Book Schema

TL;DR — Book schemaBook schema (schema.org/Book) is structured data describing a book's title, author, ISBN, and format. It can power Google's review-snippet rich result and — for approved feed providers — the registration-gated Book Actions program, and it helps Google disambiguate book entities even without a visible rich result. is code you add to a page that labels a book’s details for search engines — “this is the title,” “this is the author,” “this is the ISBN.” There’s a lot of confusion online right now that “Google removed Book schema in 2025.” That’s not what happened — Google announced it, then didn’t actually pull it. Adding Book markup still helps search engines understand your book, and it can make a book-review page eligible for star-rating snippets. Just don’t expect the fancy “buy or borrow” buttons from markup alone — those come from a separate, invite-only Google program.

What Book schema is

When you read a book’s page, you can tell the title from the author from the price just by looking. A search engine sees plain text and has to guess. Book schema spells it out in code, using the shared schema.org vocabulary — it tags each part of the page with what it actually is: the name (title), the author, the isbn, when it was published, how many pages, what format.

It’s almost always written as JSON-LDJSON-LD (JavaScript Object Notation for Linked Data) is a script-based structured data format, typically paired with the schema.org vocabulary to describe page content for search engines and AI systems. Google recommends it over Microdata and RDFa because it's the easiest format to implement and maintain at scale — but all three work, and structured data isn't a ranking signal. — a small block of code that sits in the page without changing how the page looks.

Book schema vs. “Book Actions” — not the same thing

Here’s the first thing that trips people up. There are really two separate ideas:

• Plain Book schema — the schema.org/Book markup any site can add to describe a book. This is the ordinary structured dataStructured data is a standardized way of labeling page content (using the schema.org vocabulary in JSON-LD, Microdata, or RDFa) so search engines can understand its meaning. It's not a direct ranking factor — its value is rich results and entity understanding. you’d expect.
• Book Actions — a special Google feature that puts “buy” or “borrow” buttons in search results. This is not something you turn on by adding markup to a page. It’s an invite-only program that big book retailers and library systems apply to and submit a data feed for — much more like uploading a catalog than sprinkling schema on a page.

Most people reading this aren’t eligible for Book Actions, and that’s fine — plain Book markup is still worth doing.

About that “removed in 2025” thing

You may have seen a headline that Google “removed Book schema” (or Book Actions) in 2025. Here’s the honest version: Google announced in June 2025 that it planned to phase out seven structured data types, and Book Actions was on the list. But when it came time to actually pull them, Book Actions was the one type Google left alone — and by November 2025 Google had even removed the “this is going away” banner from its own Book documentation. So the feature is still alive.

The takeaway for a beginner: don’t rip your Book markup out because of a stale headline. It still works fine as structured data.

What Book markup is still good for

Even if you’ll never touch the invite-only Book Actions program, plain Book markup earns you two real things:

• ⭐ Star ratings on book-review pages. Google explicitly counts Book as one of the things you’re allowed to review for its rating-stars snippet — no invite needed.
• 🧠 Better understanding of your book as an entity, especially if you link it to its Wikipedia, Wikidata, or Goodreads page with sameAs.

What it does not do: it doesn’t get your book into Google Books (that’s a totally separate publisher program), and — like all schema — it’s not a magic ranking boost. It makes pages eligible for extras; Google still decides whether to show them.

Want the full version — the feed-vs-markup distinction, the exact 2025 timeline, the ISBN rules, and what to actually do? Switch to the Advanced tab.

TL;DR — schema.org/Book (a child of CreativeWork) is structured dataStructured data is a standardized way of labeling page content (using the schema.org vocabulary in JSON-LD, Microdata, or RDFa) so search engines can understand its meaning. It's not a direct ranking factor — its value is rich results and entity understanding. describing a book — name, author, isbn, datePublished, bookFormat, numberOfPages, bookEdition, illustrator. It is not the same as Google’s Book Actions: a registration-gated, feed-based “buy/borrow” program for approved book providers and library systems, structurally closer to Merchant CenterGoogle Merchant Center (GMC) is a free platform where retailers upload and manage product data so their products can appear across Google — Shopping, organic Search product grids, Images, Lens, and AI surfaces. Since 2020 it powers free (organic) product listings, not just paid Shopping ads. than to on-page JSON-LDJSON-LD (JavaScript Object Notation for Linked Data) is a script-based structured data format, typically paired with the schema.org vocabulary to describe page content for search engines and AI systems. Google recommends it over Microdata and RDFa because it's the easiest format to implement and maintain at scale — but all three work, and structured data isn't a ranking signal.. The story most content gets wrong: Book Actions was announced for deprecation in June 2025 (one of seven types), but was the one excluded when Google actually pulled Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. reporting for the others in September 2025, and Google removed the deprecation banner from the /book doc entirely by November 2025 — so “Book schemaBook schema (schema.org/Book) is structured data describing a book's title, author, ISBN, and format. It can power Google's review-snippet rich result and — for approved feed providers — the registration-gated Book Actions program, and it helps Google disambiguate book entities even without a visible rich result. was removed” is now itself a myth. Realistic advice for almost everyone: keep/add plain Book markup for Review snippetReview schema (schema.org/Review) is structured data for a single critic's or user's evaluation of one specific thing — one author, one itemReviewed, one reviewRating — distinct from AggregateRating, which summarizes many reviews into an average. eligibility (Book is a listed reviewable type) and entity clarity via sameAs; pursue Book Actions only if you’re a high-volume provider willing to onboard. Markup doesn’t feed Google Books and isn’t a ranking factor.

