How to Compare the Level of Subtitle, Sign Language, and Audio Description Support in Broadcast Catch-Up Services

A catch-up service should not receive a high accessibility rating merely because its help page mentions subtitles, sign language, or audio description. The more useful question is how much of its catalog supports each feature, whether the quality is sufficient for real viewing, and whether viewers can find and activate it on the devices they normally use.

Coverage can vary sharply between recent and archived programmes, individual episodes of the same series, in-house productions and licensed titles, or the broadcaster’s website and smart TV application. A service may provide captions for most current dramas but almost none of its older archive. Another may offer audio description on selected flagship programmes while making those versions difficult to locate.

Accessibility should therefore be measured through five areas: catalog coverage, caption quality, sign-language presentation, audio-description quality, and device-level usability. Testing a representative sample provides a more reliable result than counting accessibility icons on a homepage.

Calculate the Coverage Rate for Each Feature

Begin with the proportion of programmes that actually support the required feature. Do not combine subtitles, signing, and audio description into one general accessibility score because each serves a different need.

Use a simple calculation:

Coverage rate = Supported programmes ÷ Programmes tested × 100

If captions are available on 42 of 50 tested programmes, the measured caption coverage is 84%. If audio description is present on six, its coverage is 12%. These figures reveal far more than a statement saying that the platform “supports accessibility.”

Korean disability broadcasting rules treat closed captions, audio description, and Korean Sign Language as separate access services and establish different provision requirements according to broadcaster category. This reinforces the need to measure each feature independently rather than treating one as a substitute for another.

Ofcom’s 2025 access-services report offers a useful example of the gap between platform capability and catalog coverage. Among responding on-demand providers, 88.9% offered subtitles somewhere on their services, but the proportion offering audio description was 40% and the proportion offering signing was 28.6%. Looking only at providers that supplied these services, subtitles covered 85.1% of programme hours, audio description 21.7%, and signing 3.8%. Ofcom notes that participation was voluntary and the responding providers varied, so these figures are indicative rather than a complete market census.

Build a test sample based on what the household actually watches. A useful set may include recent drama, entertainment, news, documentary, sports replay, an older archived episode, and licensed content produced by another company. Test at least two or three episodes from a continuing series because accessibility files may be missing from an individual installment.

Keep recent and older content in separate columns:

Content group	Titles tested	Captions	Sign language	Audio description
Current programmes	20	19	3	7
Previous seasons	15	11	1	2
Archive titles	15	6	0	0

A service with excellent support for new releases but poor archive coverage may suit viewers who follow current broadcasting while remaining unsuitable for those who regularly watch older programmes.

Distinguish Closed Captions From Dialogue Subtitles

Dialogue appearing as text does not automatically qualify as complete accessibility captioning.

Translation subtitles are often designed for viewers who can hear the soundtrack but do not understand the spoken language. They may translate the dialogue without identifying who is speaking or describing sounds that carry narrative meaning.

Closed captions should communicate the important information available through the audio track. W3C describes captions as synchronized text or visual alternatives for both speech and non-speech information needed to understand the content. This can include speaker identification, sound effects, music, laughter, and the location of an off-screen voice.

A proper test should use scenes containing more than a simple conversation. Select a section with overlapping speakers, music, environmental sounds, and an important off-screen event.

Review whether the caption track:

• Identifies a speaker when the image does not make the identity obvious.
• Describes meaningful sounds such as a door opening, an alarm, applause, or footsteps.
• Indicates music when its presence or mood affects the scene.
• Preserves essential numbers, names, and technical terms.
• Remains synchronized without appearing too early or disappearing too quickly.
• Avoids covering faces, signs, scores, or other important visual information.

W3C also notes that captions should not obscure relevant video content. A technically complete caption may still be difficult to use if its position hides a speaker’s face, an on-screen name, or information displayed at the bottom of a news broadcast.

Check the complete programme rather than the opening minute. Captions sometimes disappear during archive inserts, previews, studio links, advertisements, or material supplied by another rights holder. These gaps can be especially damaging in news and documentary programmes because the inserted footage may contain the central evidence being discussed.

Customization affects usability as well. Test whether the player allows changes to text size, background, contrast, and position. Large captions that are readable on a phone may occupy too much of a television screen, while a fixed small font may be unusable from normal living-room distance.

Evaluate Sign Language as a Visual Service

Captions and sign-language interpretation are not interchangeable. Sign languages are independent languages with their own grammar and expressive structure. W3C explains that sign-language interpretation can communicate intonation, emotion, and other information in ways that differ from written captions, and it may be essential for viewers whose primary language is a sign language.

First determine how the service provides the interpretation. It may be permanently embedded in the video, offered as a separate signed edition, or delivered through a synchronized secondary video that the player overlays on the main programme.

Each method creates a different viewing experience.

An embedded interpreter is usually easy to play but cannot be hidden or resized. A separate version can provide a larger interpreter image but may be listed away from the ordinary episode page. A selectable overlay offers greater flexibility only when the player and device support it reliably.

Judge the interpreter window according to practical visibility. Hands, arms, upper-body movement, and facial expressions need to remain clear. A small corner image may satisfy a basic availability check while becoming difficult to understand on a phone or distant television.

