Images & Video · from a Podcast episode, Oct 31, 2026 · Added
Storefront Personalization by Shopper Cohort with Amazon's AI Tools and Claude
Redesign a Brand Store for new, considering and repeat shoppers using Amazon's storefront AI tools
“conversion rate rose from 8.7 to 12%”
Per source Reported by an agency about its own client work, not independently verified.
Copy-paste prompt
Here is what Amazon's storefront Personalization Insights told me about my Brand Store [paste], plus a screenshot of my current homepage. For three audiences (new to brand, considering, repeat buyers): 1. What does each need to see in the first five sections? 2. Propose a section-by-section order and a simple wireframe for each version. 3. Point out any suggestion that conflicts with my actual best sellers: [list top 5 ASINs by sales].
The source didn't publish its exact prompt. We wrote this starter so you can try the same workflow. Replace anything in [brackets].
See the original Podcast episode
What they did
Daniela Bolzmann, whose agency Mindful Goods says it has worked with over 850 brands, uses the 'Explore AI Tools' button in Amazon Brand Stores. She prompts Personalization Insights back and forth to extract every bit of storefront performance data, then gives that data, her current design and any Helium 10 data to Claude, which suggests section order and wireframes for each audience. Using Personalize by Cohort, she duplicates the homepage for new-to-brand, considering and repeat shoppers and edits each version. She did this for the pet brand A Better Treat.
How it went
She says the store had been making about $12,000 a month with no ads pointed at it. After the redesign, the founder reported an extra $100 a day in sales in the first two weeks, and for new-to-brand shoppers visits doubled, orders tripled and conversion rate rose from 8.7 to 12%. Two of the three cohorts performed well, and average order value fell, which she plans to test next. She warns to check Claude's suggestions against actual sales data.
We only report what the source says. Results are self-reported unless noted.
Worth knowing
AM/PM Podcast #568 (Helium 10). Agency-reported. Brand Stores have no Manage Your Experiments, so this was before-and-after, not a split test. Her prompts are on her Substack (designproof.substack.com).