Aizii Learn · AI Search

[← Back to Learn](https://aizii.ai/learn)

# Do customer photos and reviews affect AI recommendations

Customer photos and reviews affect AI recommendations when they are real, visible, and specific, on a page a system can read. They do not affect recommendations because they exist. A review that says a shoe fits wide feet, a jacket keeps out rain, or a food suits a sensitive stomach is a fact an engine can quote. A review that says great product, with no detail, is a fact an engine cannot use. A customer photo that shows the product in use, on a real person, in a real room, is evidence a shopper can trust. A customer photo that shows nothing, or shows a different product, is not.

No vendor publishes a formula for how much photos or reviews affect an AI recommendation, and no vendor promises that a review will get a brand cited. Treat photos and reviews as content, not as a lever, and judge them by whether they answer a question a shopper actually asks.

## Where it goes wrong

Brands collect reviews that say nothing, or they mark up reviews that do not exist, and then they wonder why the reviews change nothing.

-   The reviews are generic. A page with fifty reviews that all say great product, fast shipping, and love it gives an engine no fact to quote, and gives a shopper no reason to choose this product over another.
-   The reviews are not visible. A review that is only in the markup, and not on the page, is not a review a shopper can read. Google says review content you mark up must be readily available to users from the marked up page, and that it must be immediately obvious to users that the page has review content.
-   The reviews are fake, or the incentive is hidden. Google says not to include fake or undisclosed incentivized reviews in markup. A fake review is a lie, and a hidden incentive is a lie by omission, and both teach a system not to trust the next review on the page.
-   The reviews are aggregated from other websites. Google says not to aggregate reviews or ratings from other websites. A rating that mixes reviews from three sites is not a rating for this product, on this page, from these customers.
-   The customer photos are not checked. A photo that shows a different color, a different size, or a damaged product, left on the page with no reply, will be read as the product, and a shopper who sees it will trust the photo over the description.

## The steps

**Step 1. Ask for a specific review, not a rating.** The marketer asks customers for the fact that mattered: did it fit, did it keep out rain, did it suit a sensitive stomach, did it last? A review that answers one of those questions is worth more than ten reviews that say great product. Put the question in the review request, in plain words.

**Step 2. Show the reviews on the page, with the rating.** The marketer checks that the reviews are visible on the product page, with the review text, the author, and the date, and that the aggregate rating is visible where the markup says it is. Google says that if you use an aggregate rating, users should be able to see that aggregate rating on the page. A rating that is only in the code is not a rating a shopper can trust.

**Step 3. Mark up only what is visible, and only for this product.** The developer marks up the reviews that are on the page, nested in the Product for that product, with an aggregate rating that matches the reviews shown. Google says to provide review information about a specific item, not about a category or a list of items, and to refer clearly to a specific product by nesting the review within the markup of another type, such as Product. If the page shows multiple individual reviews, also include an aggregate rating of those reviews.

**Step 4. Do not mark up fake, self serving, or undisclosed incentivized reviews.** The marketer checks every review before it is marked up: is it from a real customer, for this product, with no hidden payment? Google says to only accept ratings that are accompanied by a review comment and an author name, and not to include fake or undisclosed incentivized reviews. A review from the brand itself, or from a person paid to write it without disclosure, must not be marked up, and should not be shown as a customer review.

**Step 5. Reply to the photos and reviews that show a problem.** The marketer reads customer photos and reviews every week, and replies where a photo shows a different color, a damaged product, or a fit problem. A reply that says what the brand will do, in plain words, is a fact an engine can quote, and a shopper can trust. A problem left with no reply is a problem the brand has accepted.

**Step 6. Use the reviews to fix the product page.** The marketer reads the reviews for the question they answer, and adds that answer to the product page. If five reviews say the shoe fits wide feet, the product page should say it fits wide feet, in plain words, with the size guidance to match. The review is the evidence, and the page is the answer.

