Start with where the image will appear
An article cover may need space for a headline; a product image needs a clear focal point; a social post must work on a phone. Decide the aspect ratio, placement and areas that must survive cropping before generating. These choices give later revisions a concrete direction.
Describe what someone could see
For example: ‘A clear glass with one green leaf on a pale wooden table. Place the glass on the right and leave the left side empty. Soft window light from the side, with a simple background.’ Establish objects, relationships and space before adding style words.
Change one variable per round
If the composition feels crowded, adjust the number of background objects or the empty space. If the mood is wrong, change the light or color. Save the previous prompt and image with a short reason for the change. Changing everything at once makes useful results harder to reproduce.
Give references a specific job
Explain whether a reference is for composition, color or material. Use images you have permission to use and check the tool's input requirements. For recurring characters or products, keep a consistent reference set instead of describing them from memory each time.
Inspect the image in its final context
Check hands, lettering, edges, reflections and how objects connect. Place the image in the actual poster or page to review contrast and cropping. Add complex text in a layout tool when needed. The final image should support the information you want to communicate.
Methods, examples and selection advice in this article were developed by FindGoodAI. Product capabilities and access links refer to the official resources listed below.
Sources & references
Better image prompts begin with a visible scene — official reference ↗
Better image prompts begin with a visible scene — official reference ↗
Better image prompts begin with a visible scene — official reference ↗
Copy the link to save or share this article.
