Snapchat monetization advice: closed captions, 30 s, and Sume

You can burn closed-style captions into a clip with Sume video captions, and Snapchat's monetization page recommends captions. The page says Spotlight videos must be at least 30 seconds long to be eligible to earn revenue, and that closed captions provide context that help keep viewers engaged. Sume's captions job costs $0.20 per accepted job at the current estimate for clips of up to 60 seconds.

The page gives guidance and not a technical spec, so treat the advice as a checklist and not as a promise. The sections below match each line of it to a Sume setting, and they say where no match exists.

What the page says

The page is a program overview, and its video guidance is short. It also says to use the Snapchat Camera and Creative Tools, so a file made elsewhere may not fit the program's intent, and you should read the program terms before you rely on any outside workflow.

How the Sume captions job works

Burned-in captions are open captions, since they are part of the picture and cannot be turned off. Sume's caption job takes a public HTTPS video_url, a style, an optional language, and optional script_text. Without script_text, it transcribes the speech. If you pass script_text, Sume keeps the speech to text timings as the source for time and aligns your words to them.

A silent clip has no speech, so the job would fail with caption_no_speech. In that case you send cues, each with text, start and end in seconds, and Sume burns your text without speech recognition.

Style is a creative choice. Sume's docs say that if you send no style, the text of the captions sets it, with slam for Latin text. The tiktok-green and punch styles do not accept design overrides. If you want to change colors or placement, pick slam or one of the Hangul identities that support them, and read the design override table in the captions docs.

If you need to fix a wrong word without paying for a second transcription, you can restyle with source_caption_id, which reuses the stored word timings. The docs say a restyle is still a render, so the price stays the same.

A caption request

The payload below captions a clip with the punch style and a script. The style names in the docs include slam, punch, tiktok-green and black-outline. Note that punch does not support design overrides, and that Korean text on a Latin style is rejected. This sample is Latin text only.

A practical order of operations is: trim to the final range, probe the duration against the 30-second line, then caption the final cut. Captioning before the trim would burn text onto frames you later discard, and the cue times would no longer line up.

import json

payload = {
    "video_url": "https://media.sume.com/artifacts/artf_demo/clip.mp4",
    "style": "punch",
    "language": "en",
    "script_text": "Three tips for a faster morning routine.",
}
headers = {"Idempotency-Key": "snap-caption-001"}
print(json.dumps({"headers": headers, "body": payload}, indent=2))

Length and price

A 30-second floor is not only a trim question. If the clip is 40 seconds and the useful part is 25, cutting to 25 loses revenue eligibility by the page's rule. Probe the final duration, and keep it at or above 30 seconds, as a separate check from the 5-second Post button check.

The caption price covers clips of up to 60 seconds under the fixed estimate. For a longer clip, the docs tell you to check the live price in the catalog and the OpenAPI document, so do not assume $0.20 beyond 60 seconds.

What not to expect

Captions help viewers who watch without sound, but the page offers no number for how much they help, so this post makes no claim about reach or earnings. It only matches the page's advice to a tool that exists.