Google Retires 7 Structured Data Features: What You Need to Know

Last Updated: April 1, 2026

---

• Google retired seven structured data features years ago, and today the real question is not “what disappeared” but “which schemas still move the needle for traffic, AI Overviews, and revenue.”
• You do not need to panic or rip out old markup, but you should audit your schema, double down on high-impact types, and stop chasing features that no longer show in search.
• Structured data now plays a bigger role in how Google’s AI understands your entities, products, people, and brand, not just how your result looks.
• If you focus on clear content, solid technical SEO, and a tight set of supported schema types, you will be in a much stronger spot than someone trying to mark up everything.

Google’s Structured Data Deprecations: What Actually Happened

Google retired support for seven structured data based rich features: Book actions, Course info, ClaimReview snippets, Estimated salary, Learning video, Special announcement, and Vehicle listing.

These changes are long rolled out now, so we are talking about them as a historical shift that kicked off a broader clean up of low impact rich results.

Here is the key thing that sometimes gets lost.

Google stopped using these types for special rich result presentations, but it did not punish sites that still have the markup in their code.

Google no longer shows rich results for these seven structured data types, but valid schema markup will not trigger a penalty and can remain on your pages.

Your snippets lost some extras, yes, but your rankings did not suddenly tank just because the rich feature went away.

Traffic changes came from how those extras affected click through rate and attention, not from an algorithmic penalty.

Technical Meaning Of “Removed”

When Google says a structured data feature is removed, here is what that really means in practice.

• Google’s rich result systems ignore those specific types for visual enhancements in the SERP.
• Your markup might still validate in tools, but it shows as not eligible for rich results.
• Google can still read that schema as plain data for understanding, it just does not attach a special search feature to it.

This is why stripping all old schema in a rush is usually a bad idea.

The markup might still help internal tools, other platforms, or future feature experiments, and removing it early gives you nothing.

Quick FAQ: Do You Need To Remove Old Markup?

Let me cut through the noise and hit the three questions I hear most.

• Should I remove the deprecated schema? Not immediately. If it serves no internal use at all, you can schedule cleanup as part of a broader code refactor.
• Will leaving it slow my site down? For 99% of sites, the performance impact of a few extra JSON LD blocks is trivial compared to images, scripts, and fonts.
• Does it still help Google understand my content? Sometimes. Google has been clear that schema is a hint, not a guarantee, but structured facts can still help its systems form a better picture of your page, even without a rich result attached.

If you are expecting magic from markup alone, that is where you are taking the wrong approach.

Structured data works best as a helper, not the hero.

The Seven Retired Structured Data Features

Here is a quick refresher on what Google removed from rich results and what each type was trying to do.

Structured data type	Old search feature	What changed
Book actions	Preview / buy buttons for books in snippets	No direct book action buttons tied to schema
Course info	Course rich cards with instructor, duration, etc.	Those specific course rich results no longer show
ClaimReview	Fact check snippets linked to checked claims	Heavily restricted and not a general purpose snippet
Estimated salary	Salary ranges highlighted in job results	Salary no longer lifted into a distinct rich feature
Learning video	Extra signals for educational videos	Replaced by more generic video handling
Special announcement	Prominent pandemic related announcements	Program ended, schema ignored for rich display
Vehicle listing	Enhanced car listings with specs and pricing	Shifted closer to feed based and Product flows

If that list feels a bit niche, that is the point.

These features did not affect most sites, so Google cut them to keep results cleaner and reduce spam surfaces.

Schema.org vs Google Support

One thing many people still mix up is the difference between schema.org and Google’s supported rich results.

Schema.org is just the shared vocabulary, the dictionary of types and properties that different platforms can use.

The fact that Google stops supporting a type for rich results does not mean the type is “dead” in schema.org or broken for other platforms.

You can still use Book, Course, or Vehicle related schema for internal search, partner APIs, or other search engines.

Google simply chose not to reward those specific types with extra visual treatment in its own SERPs.

