SchemaForAI
Ad

Free Review Schema Generator for AI Search

Generate JSON-LD Review schema with ratings, reviewer, and review body. Optimized for ChatGPT, Google AI Overviews, and star-rating rich results. No sign-up.

Review Details

Review Schema

The name of the thing being reviewed.

Use the item's canonical name — the string users would search for.

What category of thing you're reviewing.

Pick the most specific type — Product for goods, LocalBusiness for venues, Book for books, Movie for films.

URL of the item being reviewed (optional, but strongly recommended).

Linking to the reviewed item lets AI correlate your review with the entity. Skip this and the review floats free.

Name of the person writing the review.

Named reviewers are trusted more than anonymous ones by AI engines. Use a real person, not 'Customer #1234'.

The reviewer's rating (1–5 by default, but see bestRating).

Maximum on your rating scale. 5 for 5-star, 10 for 10-point, 100 for percentage.

Minimum 50 characters. AI engines quote review bodies directly.

0 / 5000

Mention specific attributes (build quality, sound, service, etc.). Generic reviews don't get cited.

When the review was published.

Organization publishing the review (optional).

JSON-LD
{
  "@context": "https://schema.org",
  "@type": "Review",
  "itemReviewed": {
    "@type": "Product",
    "name": "Item name"
  },
  "author": { "@type": "Person", "name": "Reviewer" },
  "reviewRating": {
    "@type": "Rating",
    "ratingValue": 4.5,
    "bestRating": 5
  },
  "reviewBody": "Review text…"
}

AI Readiness Score

Get a 0–100 score showing how likely AI search engines are to parse and cite content from your page.

How to use this code:

  1. Copy the JSON-LD code above
  2. Paste inside <script type="application/ld+json">
  3. Add the script tag to your review page <head>
  4. Validate with Google Rich Results Test ↗
AI Simulator

See how ChatGPT answers — with and without your schema

Side-by-side comparison of what an AI assistant would return for a typical review query, with vs. without the structured data you generated above.

3 of 3 remaining today

Sample user query

Is this item worth buying? What do reviewers say?

Without Schema
Generic answer

Click Simulate to see the AI response without schema.

With Schema
Detailed citation

Click Simulate to see the AI response with schema.

Install Guide

How to install this schema on your site

Pick your platform. Every step includes a copy-paste-ready code block where it applies.

  1. 1

    Install and activate a schema plugin: "Rank Math SEO", "Yoast SEO Premium", or "Schema Pro". For straight JSON-LD injection without a full SEO plugin, "WPCode" (formerly Insert Headers and Footers) is the simplest option.

  2. 2

    In Rank Math: go to the post edit screen → Rank Math sidebar → Schema tab → Schema Builder → paste your JSON-LD. In Yoast Premium: the Schema tab auto-generates standard types; use the Custom Schema plugin add-on to inject custom types.

  3. 3

    For WPCode: Code Snippets → Add Snippet → Custom Code (HTML/PHP) → paste the full <script type='application/ld+json'> tag. Set Location to 'Site Wide Header' or 'Specific Post/Page' depending on scope.

    html
    <script type="application/ld+json">
    {
      "@context": "https://schema.org",
      "@type": "FAQPage",
      "mainEntity": [ /* your Q&A pairs */ ]
    }
    </script>
  4. 4

    Publish the post. Open the live URL, view source, confirm the <script type="application/ld+json"> tag appears in the <head>.

  5. 5

    Validate at search.google.com/test/rich-results.

What is Review Schema?

Review is the Schema.orgtype for a specific person's or publication's evaluation of something — a product, a business, a book, a movie, an event, or any other reviewable entity. It connects three entities explicitly: what is being reviewed (itemReviewed), who is reviewing it (author), and the numeric rating plus written reasoning (reviewRating and reviewBody). When AI search engines surface reviews in recommendation queries, they're reading this exact structure.

