You finally built the course. Now every module needs a cover, and you’re either reusing the same tired stock photo or losing an afternoon in Canva trying to make eight thumbnails look like they belong together. They never quite do.
This prompt fixes that. It turns one module title into a clean, designed cover, and because it locks your brand colors, mood, and style into every request, your online course covers for coaches stay consistent from Module 1 to Module 8. By the end of this page you’ll also understand why it works, so your next prompt is sharper.
When to use this
- You’re launching a course and need a cover for every module that looks like a set.
- You’re tired of generic stock photos that don’t match your brand.
- You want thumbnails that still read clearly when shrunk on a course platform.
- You’re refreshing an older course and want all the covers to finally match.
- You don’t have a designer and don’t want to learn one.
The prompt
Copy this whole block into ChatGPT, Claude, or Gemini:
You are an expert brand designer who creates cover art for online courses sold by coaches. Your job is to design ONE module cover image that looks professional, stays on-brand across an entire course, and reads clearly even as a small thumbnail.
Before designing, ask me up to 3 clarifying questions if anything below is unclear. Otherwise, proceed.
CONTEXT
- Course name: {{COURSE_NAME}}
- This module: {{MODULE_TITLE}}
- My coaching niche: {{NICHE}}
- My brand colors: {{BRAND_COLORS}}
- The mood I want: {{MOOD}}
- Where the image will be used: {{ASPECT_RATIO}}
TASK
Design ONE module cover image. Return it in two parts:
1. A ready-to-paste image-generation prompt (for DALL-E, Midjourney, or Gemini) that describes the scene, composition, lighting, color palette, and style in concrete visual terms.
2. A short layout note telling me where the module title text should sit and what NOT to put in the image.
CONSTRAINTS
- Use only my brand colors plus neutral tones. No rainbow palettes.
- Leave clear negative space for a text overlay; do not bake the module title into the image itself.
- Keep the composition simple enough to read at thumbnail size.
- No clip-art, no generic stock-photo handshakes, no cheesy lightbulbs or rocket ships.
- Match the style across modules so all covers look like one family.
- State the aspect ratio explicitly in the prompt.
After the cover, give me 2 alternative visual concepts I could try for the same module.
How to customize it
Replace the six {{VARIABLES}} before you send it:
| Variable | What to put | Example |
|---|---|---|
{{COURSE_NAME}} |
The full name of your course | The Confident Speaker Method |
{{MODULE_TITLE}} |
The specific module this cover is for | Module 3: Beating Stage Fright |
{{NICHE}} |
Your coaching niche | public speaking coaching for introverts |
{{BRAND_COLORS}} |
Your two or three brand colors | deep navy, warm gold, off-white |
{{MOOD}} |
The feeling the cover should give | calm, grounded, quietly confident |
{{ASPECT_RATIO}} |
Where the image will live | 16:9 for a course platform thumbnail |
Keep every variable identical across modules except {{MODULE_TITLE}}. That single change is what makes the covers a matching set instead of eight unrelated images.
See it in action (full example)
Here is the exact prompt, filled in for a public speaking coach. This is the whole input, nothing hidden:
You are an expert brand designer who creates cover art for online courses sold by coaches. Your job is to design ONE module cover image that looks professional, stays on-brand across an entire course, and reads clearly even as a small thumbnail.
Before designing, ask me up to 3 clarifying questions if anything below is unclear. Otherwise, proceed.
CONTEXT
- Course name: The Confident Speaker Method
- This module: Module 3: Beating Stage Fright
- My coaching niche: public speaking coaching for introverts
- My brand colors: deep navy, warm gold, off-white
- The mood I want: calm, grounded, quietly confident
- Where the image will be used: 16:9 for a course platform thumbnail
TASK
Design ONE module cover image... [rest of the prompt unchanged]
And here is the complete output the model returns:
Got it. Your brief is clear enough to design from, so I’ll proceed.
1. Ready-to-paste image-generation prompt
A minimalist 16:9 course cover illustration. An empty stage seen from the wings, soft single spotlight pooling on weathered wooden floorboards, an unoccupied microphone stand standing calm and centered slightly to the right. Background fades into deep navy shadow with gentle off-white haze in the light beam. Warm gold accents on the spotlight glow and the microphone rim. Flat editorial illustration style, clean vector shapes, subtle film grain, no people, no text. Generous empty negative space across the upper-left third. Calm, grounded, quietly confident mood. Color palette strictly limited to deep navy, warm gold, and off-white. 16:9 aspect ratio.2. Layout note
Place the module title text in the upper-left negative space, left-aligned, in off-white. Use “Module 3” small on top and “Beating Stage Fright” larger beneath it. Do NOT bake any text, logos, or numbers into the generated image itself, add those as a separate overlay so they stay crisp and editable. Avoid faces, audiences, and literal anxiety imagery (sweating, shaking hands); the empty, waiting stage carries the feeling on its own.
2 alternative visual concepts for this module: 1. A single deep-breath motif: a softly lit glass of water and a folded set of speaker notes on a navy table, gold morning light from the side. Quiet pre-talk calm. 2. An overhead view of two shoes on a navy floor, gold tape mark forming an X where a speaker would stand. Implies stepping up, no person required.
That is publish-ready. Paste the prompt block into your image tool, generate, then drop your module title into the upper-left space. Repeat for each module, changing only the module title, and you have a matching set.
Why this works
Three LLM principles are doing the heavy lifting here. Learn them and every prompt you write gets better:
- Role priming. The first line (“You are an expert brand designer who creates cover art for online courses”) tells the model which slice of its knowledge to draw from. “Make me a course image” pulls from the bland average of the internet; assigning a designer role pulls from the language of composition, palette, and negative space. Always assign a role.
- Specificity in, specificity out. An image model can only render what it’s told. “Professional and modern” produces a different result every time and matches nothing. Naming the scene, the lighting, the exact palette, and the aspect ratio is what makes Module 3 and Module 4 look like siblings. Notice the prompt asks the LLM to write another concrete prompt for the image tool, because vague image prompts are the number one reason covers come out generic.
- Constraints are quality control. The “no” list (no clip-art, no handshakes, no baked-in text, no rainbow palettes) isn’t decoration. Each line removes a common failure mode that ruins AI course art. Telling the model what NOT to do is as powerful as telling it what to do. And the “ask me up to 3 clarifying questions first” line lets it fill gaps by asking instead of guessing, which is the single biggest fix for off-brand results.
Do this now
- Copy the prompt above into ChatGPT, Claude, or Gemini.
- Replace the six variables with your real course name, module, niche, colors, mood, and aspect ratio.
- Send it. If it asks clarifying questions, answer them honestly.
- Paste the image prompt it returns into your image tool, generate, and add your module title as a text overlay. Done beats perfect.
Pro tips
- Lock everything but the module title. Reuse the exact same colors, mood, and style across every module. Changing only
{{MODULE_TITLE}}is what keeps the set consistent. - Always overlay your text separately. Image models still misspell baked-in words. Generating clean negative space and adding the title yourself keeps it sharp and editable.
- Save the winning image prompt. Once a cover looks right, keep that generated prompt as your template and swap the module-specific scene for the next one.
- Generate two ratios at once. Ask for both a 16:9 thumbnail and a 1:1 version so the same cover works on your course platform and on social.
0 comments
No comments yet.