Per Type: What You Should Do Now

Let’s walk through each retired feature and what makes sense now, from a practical point of view.

1. Book actions

Book action markup used to trigger preview or buy buttons directly in results, which was nice for publishers and stores.

Today you are better off focusing on more general schemas.

• Mark your books with Product, Offer, and Review where relevant.
• Make sure your book pages have clear purchase CTAs, structured URLs, and clean titles and meta descriptions.
• If you use feeds like Google Books or Merchant Center, keep those clean, because feeds now drive more rich experiences than niche schema types.

If your dev team is cleaning legacy code, they can remove Book actions markup during a natural upgrade cycle, but there is no emergency.

2. Course info

Course rich results tried to highlight course metadata like instructors and duration directly in search cards.

Those visuals are gone, yet the underlying Course type still has value outside Google.

• Keep Course markup if your LMS, internal search, or partner sites use it.
• For Google, focus on making each course page clear, crawlable, and structured with basic schemas like Organization, Product, and BreadcrumbList.
• Use strong titles that match what real students search for, such as “SEO course for beginners” instead of abstract labels.

One pattern I see with course sites is over reliance on schema as a shortcut instead of improving copy, structure, and navigation.

The sites that win treat schema as support, not the main weapon.

3. ClaimReview

ClaimReview is the trickiest one, because it ties into news, politics, and misinformation.

Google has narrowed when and how ClaimReview surfaces, and for many publishers the generic fact check rich snippets are simply not a thing anymore.

• If you are not a dedicated fact checking site, stop expecting ClaimReview to give you a special result type.
• If you are, follow Google’s fact check guidelines closely and focus on News SEO fundamentals: fast crawling, clear headlines, and tight topical authority.
• Use ClaimReview primarily as a signal of structure for machines, not as a promise of traffic.

I know some newsrooms that built entire strategies around fact check snippets and saw them vanish overnight.

That is a reminder that any single surface owned by Google is a risky dependency.

4. Estimated salary

Estimated salary rich results lifted salary ranges into the SERP next to job postings.

Those highlights are gone, but intent from job seekers around salary did not vanish.

• Keep using JobPosting schema with accurate details.
• Publish clear salary info on page where you can, since users and modern regulations often expect it.
• Measure job detail page CTRs and on page engagement instead of chasing a salary badge that no longer exists.

Recruitment teams that still hide salary are fighting user behavior and regulation, not just Google.

No schema trick will fix that.

5. Learning video

Learning video markup gave Google extra hints for educational videos, but it never became a major feature across most niches.

Today, generic video markup does the heavy lifting.

• Use VideoObject as your baseline schema for any video content.
• If your video teaches a step by step process, consider HowTo on the written version.
• Invest time in strong thumbnails, titles, and descriptions that map to user searches and AI summarized answers.

From what I see across client data, good VideoObject markup plus solid content beats clever, niche types that hardly anyone supports.

6. Special announcement

Special announcement schema was a kind of emergency tool, mostly during COVID.

It was always meant to be temporary and tightly controlled.

• Do not plan future communication strategies around special structured data flags like this.
• For time sensitive alerts, focus on fast crawling: news sitemaps, frequent updates, and clear site structure.
• If you are an organization that handles emergencies, build email, SMS, and app push channels, not just search dependencies.

Google will keep experimenting with temporary programs like this, but they come and go.

Your core communication stack needs to be independent of any one rich result program.

7. Vehicle listing

Vehicle listing schema tried to make car inventories richer directly in search, similar to products.

Over time, Google has leaned more on feeds and vertical programs for cars, just like for shopping.

• Use Product markup for each vehicle with price, brand, model, and availability where it makes sense.
• Connect to Google’s vehicle or inventory programs in your region through Merchant Center or partner feeds.
• Clean internal data first, because feeds built on messy inventory data never perform well, no matter how much schema you paste on top.

Dealers who fix their inventory data model usually see better search performance than those who obsess over microdata details alone.