## The machine-readable piece

This is the shape of a Product block with a review summary, using the schema.org vocabulary. The review part is only used when the page visibly shows the reviews and the aggregate rating. Replace the placeholders with your own product facts, and only use values that match the visible page.

```
<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Product",
  "name": "Example Pour Over Coffee Maker",
  "image": "https://[yourdomain.com]/images/pour-over-coffee-maker.jpg",
  "description": "A pour over coffee maker for a small kitchen, with a glass carafe and a reusable filter.",
  "brand": {
    "@type": "Brand",
    "name": "Example Coffee Brand"
  },
  "offers": {
    "@type": "Offer",
    "price": "49.00",
    "priceCurrency": "USD",
    "availability": "https://schema.org/InStock"
  },
  "aggregateRating": {
    "@type": "AggregateRating",
    "ratingValue": "4.6",
    "reviewCount": "128"
  },
  "review": {
    "@type": "Review",
    "author": {
      "@type": "Person",
      "name": "Example Customer"
    },
    "reviewRating": {
      "@type": "Rating",
      "ratingValue": "5"
    },
    "reviewBody": "The maker fits a small kitchen, and the coffee stays hot in the carafe."
  }
}
</script>
```

Check the block after you publish it. Open the page and confirm the name, the price, the availability, the aggregate rating, and the review text match what the shopper sees. Then run the page through the Rich Results Test and fix any error it reports. A review that is visible, specific, and marked up correctly is a review an engine can quote. A review that is only in the code is not.

## What it looks like when it works

Say you run a home brewing brand, selling pour over makers, grinders, and kettles, with a catalog of about forty products. The marketer asks customers for the fact that mattered: does the maker fit a small kitchen, does the coffee stay hot, does the filter last? The reviews come back with specifics, and the marketer shows them on the product page, with the aggregate rating visible, and marks up only what is shown.

One review says the maker fits a small kitchen and the coffee stays hot in the carafe. The marketer adds that fact to the product page, in plain words, with the size of the carafe. A shopper asks an assistant which pour over maker suits a small kitchen, and the assistant can find the page, read the review and the page together, and cite the brand with the fact as written. The review did not make the sale on its own. It answered the question, in a customer's words, on a page a system could read.

### The fundamentals

These mechanics are shared across every engine and covered in full in their own guides:

-   •The llms.txt format and a complete sample, see our [guide to writing a good llms.txt file](https://aizii.ai/learn/ai-search/how-to-write-a-good-llms-txt-file)
-   •The markup AI systems read to verify facts, see our [product schema guide](https://aizii.ai/learn/ai-search/how-to-add-product-schema-markup-for-ai-search)
-   •The complete crawler rule reference, see our [robots.txt guide for AI search](https://aizii.ai/learn/ai-search/how-to-write-robots-txt-for-ai-search)

## What this guide does not cover

This guide covers whether customer photos and reviews affect AI recommendations, and how to show them so a system can read them. It does not cover how to collect reviews at scale, the full technical audit that checks every crawler and every page, or the transaction layer that lets an agent complete a purchase. It does not cover how to run a review platform, or how to handle a legal dispute about a review. Those are separate jobs, and this guide is the map, not the whole trip.

## Check your own store now

Run the free scan. One click, no signup, and it shows whether AI crawlers can read your store today: aizii.ai/aeo-scan

[Run the free scan](https://aizii.ai/aeo-audit)

## Sources

-   Google Search Central, Optimizing your website for generative AI features on Google Search (specific): [https://developers.google.com/search/docs/fundamentals/ai-optimization-guide](https://developers.google.com/search/docs/fundamentals/ai-optimization-guide)
-   Google Search Central, Review snippet structured data (specific): [https://developers.google.com/search/docs/appearance/structured-data/review-snippet](https://developers.google.com/search/docs/appearance/structured-data/review-snippet)
-   Google Search Central, Product structured data (specific): [https://developers.google.com/search/docs/appearance/structured-data/product](https://developers.google.com/search/docs/appearance/structured-data/product)
-   Schema.org, Product type (specific): [https://schema.org/Product](https://schema.org/Product)
-   Schema.org, Review type (specific): [https://schema.org/Review](https://schema.org/Review)
-   Schema.org, AggregateRating type (specific): [https://schema.org/AggregateRating](https://schema.org/AggregateRating)

Note: links marked specific point to the page that supports the claim in this guide. Links marked general docs point to the vendor's general docs section for that topic, not to a page that proves every step. Check the vendor docs for the current path before relying on a menu name or click path in this guide.