A companion type, AggregateRating, summarizes many Reviews into a single average rating and review count. Most product and business pages use both: AggregateRating for the star-display summary at the top, plus a selection of individual Reviews to provide qualitative context. Here is a canonical editorial Review:

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Review",
  "itemReviewed": {
    "@type": "Product",
    "name": "Acme Wireless Headphones"
  },
  "author": {
    "@type": "Person",
    "name": "Jane Doe"
  },
  "reviewRating": {
    "@type": "Rating",
    "ratingValue": 4.5,
    "bestRating": 5
  },
  "reviewBody": "The noise cancellation is excellent...",
  "datePublished": "2026-01-15"
}
</script>

Notice the nested itemReviewed uses its own @type — in this case Product, but it could just as easily be LocalBusiness, Book, Movie, SoftwareApplication, or any other reviewable schema type. This nesting is what lets AI engines correctly attribute the review to the specific entity.

Ad

Why Review Schema Matters for AI Search

Review schema has become even more valuable since the shift to AI-powered search. When a user asks Perplexity "what are the best 4K TVs under $1000," the AI synthesizes recommendations from review content across the web — and the reviews that make it into the synthesis are overwhelmingly ones with clean Review schema. A thoughtfully reviewed product with no schema is invisible to AI recommendation engines; a mediocre review with perfect schema will get cited.

AI engines reward specific properties of Review schema:

  • Named reviewer authority: A Review attributed to a named Person with a sameAs-linked identity (LinkedIn, bio page, other review authorship) dramatically outperforms anonymous reviews. Editorial reviews from recognized publications carry the most weight.
  • itemReviewed disambiguation: Linking the review to the correct entity via itemReviewed.url or sameAs is what lets AI engines confidently attribute your review to the right product or business. Ambiguous items get filtered out.
  • Rating precision: A rating of 4.7 is more informative than a rating of 5. AI engines treat rounded ratings as lower-quality signals than precise ones, all else being equal.
  • Review body substance: AI engines quote specific phrases from reviewBody directly in their recommendation responses. Reviews with concrete observations ("the bass is tight, the mids are recessed") get quoted; vague ones don't.
  • Recency: Recent reviews weight more heavily in time-sensitive categories. A 2025 review of a 2026 product model gets the latest context; a 2018 review doesn't.
  • Publication credibility: Reviews with a named Publisher Organization that AI engines recognize as credible get a citation multiplier. This is how Wirecutter consistently beats random review aggregators.

The punchline: for review publications, Review schema is not optional. It is the entity declaration that makes your review discoverable by AI systems that increasingly bypass traditional search.

Review Schema Best Practices

Always link itemReviewed by URL

A Review of "Acme Headphones" is weaker than a Review of the specific Acme Headphones product page. URL linking eliminates ambiguity.

Attribute to a real Person

Anonymous or "Staff" reviews lose weight. Named reviewers with bios and external profiles earn citation.

Use precise ratings

4.7 beats 5. AI engines treat precision as a quality signal.

Write substantive reviewBody

Name specific attributes — sound, build, fit, service, taste. AI engines quote these phrases directly.

Include date stamps

datePublished on every review. Freshness matters, especially for fast-evolving categories.

Don't self-review

Google's guidelines prohibit it, and AI engines detect and discount self-reviews.

Combine Review with AggregateRating

Both together — summary stats plus individual voices — is the strongest pattern.

Scope to actual content

The review schema must match the review visible on the page. Hidden schema is manipulation.

How to Install Review Schema

Editorial review site

  1. 1Structure your CMS so each review article stores itemReviewed data explicitly — don't rely on free-text identification.
  2. 2Generate Review schema at build time from the structured review record.
  3. 3Always emit datePublished and dateModified; update the latter when you refresh the review.
  4. 4Attribute to a named Person entity, and include a sameAs Person profile on that author's bio page.
  5. 5Validate the review page URL with Google Rich Results Test.

E-commerce product page (with user reviews)

  1. 1Use AggregateRating for the summary stats visible at the top of the product page.
  2. 2Emit 5–10 individual Review objects for the most helpful reviews.
  3. 3Pick helpful reviews deterministically (by upvote count) so schema is cache-friendly.
  4. 4Never expose reviews in schema that aren't also visible on the page.
  5. 5Validate a sample of product pages to ensure both AggregateRating and Review schema render correctly.

WordPress

  1. 1For review blogs, plugins like WP Review Pro, Rank Math, or Schema Pro auto-generate Review schema from custom fields.
  2. 2For WooCommerce stores with user reviews, WooCommerce itself handles AggregateRating; augment with review-specific plugins if you need individual Reviews in schema.
  3. 3Validate on the live URL — WordPress schema plugins sometimes emit Review schema without itemReviewed, which fails validation.
  4. 4Ensure review dates come from the post or comment timestamp, not the page rendering time.