schema.org/Book — the vocabulary

Book is a child of CreativeWork, so it inherits a lot and adds a few book-specific properties. The six that are specifically Book:

Plus the inherited CreativeWork/Thing staples you’ll actually use: name (title), author, publisher, datePublished, inLanguage, genre, aggregateRating, review, image, url, sameAs, and identifier. Two subtypes worth a mention: Audiobook (for audio editions) and SequentialArt (comics/graphic novels).

Book schema vs. Book Actions — the distinction the whole topic hinges on

This is the confusion I most want to kill, and it’s the opposite of how Product schemaProduct schema (schema.org/Product) is structured data that tells search engines a page's product name, price, availability, and reviews so it can appear in Shopping-style rich results. It's separate from a Google Merchant Center feed, though Google reconciles the two. works, so slow down here.

With Product schema, you add JSON-LDJSON-LD (JavaScript Object Notation for Linked Data) is a script-based structured data format, typically paired with the schema.org vocabulary to describe page content for search engines and AI systems. Google recommends it over Microdata and RDFa because it's the easiest format to implement and maintain at scale — but all three work, and structured data isn't a ranking signal. to a page and you’re eligible for a rich result. Book Actions is not like that. Book Actions is a feed submission program for onboarded providers — structurally much closer to a Google Merchant CenterGoogle Merchant Center (GMC) is a free platform where retailers upload and manage product data so their products can appear across Google — Shopping, organic Search product grids, Images, Lens, and AI surfaces. Since 2020 it powers free (organic) product listings, not just paid Shopping ads. feed than to on-page markup. Google’s own framing: Book Actions make Search “an entry point to discover books and authors,” letting users “quickly buy or borrow” a book directly from results, and “as a provider of books, you can provide a feed of data to Google.” The two actions are ReadAction (buy) and BorrowAction (borrow, for libraries).

And it’s gated. Straight from the doc: the feature “is currently limited to book providers with a wide selection of available books,” and “the use of this feature is limited to book providers that have filled out the interest form and have been onboarded.” You register interest; interest doesn’t guarantee participation. If you’re a book blogger, an author with a handful of titles, or an SEO auditing a client’s markup, you are almost certainly not going to earn a Book Actions rich result from on-page JSON-LD — that’s not a limitation of your implementation, it’s how the program is designed.

Work vs. Edition — how Book Actions models a book

Book Actions uses a nested-entity model that’s unusual relative to Product’s flatter shape. Google distinguishes the Work (the abstract book — title, author, original language) from the Edition (a concrete copy — year, edition name, ISBN). Library lending adds a Library / LibrarySystem layer on top for the borrow flow. You don’t need this for plain entity markup — it’s specific to the feed program — but it explains why Book Actions docs look heavier than “add some JSON-LD.”

ISBN is the reconciliation key (ISBN-13 only)

In the feed program, ISBN is the matching signal — directly analogous to GTINProduct identifiers are the standardized values — GTIN (Global Trade Item Number), MPN (Manufacturer Part Number), and brand — that shopping feeds like Google Merchant Center and Microsoft Merchant Center use to match a product listing to the correct item in their catalog. A GTIN alone is usually enough; without one, brand + MPN is the fallback; products with none declare that with the identifier_exists attribute. in Product. Google: “The ISBN is the main reconciliation signal when Google Search matches your feed data to Google’s data,” and without it “it’s likely that your books can’t be matched and therefore won’t be served.” One hard rule people miss: Google doesn’t accept ISBN-10. Its warning is explicit — for ISBN-10-only books, convert to ISBN-13 before sending the feed.

If you’re actually onboarded: the provider checklist

For the small slice of readers who are genuinely a high-volume book provider, here’s the shape of the work, condensed:

1. Onboarding — fill out the interest form; indicating interest doesn’t guarantee participation, and Google decides who it onboards.
2. Feeds — sellers upload a Book feed; lenders upload separate Book and Library feeds, using the upload details Google’s support team supplies once you’re onboarded.
3. Work/Edition — model the abstract Work once, then group each concrete Edition under it with workExample rather than flattening everything into one entity.
4. Supported identifiers — ISBN-13 is preferred and ISBN-10 is rejected; Google also accepts a few alternates (OCLC, LCCN, JP e-code) for books that genuinely lack an ISBN.
5. Action targets — set ReadAction for buy and BorrowAction for borrow. Note that Google personalizes and orders which providers actually show to a given user; which provider “wins” a slot isn’t something any single feed controls.
6. Validation — run the feed’s JSON-LD through the Schema Markup Validator before submission, same as you would for plain on-page markup.

The confusing 2025 timeline, explained

This is the part almost no other content has gotten right, so here’s the accurate sequence. Google spent 2025 trimming the SERP, and Book Actions got caught in the announcement but not the execution.

