Subtitling and Captioning: How Much Demand Has the Video Boom Created?

My first subtitling job was an 11-minute video. I told myself it would take forty minutes.

It took three hours and twenty minutes. Then the client wrote back saying the timing was off in 14 places, and I learned that subtitling is not "listen and type." It is closer to editing than to typing, and nobody warns you about that part.

---

Why demand grew in the first place

Most people now watch video with the sound off. On a bus, at a desk, in bed next to someone sleeping. On short-form platforms a large share of viewing happens muted, and a video without on-screen text loses people in the first two seconds.

So creators need captions. Not just big channels, small ones too.

Language is the second push. A Hindi channel that adds English subtitles reaches people who follow the content but would never search for it in Hindi. One client of mine had about 38,000 subscribers, and a few weeks after adding English text he told me a noticeable slice of his new views came from outside India. I can't verify his analytics, so take that as his claim, but it was enough to make him order subtitles for everything after.

The third source is less exciting and pays more steadily: online course creators and companies with training videos. Many of them need captions for accessibility, and some have it written into their requirements. These clients want clean work and tend to pay on time. They also send back corrections, a lot of them.

---

The first job, and the mistake I made

The client ran a cooking channel. I sent a message on Instagram offering to subtitle two videos for free so he could judge the quality.

The first video was 6 minutes 40 seconds. I ran the auto-captioning tool and figured I would just tidy it up.

That was wrong. The tool turned "kadhai" into "karaai," "tadka" into "tada," and "pyaaz" into "pyas." I counted 61 errors in that single video, and these are not subtle mistakes. They sit on screen in large letters where every viewer sees them.

So I treated the auto output as a rough draft and listened to every line. It took 2 hours and 5 minutes.

He liked it. The following week he sent nine more videos, and I remember feeling oddly relieved and slightly dreaded at the same time, because I had no idea how I would fit that much listening into my evenings.

---

Where the real work hides

Not in typing. In timing.

A subtitle that stays on screen for less than a second cannot be read. One that is too long pulls the viewer's eyes off the video and onto the text. My own rule settled at about 42 characters per line and never more than two lines.

Where you break a line matters more than I expected. "I went to the market / yesterday" reads differently from "I went to the / market yesterday," and the second one makes viewers stumble. I didn't notice this at all in the first month.

Translation adds another layer. Go word for word and the English sounds stiff. Go by meaning and the line grows longer than the timing allows. I still get stuck on this, and I suspect experienced people do too.

---

Tools, hours, and money

Subtitle Edit is free and shows the audio waveform, so you can see where speech starts rather than guessing. Aegisub does similar things but feels dated to me. YouTube Studio's editor is fine for a quick job.

On time, here is what my first two months looked like. Same-language captioning took about 8 to 12 minutes of work per minute of video. With translation, 15 to 25 minutes. I am faster now, but not dramatically.

On pay: Rs 40 to Rs 90 per video minute from my first six clients. The best month brought in Rs 6,140, and dividing by hours worked gave roughly Rs 110 an hour. Not bad for a skill you can pick up in a couple of weeks. Not life-changing either.

I used to think knowing two languages was the main thing clients wanted. Actually, that's not quite right. Bilingual helps you get in the door, but what earned me repeat work was turnaround. The client who gets a file within 24 hours comes back. The one who waits four days finds someone else.

---

Who this suits, and who should skip it

If listening to the same few seconds again and again makes you want to throw the laptop, skip it. I once replayed a 4-second clip nine times because I couldn't catch a single word. (It was "mawa." Another cooking video, naturally.)

If you already do voice-over work or translation, subtitling fits next to both, and some clients ask for the two together. For people learning video editing, it is an easy way in, since the client's footage lands in your hands and bigger requests tend to follow.

I honestly don't know whether auto captions will get good enough in two years to shrink this market. English has improved a lot already. Hindi and mixed-language speech still lag. So for now I treat subtitling as one slice of my income, not all of it.

Anyway, one client still sends me four videos every week, and last night one of them had a word I haven't been able to place. I've been wondering since 11 p.m. whether it was "chaasni" or something else entirely...