Most marketing teams first meet captions as a tactic for silent social feeds. Captions now also appear in procurement checklists and public-sector briefs. If your agency delivers a film to a city council, a university or a federal client, the caption file is part of the deliverable.
Why captions matter for business video
The audience is large. The World Health Organization’s fact sheet (March 2026) says about 430 million people, over 5% of the world’s population, need rehabilitation for disabling hearing loss, and projects more than 700 million by 2050.
The W3C Web Accessibility Initiative adds that captions also serve people who process written information better than audio. That includes viewers in noisy places, viewers watching in a second language and anyone scrolling with the sound off. For a brand team, captions make the message reachable, meet the standards clients expect, and give you a clean text version of every video.
The WCAG success criteria for video
The Web Content Accessibility Guidelines (WCAG) are the technical standard behind most laws below. Guideline 1.2 covers time-based media, and WCAG 2.1 and 2.2 share the same 1.2 criteria.
Source: W3C, How to Meet WCAG 2.2 (Quick Reference); W3C, Understanding SC 1.2.2.
Most laws point to Level AA, so a typical covered video needs captions and audio description, and a livestream needs live captions. W3C’s guidance on 1.2.2 says captions go beyond dialogue: they identify speakers and include meaningful sound effects.
The rules by market (US, EU, UK, Canada)
This section is general information, not legal advice; ask a qualified lawyer how these rules apply to you.
US: ADA Title II web rule
The Department of Justice published its web and mobile app rule for state and local governments on April 24, 2024. The standard is WCAG 2.1 Level AA, and ADA.gov names captions for videos as an example of what it covers. An April 2026 interim final rule delayed the compliance dates by one year without changing the requirements. The DOJ noted that generative AI “does not yet reliably automate the remediation of inaccessible content at scale.”
Exceptions include qualifying archived content and social media posts made before the compliance date. ADA.gov adds that the effective communication duty still applies, for example captioning an archived video when someone asks.
US: Section 508
Federal agencies follow the Revised Section 508 Standards, which Section508.gov says work “in concert with WCAG 2.0 success criteria.” Agencies need captions for prerecorded and live media, audio description or a media alternative, and player controls for captions and description at the same menu level as volume.
EU: European Accessibility Act
The European Accessibility Act (Directive (EU) 2019/882) applies from 28 June 2025. Covered services include services providing access to audiovisual media, e-commerce, e-books and consumer banking. The directive describes access services as including subtitles for the deaf and hard of hearing, audio description, spoken subtitles and sign language interpretation. Microenterprises providing services (fewer than 10 staff and no more than EUR 2 million turnover or balance sheet) are excluded.
UK: Public sector regulations and the Equality Act
GOV.UK says websites and apps covered by the Public Sector Bodies (Websites and Mobile Applications) (No. 2) Accessibility Regulations 2018 should meet WCAG 2.2 AA. Prerecorded media published before 23 September 2020 and live media are exempt. The regulations build on the Equality Act 2010 duty to make reasonable adjustments.
Canada: Ontario’s AODA
Since January 1, 2021, Ontario’s public sector and organizations with 50 or more employees must make public websites meet WCAG 2.0 Level AA. Live captions (1.2.4) and prerecorded audio description (1.2.5) are excluded, so prerecorded captions remain in scope.
Source: ADA.gov and 2026 IFR; Section508.gov; EUR-Lex; GOV.UK; Ontario.ca.
Closed captions vs open (burned-in) captions
W3C defines closed captions as captions viewers can turn on and off, and open captions as captions that cannot be turned off.
- Closed captions travel as a separate file. They can be fixed after publishing, swapped by language and read as text. GOV.UK’s publishing guidance tells government teams to use them.
- Open captions are rendered into the picture, so they show everywhere, including muted autoplay feeds. Fixing them means a re-export, and they can collide with platform buttons and graphics.
For most campaigns the answer is both. Our social media video editing guide covers how captions sit inside vertical safe zones.
Caption file formats (SRT, WebVTT) and platform support
W3C lists WebVTT as the most common web format, alongside SRT and TTML.
- WebVTT (.vtt): MDN says the file must begin with “WEBVTT”, use UTF-8 and the MIME type text/vtt. Timestamps use a period before milliseconds (00:00:01.000), cue settings control position and alignment, and the file attaches to HTML video with the track element.
- SRT (.srt): SubRip is a simple plain-text format. YouTube supports basic SRT only, recognizes no style markup and requires plain UTF-8.
Source: YouTube Help; Meta for Developers, Video Captions; TikTok Ads Help; MDN, WebVTT.
Platform features change often, so confirm LinkedIn’s current caption upload rules in its help center before delivery.
Caption quality standards: accuracy, sync, speakers and sound
The FCC adopted caption quality standards for TV in February 2014. They are written for broadcasters, but the four tests suit any video:
- Accuracy: captions accurately convey dialogue and sounds.
- Completeness: captions run from beginning to end.
- Timing: captions do not lag far behind dialogue.
- Placement: captions do not block important information.
The Described and Captioned Media Program (DCMP) Captioning Key adds detail:
- No more than two lines per caption is preferred.
- Minimum duration of 40 frames (1 second and 10 frames), maximum of 6 seconds.