June 12, 2025 — the announcement. Google (Henry Hsu) posts that seven structured data types will be phased out, “because our analysis shows that they’re not commonly used in Search.” The seven, verbatim: “Book Actions, Course Info, Claim Review, Estimated Salary, Learning Video, Special Announcement, Vehicle Listing.” Note the scope caveat in the same post: “The use of these structured data types outside of Google Search (and dependent features) is not affected.”

September 9, 2025 — the execution, minus one. Google actually removed Search Console reporting, Rich ResultsRich results (formerly 'rich snippets') are enhanced search listings — stars, images, prices, breadcrumbs, video thumbnails, and more — that Google and Bing build from structured data. They're a display feature, not a ranking factor, and eligibility never guarantees they'll show. Test support, and API fields for the deprecated types. But — as Barry Schwartz reported at Search Engine Land — Google only removed reporting for six of them: Course Info, Claim Review, Estimated Salary, Learning Video, Special Announcement, and Vehicle Listing. His line is the load-bearing fact of this whole article: “Google did not mention Book Actions in this announcement.” Book Actions was quietly left off the removal list.

November 5, 2025 — the banner comes off. That’s Google’s own changelog date for removing the deprecation banner from the /book documentation entirely (Barry Schwartz’s Search Engine Roundtable piece reporting on it published the next day, November 6). Schwartz’s read, reporting what Google told him: Google did this “as there’s still a feature using the markup in Google Search.” In other words, the feature survived — the June announcement was the plan, not the outcome. To be precise about scope: this changelog entry confirms specifically that Book Actions still uses the markup — it isn’t independent proof of every other claim about plain schema.org/Book markup made elsewhere in this article.

December 2025 onward — where it stands. The /book doc remains actively maintained (its own “last updated” date is December 10, 2025), it carries a “Limited Access” flag in Google’s own docs navigation, and it’s still feed-based and provider-gated. So the accurate status is: alive, but gated — not general-availability, and not removed.

If you saw Book Actions listed somewhere as “removed in June 2025” (including in older versions of this site’s own structured-data hub), that’s the announcement being mistaken for the result. The record needs correcting, and this article is that correction.

Should you bother with Book Actions specifically?

Honest answer: only if you’re a high-volume book retailer or a library system with a wide selection, willing to register interest, get onboarded, and submit and maintain a data feed. For that audience, it’s a real, still-live surface worth pursuing. For everyone else — the large majority of people who land on a “book schema” search — Book Actions isn’t a realistic target, and chasing it via on-page markup is wasted effort.

What plain Book markup is still good for — no gatekeeping

Set Book Actions aside and Book markup still earns its keep:

• Review snippet eligibility. Book is one of the entity types Google’s Review snippet rich result explicitly supports — listed right alongside Recipe, Movie, Product, SoftwareApplication, and LocalBusiness. This is a fully general-availability path (no onboarding gate) for book-review sites, book bloggers, and retailers with genuine reader ratings. If you have real ratings, mark them up with aggregateRating / review — the same properties and the same genuine-review discipline covered in the Review schemaReview schema (schema.org/Review) is structured data for a single critic's or user's evaluation of one specific thing — one author, one itemReviewed, one reviewRating — distinct from AggregateRating, which summarizes many reviews into an average. and AggregateRating schemaAggregateRating schema (schema.org/AggregateRating) is structured data that represents the average of many ratings or reviews of an item, usually nested inside a parent type (or standalone with itemReviewed) to power star-rating rich snippets in search results. siblings. Eligible doesn’t mean guaranteed — Google still decides whether to render the stars.
• Entity / Knowledge GraphThe Knowledge Graph is Google's database of entities — people, places, organizations, and things — and the factual relationships between them. It's separate from any single website's structured data: your schema markup is one of many possible inputs to the graph, not the graph itself. clarity. Even with zero rich-result ambition, linking a Book (or its author Person) to Wikipedia, Wikidata, or Goodreads via sameAs is a low-effort, ungated way to help Google disambiguate which book and which author you mean. (I won’t claim a specific AI-citation or ranking effect from this — there’s no documented evidence of one — but disambiguating your entity is sound practice on its own terms.) This is under-covered relative to the property-table content that dominates search results, and it’s where a lot of the quiet value of Book markup actually lives now.

Two things it does not do, to be precise:

• It doesn’t feed Google Books. There’s no documented connection between on-page schema.org/Book markup and Google Books ingestion — Google Books’ catalog comes from its own partner/publisher pipeline. Don’t add markup expecting it to get your book listed there.
• It isn’t a ranking factor. Like all structured data, it aids understanding and rich-result eligibility. Eligibility isn’t display, and neither is a ranking boost.

Example: plain entity-clarity JSON-LD

This is ordinary Book markup for entity understanding and Review-snippet eligibility — note the sameAs links and the aggregate rating. This is not the Book Actions feed format, which is a separate spec you’d only touch as an onboarded provider.

