Pick 1× or 2× before you caption

Pick 1× or 2× before you caption. The 21 basic presets bill at the standard rate. The 9 dynamic presets — glass, whisper, glide, fusion, terminal, handwritten, backdrop, and the rest of that list — bill at 2×. Transcription is the same work. The multiplier is the look. Choosing Glass because it looked pretty on a thumbnail, then captioning thirty clips, is how a month of captions becomes two months of caption renders.

What captions cost across a month of clips is the arithmetic: credits = caption_render × preset_multiplier × clips. This page is the weekly radio: pick the tier before the first full render, not after you have already 2×’d a batch.

This is not a tour of thirty looks. Caption styles is nine families of five — Classic, Paper, Ink, and the rest — a different catalogue from the VEED basic/dynamic split the studio actually burns. Do not confuse a family name with a tier. The bill cares about 1× versus 2×.

The multiplier is the only aesthetic price on the hub

Captioning is the job where a purely visual choice is a pricing choice. Speech-to-text and the timing pass do not change when you pick Glide instead of Plain. What changes is the charge. There is no third tier and no partial multiplier. Basic is 1. Dynamic is 2.

On one clip that is easy to ignore. At 30 clips a month it is the second-largest recurring decision in a short-form workflow after the video model itself, and unlike the model it changes nothing about the footage. Thirty clips on a basic preset is thirty base renders. Thirty clips on a dynamic preset is sixty. That is the whole leftover.

Agencies inherit this the noisy way: the agent default lands on a dynamic look, nobody names a tier in the brief, and a retainer quietly 2×s. The fix is a house default written down. One channel, one tier, unless a single hero clip is worth the extra render.

Dynamic is not “better captions.” It is animation: glass, fusion, terminal, handwritten motion, a look that can fight a face or a UI. Basic is legible type. Most talking-head and screen-recording work wants basic. A music-video hook or a channel that already owns a kinetic look can justify dynamic. Justify it on this clip, not as a vibe.

If you do not know which look you want, you are not ready to pick 2×. Preview first — that is a different radio, on a few seconds of the real file. Do not discover the look by captioning three whole videos and keeping one. That is three full-price jobs to make an aesthetic decision.

What you are actually picking this week

Open the slate. Count the clips that will be captioned, not the clips you might generate. For each clip, or for the whole channel if you have a house style, answer:

• 1× / basic. Speech is on the file. You need words on screen for mute. You do not need the type to perform. 21 looks. This is the default for courses, podcast clips, screen recordings, and most UGC.
• 2× / dynamic. The caption is part of the brand motion. You have previewed it on this background. You accept double the caption line for the same transcript.

Do not mix tiers inside one series without a reason. A week that is Plain on Monday and Glass on Thursday does not look like a series. It looks like two tools. Pick one multiplier for the calendar, then exception a single hero if you must.

Position is not a tier. Top, middle, or bottom is a layout field. Safe area is a layout field. Colour is a layout field. None of those flip 1× to 2×. Animation does.

Language is not a tier. Add captions transcribes in 165 language codes. It does not translate. You still pick 1× or 2× on top of that. A Spanish basic burn is 1×. A Spanish dynamic burn is 2×. The multiplier does not care which language the transcript is in.

Burn-in versus SRT is also not this radio. Sidecars do not have a Glass look. If this week’s export is an SRT for a course, you are not on a caption-tier decision at all. You are on words and times. Come back here when the destination is a burned MP4.

How the week gets expensive

Three habits 2× a budget without anyone meaning to:

1. Picking from a thumbnail. Dynamic presets look like music videos in a picker. On a talking head they can strobe the mouth. Preview on the real clip, then pick the tier. If the preview already annoys you, you do not owe the full 2× render.
2. Captioning the VOD, then cutting clips. A one-hour file on dynamic is a long 2× job, and the clips you actually post still need a recrop and often a recaption. Trim first. Caption the keepers. Apply the tier to the keepers.
3. Leaving last week’s dynamic preset in the session. The studio will happily 2× whatever is selected. Reset to the house 1× at the start of the batch. Dynamic should be a choice you can point at, not a leftover.

The AI caption generator is the transcribe-and-burn path. It will not warn you that Glass is a lifestyle. You have to bring the radio. Write “basic, 1×, [preset name]” in the brief if someone else is running the agent. Unspecified style is how Glass happens.

Transcription errors are independent of tier. Product names, numbers, and speaker IDs are wrong at 1× and still wrong at 2×. Read the transcript before you burn. Paying double to animate a misspelled SKU is a complete sentence.

If the clip has no speech, you do not have a caption-tier decision. Auto-captions transcribe audio. A mute plate that needs a headline is overlay, not a 1× or 2× burn. Do not send it here to be safe.

The Monday radio

1. Count captionable clips for the week (speech on the file, destination is a feed or a burned master).
2. Name the house tier: 1× unless a written exception exists.
3. Name the preset inside that tier so the series matches.
4. Preview the exception clips if any of them are 2×.
5. Caption the batch. Do not change tier mid-batch because one hook felt flat. Flat hooks are copy and timing, not Glass.

Thirty clips at 1× is thirty caption renders. Thirty at 2× is sixty. The cost page will recompute the live per-clip figure; this page only needs you to pick the multiplier while the number is still a choice.

If you need the studio to burn the week at the tier you named, sign up and select 1× or 2× before the first full caption. The look is allowed to be boring. The invoice should not surprise you.

FAQ

Can I caption a few clips at 2× and the rest at 1×?

Yes, if the series can tolerate two looks. Most cannot. Exception a single hero, or keep the whole week on one multiplier. Random 2× is how the channel looks unbriefed.

Does previewing a dynamic look already bill 2× on the whole file?

No. A short preview is billed on the sample, not on the full duration. That is why you preview before you commit. The 2× lands when you caption the whole clip in a dynamic preset.

Are the 21 basic looks all equally cheap?

They share the 1× multiplier. They are not the same typeface or weight. Pick for contrast on this background. The bill does not change inside the tier. The read does.