The foundation wins here.

How These Changes Affect Rankings, CTR, And Traffic

A lot of people heard “rankings will not change” and took that to mean “impact is zero.”

That is not quite right.

When a rich feature disappears, your blue link might hold the same position, but your click-through rate and visibility can still drop.

Think about a job listing that used to show a clear estimated salary or a vehicle with rich specs.

Those visuals gave users a reason to click you instead of the result above or below.

Where The Real Impact Shows Up

The main places you feel these removals are.

• CTR changes for pages that previously surfaced rich snippets.
• Less SERP real estate for your brand on affected queries.
• Slight shifts in user behavior, especially in verticals like jobs, autos, or fact checking.

That is not a ranking penalty, but it still matters.

I have seen sites lose a noticeable share of clicks when a beloved snippet vanished, even though their average position chart looked flat.

How To Measure The Impact Properly

Instead of guessing, you can check how much those features mattered to you.

1. Open Google Search Console and go to Performance.
2. Use the “Search appearance” filter to see which rich results you previously qualified for.
3. Compare impressions and CTR for relevant appearances before and after the feature stopped showing.

In some cases, you will see almost no difference, which means those schemas were nice to have but not core.

In other cases, like a job board that leaned on salary snippets, you might see a real drop that forces you to adjust strategy.

Schema.org Types Still Valid, Even If Google Ignores The Feature

I want to stress this one more time because developers in particular worry about doing things “wrong.”

These schema types are still valid schema.org vocabulary.

• They will usually pass validation in structured data tools.
• They can still power internal tools, site search, partner APIs, or other search engines.
• They might be used quietly by Google for understanding, even without a public rich result.

So ripping them out instantly just to appease Google is overkill.

Plan cleanup on your timeline, not on a panic timeline.

Structured Data In The Age Of AI Overviews

The biggest shift since these deprecations is not which badge you get.

It is how Google’s AI systems now assemble answers pulled from across the web.

Structured data is fuel for AI Overviews: it helps Google identify entities, facts, relationships, and key attributes fast.

You are not just feeding a single blue link snippet anymore.

You are feeding a system that tries to summarize, compare, and quote across multiple pages.

How Structured Data Feeds AI Overviews

From what we see in the wild, a few patterns stand out.

• Well marked up entities like products, people, organizations, and events are more likely to appear in AI summaries and comparison blocks.
• Clear factual properties like price, rating, duration, and dates become ingredients for AI to answer “which, what, when, how much” type questions.
• Pages that mix clean schema with strong on page explanations tend to show up as cited sources more often.

There is no public, guaranteed formula here, and I would not trust anyone who says there is.

But consistent structured data on key entities clearly helps Google map your site into its knowledge graph.

Schema Types That Matter Most For AI Overviews

While every site is different, some schema types are showing up again and again as strong bets.

• Product for ecommerce, with price, availability, brand, and offers.
• Organization and LocalBusiness for brand and local signals.
• Person, Author, and ProfilePage for experts, content creators, and E E A T.
• Event for concerts, webinars, and conferences.
• Recipe, HowTo, and strong VideoObject where relevant.
• JobPosting for hiring and talent acquisition.

Notice what is not on that list anymore.

FAQ schema used to be a favorite trick, but Google has made that surface very limited now and treats it far more conservatively.

FAQ Schema’s Reduced Role

For a while, people spammed FAQ markup on almost every page just to claim more SERP space.

Google dialed that back hard.

• FAQ rich results now show for a small set of authoritative sites in specific contexts.
• Most generic FAQ implementations do not get any visible snippet at all.
• The markup still helps clarify Q&A structure, but it is not a growth hack for CTR anymore.

So if you are spending hours marking up every random FAQ section only for Google, you are misallocating energy.

Use FAQ where it genuinely improves UX and helps your own on site search or support, not as a primary SEO tactic.

Structured Data And E E A T: Marking Up People, Brands, And Trust

One of the biggest gaps I see on many sites is weak markup around who is behind the content.