{
  "@context": "https://schema.org/",
  "@type": "Book",
  "name": "The Left Hand of Darkness",
  "author": {
    "@type": "Person",
    "name": "Ursula K. Le Guin",
    "sameAs": "https://www.wikidata.org/wiki/Q181677"
  },
  "isbn": "9780441478125",
  "bookFormat": "https://schema.org/Paperback",
  "numberOfPages": 304,
  "datePublished": "1969",
  "inLanguage": "en",
  "publisher": {
    "@type": "Organization",
    "name": "Ace Books"
  },
  "sameAs": [
    "https://en.wikipedia.org/wiki/The_Left_Hand_of_Darkness",
    "https://www.wikidata.org/wiki/Q909751"
  ],
  "aggregateRating": {
    "@type": "AggregateRating",
    "ratingValue": "4.2",
    "ratingCount": "1187"
  }
}

The isbn here uses ISBN-13. The aggregateRating is what makes a genuine book-review page eligible for stars; the sameAs links are the ungated entity lever. None of this requires — or produces — Book Actions.

Bing and other engines

Keep this short: Bing has no Book-specific rich result and no Book Actions equivalent. Its structured-data help is generic — it reads schema.org/Book as ordinary entity data for its Satori knowledge base and general understanding, validates value shapes broadly, and publishes no book-specific requirements. So Book markup is useful to Bing as entity data, not as a route to any dedicated book feature.

For where Book schema sits in the wider picture, see the broader Schema MarkupSchema markup is code that uses the schema.org vocabulary to label what your content means so search engines can understand it and show rich results. It's most often written in JSON-LD, and it's not a direct ranking factor. and Structured Data hubs this article nests under, and the Product schemaProduct schema (schema.org/Product) is structured data that tells search engines a page's product name, price, availability, and reviews so it can appear in Shopping-style rich results. It's separate from a Google Merchant Center feed, though Google reconciles the two. / ProductGroup schema siblings for the useful structural contrast: any site can add Product markup for eligibility, whereas Book Actions requires provider onboarding — so for a book, the realistic rich-result path is the Review snippet, not Book Actions.

AI summary

A condensed take on the Advanced version:

• What it is: schema.org/Book (child of CreativeWork) is structured dataStructured data is a standardized way of labeling page content (using the schema.org vocabulary in JSON-LD, Microdata, or RDFa) so search engines can understand its meaning. It's not a direct ranking factor — its value is rich results and entity understanding. (usually JSON-LDJSON-LD (JavaScript Object Notation for Linked Data) is a script-based structured data format, typically paired with the schema.org vocabulary to describe page content for search engines and AI systems. Google recommends it over Microdata and RDFa because it's the easiest format to implement and maintain at scale — but all three work, and structured data isn't a ranking signal.) describing a book — name, author, isbn, datePublished, bookFormat, numberOfPages, bookEdition, illustrator, plus inherited aggregateRating/review/sameAs.
• Book schemaBook schema (schema.org/Book) is structured data describing a book's title, author, ISBN, and format. It can power Google's review-snippet rich result and — for approved feed providers — the registration-gated Book Actions program, and it helps Google disambiguate book entities even without a visible rich result. ≠ Book Actions. Book Actions is a registration-gated, feed-based “buy/borrow” program (ReadAction/BorrowAction) for onboarded book providers and library systems — structurally closer to Merchant CenterGoogle Merchant Center (GMC) is a free platform where retailers upload and manage product data so their products can appear across Google — Shopping, organic Search product grids, Images, Lens, and AI surfaces. Since 2020 it powers free (organic) product listings, not just paid Shopping ads. than to on-page markup. You can’t earn it from page JSON-LDJSON-LD (JavaScript Object Notation for Linked Data) is a script-based structured data format, typically paired with the schema.org vocabulary to describe page content for search engines and AI systems. Google recommends it over Microdata and RDFa because it's the easiest format to implement and maintain at scale — but all three work, and structured data isn't a ranking signal. alone; providers register interest and get onboarded.
• Work vs. Edition: Book Actions models the abstract book (Work: title, author, language) separately from a concrete copy (Edition: year, ISBN); library lending adds Library/LibrarySystem. ISBN-13 is the reconciliation key — ISBN-10 is rejected and must be converted.
• The 2025 timeline (the core, differentiated story): June 12, 2025 — announced for phase-out with six other types. September 9, 2025 — Google removed Search Console reporting for the other six; Book Actions was excluded (“Google did not mention Book Actions in this announcement”). November 5, 2025 — Google’s changelog date for removing the deprecation banner from the /book doc, because a feature still uses the markup (third-party coverage of it published November 6). Now — alive but gated (“Limited Access”), actively maintained docs. So “Book schema was removed in 2025” is a myth.
• What definitely was removed: the other six — Course Info, Claim Review, Estimated Salary, Learning Video, Special Announcement, Vehicle Listing.
• Realistic advice: pursue Book Actions only if you’re a high-volume retailer/library system. Everyone else: keep/add plain Book markup for Review snippet eligibility (Book is a listed reviewable type — ungated) and entity clarity via sameAs to Wikipedia/Wikidata/Goodreads.
• Two non-facts to correct: Book markup does not feed Google Books (separate publisher pipeline) and structured data isn’t a ranking factor — eligibility, not display.
• Bing: no Book-specific rich result or docs; reads Book as generic entity data.

Official documentation

Primary-source documentation from the search engines.

