Nano Banana for marketing: product images, ad creative and prompts
Kahu team·Updated 1 Oct 2026·6 min read
Nano Banana is the popular name for Google’s Gemini image generation and editing model. For marketing, it is most useful for editing photos you already have: changing backgrounds, creating ad variations and placing products in scenes. Start from a real product photo, describe the change clearly, and check every image for accuracy before it runs.
What it is good at
- Replacing or cleaning up backgrounds on product photos.
- Making several ad variations from one image: seasons, settings, colours.
- Keeping a product or person consistent across a set of images.
- Combining two images, such as a product and a room.
Prompts that work
- 1Start with your photo“Using this product photo, place the bottle on a timber bench in soft morning light. Keep the label exactly as it is.”
- 2Ask for variations“Make three versions for a square Instagram ad: summer, winter, and plain studio background.”
- 3Edit specifics“Remove the cable on the left. Do not change anything else.”
- 4Fit the format“Reframe as 9:16 for a Story, leaving space at the top for text.”
What to check before publishing
- Labels, logos and text are correct and readable.
- The product looks like what the customer will receive.
- No invented features, sizes or packaging.
- You have rights to any photos you uploaded.
Images made with Google’s models carry an invisible SynthID watermark. Some platforms may label AI-generated images, so avoid using them to show results you cannot deliver.
Frequently asked questions
Where can I use Nano Banana?
In the Gemini app and through Google’s AI tools and APIs. Availability and limits vary by plan.
Can I use the images in ads?
Generally yes, within Google’s terms and the ad platform’s policies. Keep the product accurate.
Is it better than real photography?
It is faster for variations. A good real photo is still the best starting point.