- White, medium-weight, sans serif text with a drop or rim shadow.
- Sound effects in brackets, naming the source, omitted when the source is visible.
- Music marked with ♪; speaker names in parentheses when placement does not make the speaker clear.
- Captions inside the safe zone, never covering graphics.
W3C warns that automatic captions often need significant editing, and that missing one word such as “not” can reverse the meaning. YouTube says its automatic captions “may misrepresent the spoken content” and tells creators to always review them.
Captions vs subtitles vs transcripts
- Captions are in the audio’s language and include speech, meaningful sounds and speaker IDs. W3C calls them an accessibility requirement.
- Subtitles are translations. W3C notes some regions use the word for both, and Section508.gov says dialogue-only subtitles do not meet the standard.
- Transcripts are plain text without time codes. They do not replace captions for video with sound, but WCAG requires them for prerecorded audio-only content such as podcasts.
A clean caption file gives you a transcript almost for free, which helps with corporate video such as town halls and training.
Audio description basics
Audio description is a spoken explanation of important visuals for people who are blind or have low vision. WCAG 1.2.5 requires it at Level AA for prerecorded video. Section508.gov advises:
- Place descriptions in natural pauses, not over dialogue, music or sound effects unless necessary.
- Describe key visuals first, such as titles, speaker changes and end credits.
- Build description into the main file or deliver a separate, clearly labeled version.
- Where a platform has no description control, publish a version with captions and description.
The cheapest description is the one you barely need. If a product name, price or URL appears on screen, have the voiceover say it too. GOV.UK says important information must be explained verbally, not only shown.
A workflow for adding captions in the edit
This is the order we recommend for commercial video editing projects.
Write it into the brief
List which versions need caption files, which need burned-in captions, and whether description is in scope.
Script for accessibility
Make key on-screen text spoken, and note speaker names and important sounds.
Lock the picture first
Captioning before lock means re-timing every cue later.
Generate a draft transcript
Automatic transcription is a fine start, treated as a draft.
Edit for accuracy
Check every word, especially names, product terms, numbers and negatives.
Add speakers and sound
Identify speakers, bracket meaningful sounds and mark music.
Time and break lines
Keep to two lines, respect durations and break at natural phrases.
Style burn-ins per ratio
Keep captions inside each platform’s safe area and clear of lower thirds.
Export sidecar files
Export SRT and WebVTT from the final timeline in UTF-8, named to platform rules.
QC on the real platform
Upload privately, turn captions on and watch end to end on desktop and mobile.
Deliverables matrix
Use this as a starting point when scoping a multi-platform campaign or an online video ad set.
Source: W3C WCAG 2.2; Meta for Developers; YouTube Help; Section508.gov.
Common caption mistakes
- Publishing raw auto-captions. Federal guidance says they would not meet the minimum standard alone.
- Dialogue only. Without speaker names and sounds, captions become subtitles.
- Captioning before picture lock. Every later trim knocks cues out of sync.
- Burn-ins under the interface. Captions behind buttons or over lower thirds fail the placement test.
- Wrong encoding or file name. YouTube expects UTF-8 SRT; Facebook rejects files that break its naming pattern.
- Text-only key facts. A price shown but never spoken creates a description gap.
- One position for every ratio. A 16:9 caption position rarely works on a 9:16 cut.
How Cut Pro Media handles captions
Captions are part of our multi-platform package: the standard edit plus 16:9, 9:16, 1:1 and 4:5 exports with intentional reframing per ratio, captions as SRT/VTT files and burned in, end-card CTAs per platform, UTM-ready slates and multiple thumbnail frames. It is custom quoted and delivered in 2–4 working days.
We edit in Adobe Premiere Pro, with DaVinci Resolve handover via XML. You review on Frame.io, Vimeo or a private YouTube link, a good place to test caption files on a real player, and caption styling follows your brand guidelines.
To be clear about scope: we do not certify accessibility compliance, offer audio description or translation, or guarantee that a video meets any law. Your legal or accessibility lead should sign off on compliance.
Common questions
Sources: W3C WAI, Understanding SC 1.2.2 Captions (Prerecorded), WCAG 2.2; W3C WAI, How to Meet WCAG 2.2 Quick Reference; W3C WAI, Captions/Subtitles (17 September 2024); ADA.gov, Fact Sheet: New Rule on Web Content and Mobile Apps of State and Local Governments (April 2024, updated 2026); U.S. DOJ, Title II Web Rule Interim Final Rule (effective April 20, 2026); Section508.gov, Video and Other Synchronized Media (October 2026); EUR-Lex, Accessibility of products and services summary and Directive (EU) 2019/882; GOV.UK, Understanding accessibility requirements for public sector bodies (30 September 2024); GOV.UK, Videos publishing guidance; Ontario.ca, How to make websites accessible (September 21, 2026); MDN, WebVTT format (November 4, 2025); YouTube Help, Supported subtitle and closed caption files; YouTube Help, Use automatic captioning; Meta for Developers, Video Captions reference; TikTok Ads Help, Captions and Translations for Spark Ads (January 2026); FCC, Closed Captioning Quality Standards (February 20, 2014); DCMP, Captioning Key: Text and Captioning Tip Sheet; WHO, Deafness and hearing loss fact sheet (3 March 2026).