Google

• Book actions (Book) structured data — the feature doc for the feed-based Book Actions program: ReadAction/BorrowAction, the Work/Edition model, the ISBN-13 requirement, the registration/onboarding gate, and library lending. (Actively maintained; carries a “Limited Access” flag; no deprecation banner as of this writing.)
• Simplifying the search results page (Search Central Blog, June 12, 2025, Henry Hsu) — the announcement that named Book Actions among seven types to be phased out.
• Here’s an update on our efforts to simplify the search results page (Search Central Blog, November 5, 2025) — broader “simplify the results page” policy context (not Book-specific).
• Review snippet (Review, AggregateRating) structured data — the general-availability path for stars; Book is a listed reviewable entity type.
• Search Gallery — the general supported-types grid (Book Actions is not in it; Book appears only as a Review-snippet-eligible entity).
• schema.org/Book — the canonical vocabulary: property list and CreativeWork/Thing inheritance.

Bing / Microsoft

• Marking up your site with structured data — Bing’s generic structured-data support; no Book-specific requirements or Book Actions equivalent.

Quotes from the source

On-the-record statements from Google. Where a source page exposes the text, the link is a deep link that jumps to the quoted passage.

Google docs — what Book Actions is

• “Book actions make Google Search an entry point to discover books and authors. This enables Search users to quickly buy or borrow the books that they find directly from Search results… As a provider of books, you can provide a feed of data to Google with the structured dataStructured data is a standardized way of labeling page content (using the schema.org vocabulary in JSON-LD, Microdata, or RDFa) so search engines can understand its meaning. It's not a direct ranking factor — its value is rich results and entity understanding. schema provided here. Our specification provides ReadAction to let users buy a book and BorrowAction to borrow a book.” Jump to quote

Google docs — the registration gate (why this isn’t just “add JSON-LDJSON-LD (JavaScript Object Notation for Linked Data) is a script-based structured data format, typically paired with the schema.org vocabulary to describe page content for search engines and AI systems. Google recommends it over Microdata and RDFa because it's the easiest format to implement and maintain at scale — but all three work, and structured data isn't a ranking signal.”)

• “Note: To ensure robust coverage and better serve Search users, this feature is currently limited to book providers with a wide selection of available books. If you’d like to participate, register your interest. If you indicate interest in this feature, it doesn’t guarantee your participation.” Jump to quote
• “The use of this feature is limited to book providers that have filled out the interest form and have been onboarded.” Jump to quote

Google docs — Work vs. Edition

• “Work: The abstract concept of a book. Specifically, metadata such as the title, author, and original language are attributes for a work. Edition: A concrete copy of the book. Specifically, metadata such as the year of publication, name of the edition, and International Standard Book Number (ISBN) are attributes for the edition.” Jump to quote

Google docs — ISBN as the reconciliation key

• “The ISBN is the main reconciliation signal when Google Search matches your feed data to Google’s data. You must provide an ISBN or other supported identifier for all the books you wish to have surfaced in Search results. Without it, it’s likely that your books can’t be matched and therefore won’t be served.” Jump to quote
• “Warning: Google Search doesn’t accept ISBN-10. For books that have only ISBN-10 information, use an ISBN conversion tool to convert to ISBN-13 before you send your feed to Google.” Jump to quote

Google Search Central Blog — the June 2025 announcement

• “The following structured data types will no longer be supported in Google Search results and will be phased out over the coming weeks and months: Book Actions, Course Info, Claim Review, Estimated Salary, Learning Video, Special Announcement, Vehicle Listing.” Jump to quote
• “The use of these structured data types outside of Google Search (and dependent features) is not affected.” Jump to quote

Book Actions, or just plain Book markup?

The most useful decision here isn’t “which property” — it’s figuring out whether the gated Book Actions program is even relevant to you, or whether you just want entity-level Book markup. Walk the tree.

Book schema myths and mistakes to avoid

The traps around this topic are unusually thick right now, mostly because of stale 2025 coverage. Here’s what’s actually true.

Myth: “Book schemaBook schema (schema.org/Book) is structured data describing a book's title, author, ISBN, and format. It can power Google's review-snippet rich result and — for approved feed providers — the registration-gated Book Actions program, and it helps Google disambiguate book entities even without a visible rich result. (or Book Actions) was removed by Google in 2025.” This is the big one, and it’s now itself a myth to correct. June 2025 was the announcement that seven types would be phased out. When Google actually pulled Search Console reporting in September 2025, it did so for six of them — Book Actions was excluded (“Google did not mention Book Actions in this announcement”). By November 2025 Google removed the deprecation banner from the /book doc entirely, because a feature still uses the markup. The feature is alive, just gated. Don’t rip out Book markup on the strength of a stale headline.

Myth: “Adding schema.org/Book markup to my page will get me a Book Actions rich result.” Book Actions is a registration-gated, onboarded-provider feed program — not something that activates from on-page JSON-LDJSON-LD (JavaScript Object Notation for Linked Data) is a script-based structured data format, typically paired with the schema.org vocabulary to describe page content for search engines and AI systems. Google recommends it over Microdata and RDFa because it's the easiest format to implement and maintain at scale — but all three work, and structured data isn't a ranking signal., regardless of the removal drama. Adding markup to a page is not the mechanism.