Google’s focus on experience, expertise, authoritativeness, and trust has grown, and schema can help signal those things more clearly.

If your site talks like an expert but your structured data hides the author and brand, you are sending mixed signals.

Think about how you present your people, not just your products.

High Value E E A T Related Schema Types

If you publish content in YMYL spaces or any topic where trust matters, these types are worth your time.

• Person for authors, reviewers, and subject matter experts.
• Author and ReviewedBy properties on articles and guides.
• ProfilePage on author bio pages that link out to social profiles and credentials.
• Organization with clear legal name, logo, sameAs social links, and contact info.

A medical article marked up with correct Person and ReviewedBy details sends a clearer signal than an anonymous page.

That does not replace real world expertise, but it helps machines see it.

How To Apply This Without Overcomplicating It

You do not need a bloated schema graph to benefit from E E A T related markup.

Focus on simple, consistent patterns.

• Each article or guide should reference its author with a Person entity, even if that entity lives on a separate author page.
• Use the same Organization entity across the site instead of creating slight variations on every page.
• Link your schema to real world profiles where you can, like LinkedIn, professional directories, or respected associations.

Think of schema like a map of who you are, not just what you sell.

If that map is missing, AI systems will guess, and they are not always kind.

A Modern Strategy For Structured Data: Less Random, More Precise

A mistake I see a lot is treating schema like a sticker collection.

People try to add every type they can just to feel complete.

Good schema strategy is not “how many types can we add,” it is “which facts matter most for how users decide.”

That mindset shift alone can save your team a lot of wasted effort.

Three Guiding Questions For Schema In 2026

Before adding any new markup, run it through these questions.

• Does this markup clearly state a key fact that an AI or user needs to answer a specific question?
• Does this markup help clarify who is behind the content and why they are credible?
• Does this markup expose information that is critical for decisions, like price, availability, dates, or location?

If the answer is yes to at least two, you might be onto something useful.

If you are stretching just to justify it, probably skip it.

A Simple Decision Framework For Your Team

Here is a straightforward checklist you can use this quarter.

1. Audit: Export all schema types you currently use across your site.
2. Map: Compare them against Google’s current Search Gallery of supported features.
3. Measure: Use Search Console to review impressions and CTR by search appearance for each type.
4. Prioritize: Keep and improve types that are supported, drive clicks, and help your business beyond Google.
5. Deprioritize: Flag deprecated or zero impact types that are only there for Google and not used anywhere else.

This is not glamorous work, but it is what separates disciplined SEO programs from random tinkering.

You do not need 40 schema types; you need the right 6 to 10 done well.

Examples Of High Value Structured Data Now

Let me highlight a few types that, in my experience, still carry strong value in 2026.

• Product + Offer + Review: Core for ecommerce, impacts standard snippets, AI Overviews, and shopping surfaces.
• Organization / LocalBusiness: Key for brand visibility, local packs, and knowledge panels.
• Event: Helps events show with dates and locations, and feeds AI answers for “what is happening” queries.
• VideoObject: Supports video rich results, carousels, and AI summaries that embed or reference video content.
• JobPosting: Important for hiring, especially when combined with feeds and job vertical integrations.

These types have survived multiple waves of cleanup for a reason.

They tie directly to user tasks where structured facts matter.

Guidance For Developers And Technical SEOs

If you are the one touching the code, your questions are more practical.

You care about maintenance, performance, and not breaking anything while keeping marketing happy.

Should You Strip Unused Markup?

My honest view is a bit boring.

You should, but only as part of a normal cleanup cycle, not as a panic task after every Google announcement.

• If a schema type is used only for a deprecated Google feature and has zero internal use, schedule its removal during your next template refactor.
• If it powers internal tools, mobile apps, or other platforms, document that clearly and leave it.
• When you remove it, test affected templates in multiple tools to be sure you do not accidentally break supported schema around it.

Think of schema like any other code dependency.

