Listings & SEO · Podcast episode · source Mar 24, 2026
Listings and Main Images for Dozens of Launches with ChatGPT and Nano Banana
Write listings and refresh main images fast when you launch many products
“helped her get through 33 launches”
Per source Self-reported by the source, not independently verified.
Copy-paste prompt
Create an Amazon listing with a title and five bullet points for [product] that integrates all of these keywords: [20 keywords]. Keep it readable: if a keyword can't fit naturally, list it separately for backend search terms instead of forcing it in.
The source didn't publish its exact prompt. We wrote this starter so you can try the same workflow. Replace anything in [brackets].
What they did
A solo founder running over a hundred women's accessories products pulls the best keywords for each product from Cerebro or a search query tool and asks ChatGPT for a listing with five bullets that works in all 20 of them. For images she shoots the product at home, uploads the photos to Nano Banana and asks for changes such as a different pose, better lighting and colour, and tests the results in polls of female Amazon shoppers before switching.
How it went
She says ChatGPT's copy is 'quite decent' and helped her get through 33 launches, though the more keywords you force in, the worse it reads. She also says some products 'were going to die' and she saved them by changing the main image with Nano Banana: click-through rose, rankings followed, and a new image takes a minute instead of a photo shoot.
We only report what the source says. Results are self-reported unless noted.
Worth knowing
Said on Helium 10's podcast, with no CTR or conversion figures. Two habits worth copying: cap the forced keyword list, and start AI images from a real photo of the real product.