Myth: “Book schema helps get my book into Google Books.” No documented connection. Google Books’ catalog comes from a separate partner/publisher ingestion pipeline, not from crawlingCrawling is how search engines use automated bots (like Googlebot and Bingbot) to discover URLs and download pages. A page has to be crawlable to be indexed, but crawling on its own isn't a ranking factor. schema.org/Book JSON-LDJSON-LD (JavaScript Object Notation for Linked Data) is a script-based structured data format, typically paired with the schema.org vocabulary to describe page content for search engines and AI systems. Google recommends it over Microdata and RDFa because it's the easiest format to implement and maintain at scale — but all three work, and structured data isn't a ranking signal. on arbitrary sites. Adding markup won’t list you there.

Myth (from stale competitor guides): “Book schema powers a general Book Actions carousel and Google Assistant results for any implementer.” Some guides still present Book Actions as a normal, generally-available feature any site earns from markup — and reference “Google Assistant,” a brand Google has been winding down. Both are staleness tells. Book Actions is gated to onboarded providers; there’s no open carousel you unlock with page markup.

Myth: “Since Book Actions is basically dead, there’s no reason to add Book schema at all.” Book remains a fully-supported Review snippetReview schema (schema.org/Review) is structured data for a single critic's or user's evaluation of one specific thing — one author, one itemReviewed, one reviewRating — distinct from AggregateRating, which summarizes many reviews into an average. entity type (ungated) and legitimate entity/Knowledge-Graph infrastructure via sameAs, entirely independent of the Book Actions program’s status. The value didn’t disappear.

Mistake: sending ISBN-10 to the Book Actions feed. Google explicitly rejects ISBN-10 — convert to ISBN-13 first. ISBN is the reconciliation key; get it wrong and your books can’t be matched.

Mistake: treating “valid markup” as “displayed rich result.” Eligibility isn’t display. Valid Book markup makes a review page eligible for stars; Google still decides whether to show them. And it’s never a ranking boost.

Prove your Book markup actually did what you intended

Adding Book JSON-LDJSON-LD (JavaScript Object Notation for Linked Data) is a script-based structured data format, typically paired with the schema.org vocabulary to describe page content for search engines and AI systems. Google recommends it over Microdata and RDFa because it's the easiest format to implement and maintain at scale — but all three work, and structured data isn't a ranking signal. proves nothing by itself — “the tag is on the page” and “Google recognizes it as eligible for something” are two different facts. These tests check the outcome, not just the presence of markup, for the two changes almost everyone reading this article is actually able to make: adding review markupReview schema (schema.org/Review) is structured data for a single critic's or user's evaluation of one specific thing — one author, one itemReviewed, one reviewRating — distinct from AggregateRating, which summarizes many reviews into an average., and getting the ISBN field right.

Test 1 — Your book-review page is Review-snippet eligible

• Test to run — Paste the page’s JSON-LDJSON-LD (JavaScript Object Notation for Linked Data) is a script-based structured data format, typically paired with the schema.org vocabulary to describe page content for search engines and AI systems. Google recommends it over Microdata and RDFa because it's the easiest format to implement and maintain at scale — but all three work, and structured data isn't a ranking signal. (or fetch the live URL) into the Rich-Result Eligibility Checker, or run it through Google’s Rich ResultsRich results (formerly 'rich snippets') are enhanced search listings — stars, images, prices, breadcrumbs, video thumbnails, and more — that Google and Bing build from structured data. They're a display feature, not a ranking factor, and eligibility never guarantees they'll show. Test. Look specifically at the Book entry and whether aggregateRating/review register as present and valid.
• Expected result — The checker shows Book as eligible for the Review snippet, with no missing-required-field errors on ratingValue, ratingCount/reviewCount, or the author/name fields the rating hangs off of.
• Failure interpretation — A missing-field error almost always means the aggregateRating block isn’t nested correctly under the Book, or the rating values are missing/non-numeric. An “ineligible” result with no field errors usually means the page doesn’t actually carry genuine user ratings — don’t fabricate one to pass the test.
• Monitoring window — Immediate — this checks what you’re serving right now, not Google’s crawl/render cycle.
• Rollback trigger — Any required-field error, or a rating that isn’t backed by real reader reviews — pull the aggregateRating block rather than ship a rating you can’t stand behind; a fabricated rating is a policy problem, not just an SEO one.

Test 2 — Your ISBN is a valid, submission-ready ISBN-13

• Test to run — Run the isbn value through the Schema Markup Validator (it flags malformed ISBN strings) and confirm by eye that it’s 13 digits, not 10.
• Expected result — A 13-digit ISBN with no formatting errors flagged. This matters most if you’re one of the readers actually pursuing Book Actions, where — per Google’s own doc — ISBN is “the main reconciliation signal” and ISBN-10 submissions are rejected outright.
• Failure interpretation — A 10-digit isbn value means you’re shipping a format Google’s Book Actions feed will not accept; for plain entity markup it still validates, but it’s worth fixing anyway since ISBN-13 is the modern standard.
• Monitoring window — Immediate — ISBN format is a static property, not something that needs time to resolve.
• Rollback trigger — Any ISBN-10-only value on a page you intend to keep long-term — convert it to ISBN-13 before treating the markup as done.

Tools for building and checking Book markup

Three of my own free tools cover the actual workflow for this article — write the markup, confirm it’s valid, then confirm what it’s eligible for.

• Schema Markup Generator — the place to start if you’re writing Book markup from scratch. It has a form for the Book type (and its inherited CreativeWork/Thing properties like author, isbn, datePublished, aggregateRating), so you’re not hand-writing JSON-LDJSON-LD (JavaScript Object Notation for Linked Data) is a script-based structured data format, typically paired with the schema.org vocabulary to describe page content for search engines and AI systems. Google recommends it over Microdata and RDFa because it's the easiest format to implement and maintain at scale — but all three work, and structured data isn't a ranking signal. from memory, and it exports clean, copy-pasteable output.
• Schema Markup Validator — paste your Book JSON-LDJSON-LD (JavaScript Object Notation for Linked Data) is a script-based structured data format, typically paired with the schema.org vocabulary to describe page content for search engines and AI systems. Google recommends it over Microdata and RDFa because it's the easiest format to implement and maintain at scale — but all three work, and structured data isn't a ranking signal. (or a full page) here to catch structural problems before Google does: malformed ISBNs, missing required fields, a misnested aggregateRating, or an @id graph that doesn’t connect properly to the rest of the page’s schema.
• Rich-Result Eligibility Checker — this is the one that answers “am I actually eligible for anything?” — it tells you per-type whether your Book markup qualifies for the Review snippetReview schema (schema.org/Review) is structured data for a single critic's or user's evaluation of one specific thing — one author, one itemReviewed, one reviewRating — distinct from AggregateRating, which summarizes many reviews into an average., and if not, exactly which required field is missing. Since Book isn’t in Google’s general Search Gallery and only qualifies through the Review snippet path, this is the realistic eligibility check for almost everyone reading this article — Book Actions itself isn’t something any public tool can test, since eligibility there depends on Google’s provider-onboarding decision, not on markup validity.

Metrics for Book markup

Valid eligible book-review pages

Metric: canonical review pages with valid Book plus supported review markupReview schema (schema.org/Review) is structured data for a single critic's or user's evaluation of one specific thing — one author, one itemReviewed, one reviewRating — distinct from AggregateRating, which summarizes many reviews into an average.. What it tells you: whether the implementation remains eligible for the review treatment it can actually support. How to pull it: crawl JSON-LDJSON-LD (JavaScript Object Notation for Linked Data) is a script-based structured data format, typically paired with the schema.org vocabulary to describe page content for search engines and AI systems. Google recommends it over Microdata and RDFa because it's the easiest format to implement and maintain at scale — but all three work, and structured data isn't a ranking signal. and validate representative templates. Benchmark / realistic range: every intended eligible page should pass; ordinary book-detail pages without reviews belong in a separate cohort. Cadence: every structured-data release and monthly sampling.

Structured-data coverage

Metric: intended book pages containing a stable identifier, name, author, and other modeled properties that match visible content. What it tells you: whether templates consistently describe the entity. How to pull it: compare a crawl extraction with the catalog source. Benchmark / realistic range: establish coverage by template and explain intentional omissions rather than inventing a universal target. Cadence: monthly.

Search appearance performance

Metric: impressions, clicks, and click-through rate for the affected page cohort and any applicable Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. appearance filter. What it tells you: observed search performance, not a ranking effect of schema. How to pull it: Search ConsoleGoogle's free tool for monitoring crawling, indexing, and search performance. with annotated implementation dates. Benchmark / realistic range: compare with the site’s own pre-change period and unaffected pages; eligibility does not guarantee display. Cadence: monthly.

Test yourself: Book Schema

Five quick questions on schema.orgSchema markup is code that uses the schema.org vocabulary to label what your content means so search engines can understand it and show rich results. It's most often written in JSON-LD, and it's not a direct ranking factor./Book, the Book Actions program, and the 2025 timeline. Pick an answer for each, then check.

Resources worth your time

My writing on this topic

I haven’t published a standalone Book schemaBook schema (schema.org/Book) is structured data describing a book's title, author, ISBN, and format. It can power Google's review-snippet rich result and — for approved feed providers — the registration-gated Book Actions program, and it helps Google disambiguate book entities even without a visible rich result. guide of my own — this article is the dedicated treatment. So rather than point you at something that doesn’t exist, the honest move is to send you to the primary sources below and to the related structured-data work on this site. For the broader vocabulary and where Book fits, see the Schema MarkupSchema markup is code that uses the schema.org vocabulary to label what your content means so search engines can understand it and show rich results. It's most often written in JSON-LD, and it's not a direct ranking factor. and Structured DataStructured data is a standardized way of labeling page content (using the schema.org vocabulary in JSON-LD, Microdata, or RDFa) so search engines can understand its meaning. It's not a direct ranking factor — its value is rich results and entity understanding. hubs; for the ratings side, the Review schema and AggregateRating schemaAggregateRating schema (schema.org/AggregateRating) is structured data that represents the average of many ratings or reviews of an item, usually nested inside a parent type (or standalone with itemReviewed) to power star-rating rich snippets in search results. siblings; and for the useful structural contrast (open markup vs. gated feed program), the Product schemaProduct schema (schema.org/Product) is structured data that tells search engines a page's product name, price, availability, and reviews so it can appear in Shopping-style rich results. It's separate from a Google Merchant Center feed, though Google reconciles the two. and ProductGroup schemaProductGroup schema is structured data that groups product variants — like a shirt's sizes and colors — under one parent so Google understands they're options of the same item, not separate products. It wraps individual Product markup via hasVariant/variesBy/productGroupID; it doesn't replace it. deep dives.