Dead features get pruned when you touch that part of the system, not by dropping a grenade in the repo.

Performance Considerations

Developers sometimes worry that extra JSON LD blocks will slow pages down.

In practice, HTML size from schema is rarely the real bottleneck.

• Focus first on image compression, script weight, and render blocking resources.
• Once those are handled, look at schema payload size if your pages are truly huge.
• If your JSON LD is a mess, you can consolidate multiple blocks into one well structured graph, which is cleaner anyway.

Shaving a few kilobytes of markup helps, but not as much as cutting unused JavaScript or fixing slow APIs.

Do the big wins first.

Testing And Monitoring

Given how fast Google tweaks structured data support, testing is not optional.

But it does not have to be complicated.

• Use Google’s Rich Results Test and the schema.org validator to confirm syntax and supported types.
• Set up alerts in Search Console for enhancement issues on key schemas like Product, Event, and JobPosting.
• Track search appearance metrics over time so you catch when a feature shrinks or changes behavior.

I like to keep a simple internal changelog that says, “we added X schema on these templates on this date” and then check GSC 30 to 60 days later.

Without that, it is easy to forget what you tested and whether it moved anything.

What To Watch Next In Structured Data

The seven retired types we started with are part of a pattern.

Google is pruning surfaces that are low value, spam prone, or overlapping with feed based programs.

Expect Google to keep trimming niche rich results while doubling down on types that tie to commerce, local, video, and trustworthy content.

That might sound harsh, but it actually simplifies strategy for you.

You can focus on a smaller core set of schemas instead of chasing every new experimental badge.

Trends Shaping The Next Few Years

Based on where search is heading, a few trends are worth watching closely.

• AI Overviews getting better at summarizing structured facts, which raises the value of clean Product, Event, and Organization data.
• More integration with Merchant Center and similar feeds for shopping, travel, jobs, and vehicles.
• Greater scrutiny on spammy use of schema, especially around reviews, FAQs, and fake authorship.
• Growing importance of entity clarity, where schema helps define who and what your site actually represents.

This is why subscribing to the Google Search Central blog and checking the Search Status Dashboard is worth your time.

You do not need to obsess daily, but a quick scan each month helps you avoid surprises.

Next Steps For Your Site This Quarter

If you feel behind on structured data, here is a practical short list for the next few months.

1. Run a sitewide schema audit and map each type to:
   • Supported rich result type
   • Internal or external use
   • Business owner
2. Pick 3 to 5 core schema types that match your business model, like Product, Organization, Event, Person, or JobPosting.
3. Clean and standardize those core types across all templates, reducing inconsistencies and missing fields.
4. Measure impact in Search Console and analytics, focusing on CTR, conversions, and assisted revenue where you can.
5. Plan a slow, careful cleanup of deprecated or unused schemas to keep your codebase easier to maintain.

This is not flashy, but it brings structure to what can otherwise feel like chaos.

And that calm approach tends to win over time, especially when Google keeps changing what shows on the surface.

Bringing It All Together

The retirement of Book actions, Course info, ClaimReview snippets, Estimated salary, Learning video, Special announcement, and Vehicle listing is old news now.

The real story is how you adjust your structured data strategy so it matches where Google and users are actually headed.

Treat schema as a focused way to express your most important facts to both users and machines, not as a checklist of every possible type.

That means less energy spent chasing retired or fringe features and more time tightening up Product, Organization, Person, Event, Video, and JobPosting data.

It also means being honest about what is working by checking Search Console, not just assuming that one extra property made the difference.

If you keep your markup clean, stay close to Google’s current support docs, and anchor everything in strong content, you will be in a good spot.

Not perfect, because search is never that neat, but good enough to compete while everyone else is still reacting to last year’s changes.

From there, every structured data tweak you make is grounded in clear outcomes: better understanding, better visibility, and better decisions from your users.

That is where schema really earns its keep.

Need a quick summary of this article? Choose your favorite AI tool below: