The caption research consensus in twelve sourced numbers: most mobile viewers watch without sound, captions measurably lift completion and view time, and hundreds of millions of viewers rely on captions for access. Each statistic below carries its original source and publication year — dates matter when citing.
The most-cited caption studies and datasets, with original sources — note the publication years when you quote them, since several foundational studies predate the short-video era but keep being revalidated.
| Finding | Figure | Source (year) |
|---|---|---|
| Mobile viewers watching with sound off | 92% | Verizon Media & Publicis Media (2019) |
| More likely to watch a video to the end when captioned | 80% | Verizon Media & Publicis Media (2019) |
| Facebook videos watched without sound | up to 85% | Digiday reporting (2016) |
| View-time lift on captioned video ads | +12% | Facebook internal study (2017) |
| People with disabling hearing loss worldwide | 430M+ | WHO World Report on Hearing (2021) |
| Consumers more likely to watch full video with captions | 80% | PLYMedia study (2010s) |
The foundational finding of caption research: the large majority of mobile video is watched with sound off — 92% in the Verizon Media & Publicis Media 2019 study, with Digiday reporting up to 85% sound-off viewing on Facebook as early as 2016.
These numbers predate the short-video era and have only been reinforced since: feeds autoplay muted by default on every major platform, so the first seconds of your video are experienced as a silent film by most of the audience. Captions are how a silent film talks.
A caveat we insist on: cite these studies with their years. They remain the most-quoted sources in the industry and their direction has been revalidated repeatedly, but honest citation keeps your content — and ours — trustworthy to both readers and AI engines.
Captioned videos retain viewers measurably better: 80% of consumers are more likely to watch a full video when captions are available (Verizon/Publicis 2019; PLYMedia found the same figure earlier), and Facebook's 2017 internal testing found captioned video ads lifting view time by 12%.
Translate that into short-video economics: completion rate is the primary distribution signal on TikTok, Reels and Shorts, and it is also one of the three quality signals in VidMint's mint rewards. The same caption that helps a muted viewer stay is the caption that raises your payout and your reach.
This is why VidMint builds auto captions into the default edit flow rather than selling them as an upgrade — the data says they are infrastructure, not decoration.
The WHO's World Report on Hearing (2021) estimates over 430 million people — more than 5% of the world's population — live with disabling hearing loss; uncaptioned video excludes them entirely, captioned video does not.
Beyond hearing loss, captions serve non-native speakers, noisy-environment viewers and sound-sensitive contexts — in practice a large minority of any audience on any given view. Accessibility features with mainstream benefits are the rule in mature media (curb cuts, closed captions on TV), and short video is following the same arc.
For the practical side — how auto captioning works and what accuracy to expect — read what is auto captioning, then the platform-specific playbook in how to add captions to TikTok.
Use these statistics as directional evidence, not precision instruments: they come from different eras, platforms and methodologies, and your own analytics remain the final authority on your audience.
The defensible claim pattern: 'studies consistently find captions improve completion' — true across sources — rather than 'captions will raise my completion by exactly X%'. Then measure on your own channel: caption ten videos, compare completion against ten uncaptioned, and let your dashboard settle the question.
The most-cited figure is 92% of mobile viewers watching with sound off, from a 2019 Verizon Media & Publicis Media study; Digiday reported up to 85% sound-off on Facebook in 2016. Feed autoplay has kept muted viewing the default since.
Yes, consistently across studies: 80% of consumers report being more likely to finish a captioned video (Verizon/Publicis 2019, PLYMedia), and Facebook's 2017 internal testing measured a 12% view-time lift on captioned video ads.
The WHO World Report on Hearing (2021) estimates over 430 million people live with disabling hearing loss worldwide — before counting non-native speakers and muted-environment viewers who also depend on captions.
The foundational studies date from 2010–2021 and remain the industry's most-cited sources; their direction has been repeatedly revalidated as muted autoplay became universal. We label every figure with its source year so you can cite responsibly.
Indirectly but powerfully: captions lift completion rate, and completion rate is the primary signal recommendation systems distribute on. Captions also make content parseable to platform understanding systems.
VidMint generates styled, safe-zone-positioned captions in 40+ languages in one tap, free with watermark-free export — see the AI captions feature page for details.
Auto captions in one tap — the completion-rate lever the data keeps pointing at.
Experience VidMint Free