From around the industry

• Book actions (Book) structured data (Google Search Central) — the authoritative feature doc: feed program, Work/Edition model, ISBN-13 rule, onboarding gate.
• Simplifying the search results page (Google Search Central Blog, Henry Hsu, June 2025) — the announcement that named Book Actions among seven types.
• Google drops reporting on several structured data types (Search Engine Land, Barry Schwartz, Sept 2025) — the key report: six of seven types had reporting removed, and “Google did not mention Book Actions.”
• Google Drops Support For More Structured Data Types & Search Features (Search Engine Roundtable, Barry Schwartz, Nov 2025) — the report that Google removed the deprecation banner from the Book doc because a feature still uses the markup.
• Google Retires 7 Structured Data Features To Streamline Search Results (Search Engine Journal, June 2025) — same-day coverage of the announcement.
• Review snippet structured data (Google) — the ungated path to stars; confirms Book as a reviewable entity type.
• schema.org/Book — the vocabulary itself, straight from the source.

A note on stale content you’ll run into: some widely-linked schema guides — for example the “Book Schema Markup” guide on schemavalidator.org — still describe Book Actions as a normal, generally-available feature any implementer earns from on-page markup, with no mention of the 2025 timeline, the registration gate, or the feed-vs-markup distinction (and they reference “Google Assistant,” a dated tell). I’m citing that here not as a source to trust, but as a concrete example of exactly the confusion this article exists to resolve — if you’ve read something like it, this is the correction.

Book Schema

Book schema (schema.org/Book) is structured data describing a book's title, author, ISBN, and format. It can power Google's review-snippet rich result and — for approved feed providers — the registration-gated Book Actions program, and it helps Google disambiguate book entities even without a visible rich result.

Related: Schema Markup, Structured Data, Review Schema, Product Schema

Book Schema

Book schema is the schema.orgSchema markup is code that uses the schema.org vocabulary to label what your content means so search engines can understand it and show rich results. It's most often written in JSON-LD, and it's not a direct ranking factor./Book structured-data type used to describe a book — its title (name), author, isbn, datePublished, numberOfPages, bookFormat, bookEdition, illustrator, and (via aggregateRating/review) reader ratings. It’s a child of CreativeWork, so it also inherits publisher, inLanguage, genre, image, url, and Thing-level name/sameAs/identifier.

The thing to keep straight: schema.org/Book markup is broader than the one Google feature that historically consumed it. Book Actions — a registration-gated, feed-based program that surfaces “buy” or “borrow” buttons for approved book providers and library systems — is not the same as adding on-page Book JSON-LDJSON-LD (JavaScript Object Notation for Linked Data) is a script-based structured data format, typically paired with the schema.org vocabulary to describe page content for search engines and AI systems. Google recommends it over Microdata and RDFa because it's the easiest format to implement and maintain at scale — but all three work, and structured data isn't a ranking signal.. Book Actions is closer to a Merchant CenterGoogle Merchant Center (GMC) is a free platform where retailers upload and manage product data so their products can appear across Google — Shopping, organic Search product grids, Images, Lens, and AI surfaces. Since 2020 it powers free (organic) product listings, not just paid Shopping ads. feed than to typical schema markupSchema markup is code that uses the schema.org vocabulary to label what your content means so search engines can understand it and show rich results. It's most often written in JSON-LD, and it's not a direct ranking factor.; most sites can’t earn it from page markup alone.

Book Actions has an unusually messy 2025 history that most content hasn’t caught up on. Google announced it for deprecation in June 2025 (one of seven structured dataStructured data is a standardized way of labeling page content (using the schema.org vocabulary in JSON-LD, Microdata, or RDFa) so search engines can understand its meaning. It's not a direct ranking factor — its value is rich results and entity understanding. types), but when Google actually removed Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. reporting for those types in September 2025, Book Actions was the one type excluded — and by November 2025 Google removed the deprecation banner from the /book doc entirely, because a feature still uses the markup in Search. So “Book schema was removed in 2025” is a myth: it was announced, not executed, and the feature remains live (if gated) for onboarded providers.

Outside Book Actions, Book remains a fully-supported Review snippetReview schema (schema.org/Review) is structured data for a single critic's or user's evaluation of one specific thing — one author, one itemReviewed, one reviewRating — distinct from AggregateRating, which summarizes many reviews into an average. entity type (ungated — Google lists it alongside Recipe, Movie, Product, SoftwareApplication, and LocalBusiness), and it feeds general entity/Knowledge GraphThe Knowledge Graph is Google's database of entities — people, places, organizations, and things — and the factual relationships between them. It's separate from any single website's structured data: your schema markup is one of many possible inputs to the graph, not the graph itself. understanding via sameAs links to Wikipedia/Wikidata/Goodreads. Bing reads schema.org/Book as ordinary entity data with no dedicated book rich result.

Related: Schema Markup, Structured Data, Review Schema, Product Schema

Changelog