Two viewers press play on the same film. One reads white text at the bottom of the screen because the dialogue is in a language she does not speak. The other reads similar text because he is deaf and needs every sound described. Both call it "subtitles," and that shared word causes a surprising number of production mistakes. Understanding closed captions vs subtitles is the first step to ordering the right service, staying inside accessibility rules and reaching the audience you actually want.
Closed captions vs subtitles: the core difference
Subtitles assume the viewer can hear the audio but cannot understand the language. They translate or transcribe the dialogue, and nothing more. Closed captions assume the viewer cannot hear the audio at all, or cannot hear it well. They include the speech, but also speaker labels, music cues, laughter, a door slamming and any other sound that carries meaning in the scene. The word "closed" simply means the viewer can switch them on and off, as Wikipedia's overview of closed captioning explains.
In practice, the line blurs. Many platforms label everything "CC," and some regions use "subtitles" for both. The most reliable test is to ask what the viewer needs: translation, or access to sound. Once you know that, the rest of the decision follows.
Where SDH subtitles fit
There is a third format that confuses people even more. SDH subtitles, short for subtitles for the deaf and hard of hearing, look like regular subtitles but include the sound descriptions you would expect from captions. Streaming services use them when a film is delivered in another language and still needs to be accessible. They are especially common on disc releases and in markets where a separate caption track is not standard.
If you produce content for several countries, ask your provider which of these formats each platform accepts. A single master file rarely satisfies every distributor.
Which one does your video need?
Start with your audience. If you are publishing a training video for a global workforce, subtitles in several languages let employees follow along in the language they think in. If you publish for the general public, captions are often a legal requirement as well as a courtesy. The W3C guidance on captions is a good reference for what accessible video should include, and many national broadcasting and web rules build on it.
There is also a commercial argument. A large share of people watch social video with the sound off, on trains, in offices and in bed next to a sleeping partner. Burned-in or toggleable text keeps them watching. Search engines cannot watch a video, but they can read a transcript, so captions help discoverability as well.
For most businesses the answer is both. Captions in the original language make the video accessible, and subtitles in target languages open new markets. Producing them from the same time-coded script keeps the cost reasonable.
Live programmes add one more layer to that decision. A council meeting, graduation or clinic briefing streamed online may need captions for accessibility and a spoken interpreter for residents who follow another language. Teams that book interpreting services alongside captioning give every viewer a way in, not just those who can read along in English. Treating the two as one accessibility plan avoids gaps that neither service covers alone.
How the work is actually done
Good subtitling is not transcription with a dictionary attached. A professional subtitler condenses speech so it can be read in the time available, usually capping lines at around forty-two characters and keeping reading speed comfortable. They break lines at natural phrase boundaries, so a sentence does not split in the middle of a name. They also check cultural references, because a joke that works in one language may need a different one in another.
Captioners work to different priorities. They aim for completeness and timing, and they identify speakers when the camera cannot. If you are unsure how the two disciplines relate to written transcripts, this comparison of transcription and captioning differences lays it out clearly.
Quality checks worth insisting on
Whatever you commission, ask for a review pass by a second linguist, a technical check that the files play correctly on your platform, and a style guide for recurring terms. Consistency matters most on long projects. A character who is called three different names across a season is a classic failure, and this article on keeping TV series subtitles consistent shows how glossaries and translation memories prevent it.
Finally, watch the result yourself on a phone, since that is where many people will see it. Text that looks perfect on a monitor can be unreadable on a small screen.
Common mistakes to avoid
The most frequent error is relying on automatic captions alone. They have improved, but they still stumble on names, accents, technical terms and overlapping speakers, and a caption that gets a product name wrong can do real damage to credibility. Treat automatic output as a first draft that a human must correct. Another mistake is forgetting that translated text runs longer than English. Leave room in your layout, and expect the subtitler to condense rather than cram.
Choosing a partner
Ask for samples in your language pair, find out which file formats they deliver, and confirm turnaround times before the deadline pressure arrives. A reliable team will ask about your platform, audience and tone before quoting, because those answers decide whether you need captions, subtitles or both. Get that choice right at the start and the rest of the project becomes far simpler.
