well-measured share of your audience is watching in silence by choice, and a video whose meaning lives only in its audio is invisible to them.

Definition

Sound on versus sound off describes the two viewing modes for social video. Sound on means the viewer hears voiceover, music, and effects. Sound off means the viewer watches in silence and depends entirely on visuals, text overlays, and captions to follow the content. Because both modes occur constantly and unpredictably, durable short-form video is designed to be comprehensible in either.

What It Means

The case for designing for silence rests on real survey data, not a vibe. A Verizon Media and Publicis Media study of 5,616 U.S. adults aged 18 to 54, reported by Streaming Media (https://www.streamingmedia.com/Articles/ReadArticle.aspx?ArticleID=131860), found that 69 percent of respondents watch video with the sound off in public places, and, notably, that 80 percent of people who use captions are not hearing impaired, meaning captions are mostly a sound-off convenience rather than only an accessibility tool. The widely repeated 85 percent figure has an older and narrower origin: a Digiday report dated May 17, 2016 (https://digiday.com/media/silent-world-facebook-video/) found that as much as 85 percent of video views happened with the sound off, according to publishers like LittleThings and Mic, a Facebook-feed-specific, publisher-reported number that does not generalize cleanly to today's TikTok feed. Platforms responded to the silent-viewing reality directly: TikTok rolled out auto captions to help people who are hard of hearing or deaf and later made captions default on all eligible videos (https://newsroom.tiktok.com/en-us/introducing-auto-captions). The practical takeaway across all of it is dual-track design: the message must land on mute, and the audio must reward the viewers who turn it on.

Where It Shows Up in Content Work

For short-form teams, the failure this term diagnoses is content that only works in one mode. A clip whose entire point lives in the voiceover is invisible to the majority watching in public on mute; a clip that is all text with no audio payoff gives sound-on viewers nothing extra. The working method is to script twice, once as spoken audio and once as the on-screen text and visual sequence, then merge them so each track is independently comprehensible. If a video outperforms on TikTok but not on Reels with similar content, a sound-off gap in the feed is a common culprit.

What sound on versus sound off actually describes

Sound on and sound off are the two viewing states a short-form video has to survive. Sound on means the viewer hears the full audio mix; sound off means they read and watch in silence, with text overlays and visual sequencing carrying the entire message. The split is not stable, it shifts by platform, placement, time of day, and whether the viewer is in public or has headphones, which is why the design answer is to serve both rather than to bet on one.

The reason silence is the safer default to plan for is documented. The Verizon and Publicis study found 69 percent watch with sound off in public and that most caption users are not hearing impaired (streamingmedia.com), which reframes captions from an accessibility checkbox into a mainstream comprehension feature. A video that is comprehensible on mute serves both the silent majority in public and the accessibility audience, with no cost to the sound-on viewer.

The real numbers, and where they came from

The most defensible recent figures come from the Verizon Media and Publicis Media survey of 5,616 U.S. adults, reported by Streaming Media (streamingmedia.com): 69 percent watch with sound off in public places, and 80 percent of caption users are not hearing impaired. These are survey responses about behavior, not a single platform's playback data, which makes them a better basis for cross-platform design than any one feed's number.

The famous 85 percent statistic deserves a caveat rather than a citation as gospel. It traces to a Digiday report dated May 17, 2016 (digiday.com), which found that as much as 85 percent of video views happened with sound off, according to publishers like LittleThings and Mic. In that same report, Gretchen Tibbits, then COO of LittleThings, made the design point directly, verbatim, "But while the first three seconds are critical, the video also has to be designed to capture attention without needing sound," per Tibbits. The number is real but Facebook-feed-specific, publisher-reported, and a decade old, so it should be treated as historical context, not a current TikTok benchmark.

Platforms have since built for the silent reality. TikTok introduced auto captions to help people who are hard of hearing or deaf, then made captions default on all eligible videos in supported languages (newsroom.tiktok.com). The practical effect is that assuming your TikTok audience can read your video on mute is now the safer planning assumption, which raises the bar on getting the on-screen text right rather than leaving it to auto-generated lines.

How to audit a video for both modes

Watch your own video on mute first, before you ever watch it with sound. If the point is not clear silent, you have built for the minority and lost the majority who watch in public on mute, per the Verizon and Publicis data (streamingmedia.com). The mute pass is the single most useful test, and most teams skip it because they always review with sound on.

Then check the text against the safe zones and mobile distance. Captions that fall under the interface buttons, or text too small to read at arm's length on a phone, fail the sound-off viewer even when the words are technically present. TikTok's default captions help (newsroom.tiktok.com), but auto lines often break awkwardly, so reviewed placement still matters.

Finally, confirm the audio earns the unmute. Designing for mute does not mean abandoning sound; it means the audio should add emotion, pacing, or a payoff that rewards the viewer who turns it on, the dual-track standard. A video that is fully comprehensible silent and richer with sound serves both audiences, which is what Helen Lin's framing about consumer mindset points at (streamingmedia.com).

Common mistakes

The first mistake is putting the whole message in the voiceover. With 69 percent watching on mute in public per the Verizon and Publicis study (streamingmedia.com), an audio-only point is invisible to most of the audience in the contexts where they actually watch.

The second mistake is citing the 85 percent figure as a current universal. It is a Facebook-feed, publisher-reported number from a 2016 Digiday report (digiday.com), useful as historical context but not a present-day TikTok benchmark, and overstating it undermines an argument that the better-sourced 69 percent already supports.

The third mistake is relying on auto-captions and stopping there. TikTok's default captions (newsroom.tiktok.com) are better than nothing, but unreviewed lines break awkwardly and miss emphasis, so the text that carries a mute video still needs deliberate placement and editing.

Where a planning-first tool fits

When Superdirector analyzes reference videos, it examines both the audio strategy, the voiceover pacing, music, and sound-effect placement, and the visual layer, the text-overlay frequency, caption style, and visual hook structure, which matters because teams often optimize one mode and miss the other. Its production plans pair an on-screen text sequence with the voiceover script so the dual-track coverage is planned rather than discovered in the edit. Reviewing the captions for placement and emphasis on the final cut stays operator work.

Disclosure by Bell Chen, founder of Superdirector: the reference-analysis and production-plan features mentioned here are part of the product I build. The viewing-behavior figures in this piece are sourced from the linked Streaming Media coverage of the Verizon Media and Publicis Media study and the 2016 Digiday report, and the captions behavior from TikTok Newsroom; treat the tooling note as one input among several.

Related Terms

Frequently asked questions