Prompt Almanac

Create intermediate Works with: ChatGPT, Gemini

YouTube Thumbnail and Cover Image Briefs

Before after Needs image generation Has example

Write briefs that improve click-through honestly.

Open in ChatGPTOpen in Claude

The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.

Improve my thumbnail concepts and write production briefs. Channel topic and audience: [CHANNEL_AND_AUDIENCE]. Video title and promise: [VIDEO_TITLE_AND_PROMISE]. My current thumbnail idea or existing thumbnail description: [CURRENT_THUMBNAIL]. Brand colors and fonts: [BRAND_ASSETS].

Return:
BEFORE: a description of my current concept, with problems marked in brackets ([too many elements], [text too small at phone size], [repeats the title], [low contrast], [misleading promise], [no focal point], [face unclear]).
AFTER: three alternative concepts. For each give: the single idea in one sentence, the focal subject and its position, the expression or action, background and colors with contrast notes, the text (maximum four words, large, high contrast), and why it makes the right viewer curious without promising something the video does not deliver.
PRODUCTION BRIEF: for the recommended concept, a shot list, layout grid, font size guidance, export size and file type, safe zones for platform overlays (timestamp, progress bar), and a text prompt for an image generator if useful.
TEST: how to check at 150 pixels wide, in grayscale, and next to competitor thumbnails.
A/B PLAN: which element to test first and the success measure.

No clickbait that misleads; accurate imagery only.

Fill in before use

Optional: type your details and the prompt above updates. Nothing is saved or sent.

Replace each bracketed placeholder with your own details, then paste into ChatGPT, Gemini or another assistant.

See an example result (fictional data)
BEFORE: six photos, a title repeated in text, and a tiny face [too many elements] [repeats the title]. At phone size the text is unreadable.
AFTER concept A: one face (surprised, close crop) left; right side a cracked tomato plant; text "WHY IT DIED" (3 words); background deep green; high contrast white text with a dark outline.
Concept B: split before/after of a healthy and a wilted plant; text "2 MISTAKES". Concept C: hand holding a leaf; text "THE FIX".
Production brief for A: shot on a plain background, eyes near the upper third, safe zone 100 px from the lower right (timestamp), export 1280x720 JPG under 2 MB.
Test: view at 150 px wide and in grayscale. A/B: test A versus B on the next upload.

Illustrative only, made-up data. Your results will differ and must be reviewed.

Related prompts