Guide
Caption tools bill per clip, so count last month first
The meter is how many clips you process in a month, with certain features gated by tier rather than by volume. The number you need is not your plan for next quarter. It is what you actually posted in the last thirty days, and that is almost always smaller. This page covers the meter, the agency case that breaks it, and what the same stage costs when the captions arrive inside the cut. By the end you will know which tier to buy and which number forced it.

- The unit a captioning tool tiers on
- Clips a month
- What styling one clip takes once the tool does it
- ~90 sec
The meter is clip count, with capabilities gated by tier
You upload a clip, the tool transcribes and styles the captions, and the plan allows a certain number of those a month. Read the live page for the current tiers, because both the counts and the gates move.
Two constraints run at once. The volume you need, and the feature you cannot work without. Brand-consistent styling, longer clips and export options often sit above the volume tier a solo creator would pick.
Buy the higher of the two tiers and know which one forced the decision. If it was the feature, your volume headroom is decoration you are paying for.
The unit here is friendlier than most, because a clip is a thing you can count in your own analytics rather than a quantity you have to model.
Count what you posted, not what you intend to post
Open your own account and count the last thirty days. That takes ninety seconds and it is the single most useful thing on this page.
One clip a week is about four a month. Three a week is thirteen. Daily is thirty, and daily is a cadence most accounts describe as their plan and few sustain past six weeks.
The gap between the intended number and the counted number is where over-buying lives. Buy for the counted number and top up when the counted number changes, because it is easier to move up a tier than to recover a quiet quarter.
If the count is under about ten, check whether the free tier covers you. A specialist captioning tool at low volume is often a habit rather than a purchase.
Clips a month at each cadence people claim
Count the last thirty days in your own analytics and pick the bar you actually hit, not the one you meant to.
- One a week4 clips
- Three a week13 clips
- Five a week22 clips
- Daily30 clips
This is the whole editor
Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.
The meter starts after the expensive minutes are already spent
Captioning is the last stage. Everything before it stays yours: deciding what to say, recording it, choosing which take survives, and cutting it so the point arrives early.
That matters because a caption tool measured against those minutes looks cheap and against the finished ad looks incomplete. Twenty minutes of hand-captioning becoming ninety seconds is a real saving, and it is a saving on the fastest stage of the job.
So the honest question before buying is where your own minutes go. If most of them go into styling text, this is the correct purchase and you will feel it in week one.
If most of them go into deciding the structure, a captioning tool leaves that untouched and you will be back on this search in two months looking for something else.
Where the minutes go before a caption tool opens
The meter starts at step four. If your time disappears in steps one to three, this is the wrong line item to optimise.
Decide what to say
The stage nobody can automate for you
Record it
Takes, retakes, the one that survives
Cut for structure
Get the point into the first three seconds
Caption and style
About ninety seconds with a tool
Agency work counts per client, and that is where tiers break
Five clients at ten clips a month is fifty clips, and the tier that covers a solo creator does not survive contact with that number.
Ask two questions before an agency purchase. Does the plan allow separate brand styling per client, and does it allow the client to review without consuming a seat or a clip? Those two decide whether the tool scales with your roster.
Watch for the tier where per-seat pricing starts, because at that point your bill tracks headcount rather than output. An agency with three editors and six approvers pays for nine people to do three people's work.
The workaround is boring and effective: one creation account, finished files shared out, approvals happening wherever your clients already live.
- Count clips per client per month, then sum, before reading any tier.
- Check whether brand styling is saved per client or per account.
- Reviewers rarely need their own creation access.
- Ask whether unused clips roll over. Most allowances do not.
Priced per take instead, and the captions arrive already burned in
The other way to buy this stage is to stop buying it separately.
A batch is 100 credits and covers transcription, the director pass and the b-roll search. Captions are burned in as part of the export, which bills 20 credits per output minute. A thirty-second ad is 110 credits end to end, captions included rather than added.
Six packs ship: clean-ugc, bold-impact, karaoke-flow, editorial, fast-native and minimal-lux. Colour, weight, size and position are yours, and you can emphasise a word so it pops.
The cut arrives with them. B-roll lands on the phrases you mark, the dead half seconds come out, and the pace is a setting rather than an evening. That is the stage a clip meter starts after.
The trial is 7 days and 300 credits with no card. Basic is $39.99 a month for 2,500 credits, which is about twenty-two ads from twenty-two separate takes, and Premium is $79.99 for 5,000.
The plain facts beside those numbers: one filmed take, three minutes at the ceiling, 9:16 only, and six packs rather than an open styling system.
Submagic and Cutroom, meter against meter
Two rows go to Submagic and they are the caption craft ones. The other six are what happens when the cut has to be built as well.
| Submagic | Cutroom | |
|---|---|---|
| Works on any clip you own | Upload and style anything | One take, three minutes |
| Caption styling depth | The whole product | Six packs, then colour and size |
| Decides the structure | Leaves the cut as it is | Hook, cutaways, pacing |
| Finds b-roll for you | You supply footage | Searched, cut to words |
| Changing the cut itself | Not what it is for | Re-cut free, export bills |
| Music under the voice | Effects, then levels by hand | Generated, 20 credits |
| Unit you can count in advance | Clips, from your analytics | Takes, at 100 credits each |
| Cost of the whole finished ad | One stage of it | 110 credits, upload to file |
Questions people ask
- Is a free caption tier ever enough?
- For under about ten clips a month, often yes. Free tiers usually limit length or styling rather than blocking the job. Run a month on free, count what the limits cost you in irritation, then decide.
- How do I compare a clip meter to a minute meter?
- Convert both to cost per published clip at your real cadence. A clip meter is easy to forecast because you can count clips. A minute meter needs your average length, so multiply before you compare.
- Do unused clips roll over month to month?
- Usually not, and that is worth asking directly. An allowance that resets punishes uneven schedules, which is most schedules once a launch or a holiday lands in the middle of a month.
- Should an agency buy per seat or per volume?
- Per volume while editors outnumber approvers, which is rare, and per seat only when several people genuinely create. Sum clips across every client first, because that total is what breaks the tier a solo creator would have picked.
- Is there anyone who should buy the captioning specialist instead?
- Anyone whose clips are already well structured and only look unfinished, and anyone who wants frame-level control over how each word lands. Cutroom ships six packs and will not open a clip you made elsewhere, so the specialist is the right money.
Count the last thirty days before you read a single tier. Then ask whether you are buying a caption layer or the whole ad, because 110 credits here buys the second one.