What this prompt does
This prompt directs Claude or ChatGPT to design a product review and rating system for an online store. You set [store_name] and [store_type], then describe the [catalog_size], the [review_volume], the [moderation_policy], the [max_media] per review, and the [framework]. The model designs the review summary on the product page, the review list, the submission form, a Q&A section, a moderation dashboard, review analytics, and the framework implementation.
It works because reviews are the strongest sales tool on a product page, so the prompt treats them as core rather than an add-on. By naming [catalog_size] and [review_volume], the model designs for the reality of popular items with hundreds of reviews and long-tail items with none. The [moderation_policy] variable shapes the moderation queue and fake-review detection, and the framework section specifies the JSON-LD structured data that earns rating stars in search results.
When to use it
- You're adding reviews to a store and want collection, display, and moderation designed together.
- You need the product-page review summary (rating distribution, filterable bars, AI highlights) specified.
- You want a submission form with star selection, structured pros/cons, and photo or video upload.
- You're designing a moderation workflow with automated flags and fake-review detection.
- You need JSON-LD structured data planned so reviews can earn rich-result rating stars.
- You want a Q&A section that prevents duplicate questions and supports seller answers.
Example output
The model returns a component-by-component design document. The review summary section covers a large star display, a clickable rating-distribution chart, and an AI-generated highlights block. The review list specifies per-review elements, photo lightboxes, helpful votes, and filter and sort controls. The moderation dashboard details review queues and fake-review indicators. The closing [framework] section adds an accessible star component, lazy-loaded images, optimistic helpful votes, and JSON-LD for aggregate and individual ratings. Expect a build-ready spec, not finished components.
Pro tips
- Set
[review_volume]honestly; designing for "15-200 reviews on popular items, 0-5 on long-tail" produces better empty states and pagination than assuming every product is busy. - Define
[moderation_policy]clearly so the queue, automated flags, and approval flow match how you'll actually moderate. - Use
[max_media]to set realistic upload limits in the submission form. - Keep the JSON-LD requirement front and center — it's what earns rating stars in search and is easy to overlook.
- For a multi-vendor
[store_type], ask the model to detail seller responses and per-vendor moderation. - Re-prompt the analytics section if you want sentiment trends and complaint-topic categorization expanded.