Frequently Asked Questions

What's the difference between Review and AggregateRating?+
Review is a single person's opinion of something — one reviewer, one rating, one body of text. AggregateRating is the summary statistic across many reviews — an average rating and a count. They answer different questions. A page with 200 user reviews should use one AggregateRating (average 4.5, count 200) and optionally include a selection of individual Review objects to preserve voice. Google rich results for products and businesses generally use AggregateRating for the star-count display, but AI engines read individual Reviews to quote specific feedback. Both together are stronger than either alone.
Can I add Review schema to content I publish as review content?+
Yes, and this is the primary use case. If you run a publication that reviews products, businesses, or media (think Wirecutter, Tom's Guide, restaurant critics), every review article should have Review schema where itemReviewed points to the thing you're reviewing and the reviewer is your publication or individual critic. AI engines rely on Review schema to identify editorial reviews and distinguish them from user-generated reviews. A well-marked-up editorial Review schema with a named Person author and a recognized Publisher Organization is highly cited by AI in product recommendation queries.
Should I include all my reviews on the product page?+
You don't need every review, and stuffing schema with every single review is counterproductive. The cleanest pattern is: use AggregateRating for the aggregate star count and volume, then embed the 5–10 most helpful individual Reviews as schema. The helpful-review selection should be deterministic — most upvotes, most recent, or highest-rated depending on your policy. Rotating reviews every page load breaks AI engine caching. If you have thousands of reviews, consider a separate /reviews URL with the full list of Review schema, linked from the main product page via a related or itemReviewed relationship.
What's the valid range for ratingValue?+
ratingValue can be any numeric value, but must be accompanied by a matching bestRating (and optionally worstRating) to be meaningful. The conventional 5-star scale uses bestRating: 5. A 10-point scale uses bestRating: 10. A percentage-based system uses bestRating: 100. Keep the scale consistent across reviews for the same entity — mixing 5-star and 10-point scales in your own catalog creates calculation errors when aggregating. When in doubt, use the 5-star scale: it's the most broadly understood by both users and AI engines, and it's the default assumption when no bestRating is provided.
Is it okay to include reviews for businesses on a review aggregator site?+
Yes, and it's valuable. If you run a review site (Yelp-for-an-industry, a niche review blog, a buyer's guide), your Review schema is the canonical signal telling AI engines what the business/product ratings actually are. Your site becomes a source AI engines cite in response queries. Be meticulous about itemReviewed — link each review to the correct entity URL, and if you aggregate reviews for a business also declared elsewhere on the web, use sameAs on the itemReviewed to merge with their canonical entity.
How does Google handle self-reviews?+
Google's guidelines are explicit: don't add Review schema for your own products or services as a first-party claim. A company cannot schema itself a 5-star review of its own product and have that surface in rich results. The self-reviews might still technically validate, but they won't earn rich-result display and can trigger a manual action if Google perceives manipulation. Third-party reviews (reviewer is not the company) are fine. Customer testimonials on your site get a middle treatment — they're accepted if clearly attributed to real customers (with names, dates, and verification), but increasingly filtered by AI engines in favor of independent review platforms.
Should reviews have dates?+
Always. datePublished is a critical signal for AI engines because review freshness matters. A product with mostly 2019 reviews gets cited less in 2026 than a product with recent reviews, even if the older reviews are more positive. Include datePublished on every Review object; for editorial reviews, also include dateModified when substantively updated. AI shopping assistants weight recency heavily for categories that evolve quickly (electronics, software, restaurants) and less for stable categories (books, movies, furniture).
Can I link a review to a video or image of the reviewer?+
Yes. Review schema supports media attachments via image, video, and associatedMedia. For video reviews (a YouTube unboxing, a critic's video review), link the video via video field pointing to a VideoObject. For written reviews with photos (a restaurant review with dish photos, a product review with usage shots), add images via image. AI engines increasingly cite reviews that include media because the media provides verification — and in AI surfaces that render inline images (Perplexity cards, ChatGPT's image-enabled responses), review media shows up directly.

Related Schema Generators

Ad