What Captions Cost and What Moves the Price Up or Down
A plain look at what goes into the cost of captions and on-screen text for event highlights, from running time and audio quality to speaker names and the formats you need.
Priya Nair · · 6 minutes to read
Where the cost of captions really sits
When people ask what captions cost, they usually picture the typing. In practice the typing is the smaller part. Most of the time goes into listening, checking and fitting the words to the picture so that a viewer on a small screen can read them without effort. A caption that appears too early, lingers too long or covers a speaker's face costs the audience attention, and fixing it later costs you hours.
I teach this inside the captions and delivery module, and the first habit I ask learners to build is to treat captioning as a measured task. Before you quote a price to a client or budget your own evenings, count the minutes of speech you will keep, not the minutes of footage you received. A two-hour panel that becomes a ninety-second recap holds far less spoken material than the raw length suggests.
That distinction changes every estimate that follows, and it is the first thing worth getting right.
Running time of speech, not running time of footage
The single biggest driver of effort is how much speech survives into the final cut. A recap built mostly from crowd shots, music and short reactions needs very little captioning. A recap built around three speakers explaining why an event mattered needs careful, line by line work, because every sentence must match the spoken words and stay true to what the person meant.
Work out the speech minutes after your clip selection is locked, not before. If you caption first and then trim the edit, you will rewrite timings twice. I ask learners to finish pacing, then transcribe only the lines that remain, and then time them. That order alone often saves a third of the work on a typical short project.
It also keeps the budget honest. A client who asks for a few extra seconds of a speaker is really asking for more captioning, and you can say so plainly.
- Count only the spoken lines that stay in the cut
- Lock the edit before you caption
- Treat extra speaker seconds as extra caption work
Audio quality and how clearly people speak
Clean audio is cheap to caption. Muffled audio, room echo, applause over the top of a sentence or two people talking at once is expensive, because you replay the same few seconds again and again to be sure of the words. Guessing is not acceptable when captions must match what was said.
This is why we ask learners to check audio during the source footage module before any editing begins. If a key quote is barely audible, you can decide early to drop it, pick a clearer take of the same point or ask the organiser for a better recording. Finding that out at the captioning stage is the costly way to learn it.
Accents, technical vocabulary and names of places or products also slow things down, since each uncertain term needs a check with the organiser. A short list of correct spellings supplied up front is worth real money.
- Replaying unclear lines is where hours disappear
- Ask for name and term spellings before you start
On-screen text: titles, names and lower thirds
Captions follow speech, but on-screen text is a design job. Speaker names, a title card, a date or a closing line each need a readable size, a calm position and enough time on screen to be read twice. On a vertical phone screen the safe area is narrower than most beginners expect, and text placed near the edges can be hidden by app interface elements.
Every extra text element adds a small cost, and those small costs add up quickly. Ten speaker labels are ten decisions about wording, position and timing. I suggest learners agree a short list of text elements with the client at the start, such as speaker names and one title, and treat anything beyond that as a separate request.
Consistency saves effort too. Set one font, one size range and one colour treatment once, and reuse them across every version you deliver.
- Agree the list of text elements before you begin
- Keep one style across all versions
Formats multiply the work
A horizontal, a square and a vertical export are not one video in three sizes. Captions that sit comfortably in a wide frame may be too small or too crowded in a vertical one, and line breaks often need redoing so each caption stays short enough to read at a glance. Each format means a layout check and often a retime.
If a client wants all three, plan for the captions to be adjusted in each, not simply scaled. Starting from the tightest format, usually vertical, and widening from there tends to produce cleaner results than shrinking a wide layout down.
Be clear with yourself about which versions are truly needed. A recap that will only be shared in a feed may need one format done well rather than three done quickly.
Revisions, feedback and a fair way to budget
The last cost driver is how many times the words change. Captions touch the meaning of what a speaker said, so clients often want to adjust a phrase or a name. That is reasonable, but it should be planned for. In our course the final project includes two rounds of instructor feedback, and I recommend the same rhythm for client work: a first full pass, one round of comments, then a final check.
Build a small margin for corrections into any estimate, and say how many rounds are included. Changes after the final check are new work. This keeps the relationship clear and protects the quality of the finished captions.
When you put the pieces together, a fair price reflects speech minutes kept, audio difficulty, the number of text elements, the formats delivered and the rounds of revision. None of these are hidden, and explaining them calmly is usually the best way to earn trust.
- Two feedback rounds is a sensible default
- Corrections after the final check count as new work
The cost of captions comes down to a few visible factors: how much speech remains in the cut, how clear the audio is, how many on-screen text elements you promise, how many formats you deliver and how many revisions you allow. Count these early, explain them plainly and your estimates will hold up. Careful captions that match the spoken words are worth the time they take, because they protect what each speaker actually meant.