Also inspect its position. The interpreter should not cover subtitles, sports scores, names, graphics, or other important visual information. If the window changes location during the programme, verify that those changes do not cause it to disappear or obscure another essential element.

Test whether the signed version is available on demand rather than only during the original television transmission. A schedule symbol can confirm that a signed broadcast occurred, but it does not prove that the interpreter was retained in the catch-up copy.

The service should label signed content consistently. Search for the programme title and terms used by the platform for sign language, then check whether filters or category pages return complete episodes rather than clips and promotional material.

Judge Audio Description by Relevance and Timing

Audio description should explain significant visual information that is not already communicated through dialogue or the original soundtrack.

W3C describes audio description as synchronized spoken information about actions, characters, scene changes, and on-screen text. Standard description is inserted during natural pauses in dialogue so that it adds necessary visual information without covering the original speech.

A label showing that description is available does not reveal whether it is useful. Play a visually important scene with limited dialogue and compare the standard and described audio tracks.

A strong track should clarify:

• Who enters or leaves the scene.
• Significant gestures, expressions, or silent reactions.
• Changes of location or time.
• Important objects and physical actions.
• Written text, messages, captions, or signs shown only on screen.
• Visual information needed to understand a joke, conflict, or plot development.

Information density requires balance. A sparse track may omit the action that explains the next line of dialogue. An overly detailed track may describe decorative elements while rushing past a more important expression or covering part of the original soundtrack.

Timing is equally important. The description should fit naturally into available pauses. When the narrator speaks over essential dialogue or an important sound cue, the track may become harder to follow even though it contains accurate information.

Check voice clarity and mixing through the equipment that will be used regularly. The narrator should be audible without sounding detached from the programme or overpowering the dialogue. If the description becomes difficult to hear on television speakers but works through headphones, record that as a device or mixing limitation rather than marking the feature simply as present.

Test Discovery and Controls on Every Main Device

An accessible track provides little practical value when viewers cannot find or activate it.

Test the service separately on its website, mobile application, smart TV app, streaming box, and set-top box where relevant. Ofcom reported continuing technical differences across third-party and older platforms, including cases where devices could not receive or play particular access services. Its 2025 data showed that availability also varied by device category.

Start at the search screen rather than navigating directly to a known accessible title. Determine whether the service allows users to filter for captions, audio description, or sign language. Ofcom’s electronic programme guide reporting treats highlighting and filtering of programmes with audio description or signing as an important accessibility feature because viewers need to discover suitable content before playback begins.

Check whether accessibility labels appear in all the necessary places:

• Search results
• Programme information pages
• Episode lists
• The playback screen
• Audio and subtitle menus
• Continue-watching rows

A series-level icon is not enough when support differs by episode. Each installment should identify which tracks are available before the viewer starts playback.

Once the feature is enabled, move to the next episode. Confirm whether the setting is remembered or must be selected repeatedly. Pause, rewind, exit, and resume the programme to see whether the track remains active.

Casting introduces another possible failure point. A caption or audio-description option available on the phone may disappear when playback moves to a television. The receiving device may expose a separate menu or may not support the same track.

Households that use the service across several screens should also review How Simultaneous Streaming Limits Affect Family Viewing When Watching Broadcast VOD on Multiple Devices. A platform can provide excellent accessibility on one device while its account restrictions prevent different family members from using their preferred accessible versions at the same time.

Screen-reader access should be tested throughout the process. Playback buttons, episode labels, menus, progress controls, and audio-track names need understandable text labels and a predictable focus order. A screen reader that announces several buttons only as “unlabeled” can make the player difficult to operate even when the programme includes audio description.

Create a Weighted Score Based on Actual Needs

Do not give equal weight to every feature when only one is essential to the viewer.

A person who relies on captions may assign 50% of the score to caption coverage and quality, 25% to device usability, 15% to search and labeling, and 10% to the consistency of settings. A viewer who depends on audio description should make that service the largest part of the score.

One possible comparison is:

Criterion	Weight	Service A	Service B
Required-feature coverage	35%	8/10	6/10
Quality and synchronization	25%	7/10	9/10
Device support	20%	9/10	5/10
Search and labeling	10%	8/10	4/10
Setting retention	10%	6/10	8/10

The service with the most accessible titles overall may not be the best choice if those titles do not include the programmes the viewer follows. Score the test catalog first, then use broader platform statistics only as supporting evidence.

Repeat the test periodically. Archive catalogs change, new applications are released, and accessibility files may be added or lost when distribution rights move between providers.

Report Missing Support With Reproducible Details

When a feature fails, record enough information for support staff to reproduce the problem.

Include the programme name, season, episode, device model, operating system, app version, account region, and the exact missing option. State whether the feature appears on another device or in the original television broadcast.

For example:

“Episode 4 offers Korean audio description in the web player, but the audio menu in the smart TV application contains only the standard track. The app was updated before testing. Is the described version unsupported on this television model or temporarily missing?”

A detailed report distinguishes a platform limitation from an incorrectly packaged episode. It also gives the provider evidence that can be passed to technical or content teams.

The strongest catch-up service is not the one that places the largest number of accessibility claims on its help page. It is the one that provides the required feature across a high proportion of relevant programmes, maintains usable quality throughout each episode, labels it clearly, and preserves it on the viewer’s preferred device.