Schema Markup Strategy for 2026: What to Keep, Fix, and Remove

BOOK A CALL

Schema Markup Strategy for 2026: What to Keep, Fix, and Remove

Update your schema markup strategy for 2026. Learn what schema types to keep, fix, or remove for better AI search visibility and rich results.

LoudScale Team
LoudScale Team
5 MIN READ

Schema Markup Strategy for 2026: What to Keep, Fix, and Remove

Schema markup isn’t dead—but the rules changed. On May 7, 2026, Google officially retired FAQ rich results. Combined with earlier HowTo deprecations and the January removal of seven structured data types, we now have a slimmer, stricter schema landscape. The good news? Pages with properly implemented structured data still earn 35% higher CTR from rich results. The strategy is what changed.

I’ve spent the last month auditing client sites and parsing every Search Central update. Here’s the 2026 schema markup strategy that actually works—broken down by what to keep, what to fix, and what to remove.

What Actually Changed in 2026

Let me cut through the noise. Here’s what Google actually retired:

  • FAQ rich results — Gone May 7, 2026. FAQPage schema itself is still valid; the SERP feature is dead.
  • HowTo rich results — Retired September 2023. Desktop and mobile.
  • Seven schema types — Removed from rich results: Book Actions, Course Info, Claim Review, Estimated Salary, Learning Video, Special Announcement, Vehicle Listing.

The pattern is clear: Google is simplifying the SERP by trimming niche features that got gamed at scale.

“Google is removing display features when they get aggressively scaled and stop faithfully describing pages.” — Lily Ray, VP SEO Strategy at Amsive

The 31 schema types that still generate active rich results? They share one trait: they tie directly to specific user intent signals where structured data genuinely improves the search result.

Schema Types to KEEP in 2026

These are your workhorses. Implement these first.

Article / BlogPosting Schema

Article schema remains foundational. Despite what you might have read, Google’s documentation (last updated December 2025) states: “There are no required properties.”

The recommended set still matters:

  • headline — Article title
  • image — Three aspect ratios (1:1, 4:3, 16:9)
  • datePublished and dateModified — ISO 8601 with timezone
  • author — Person schema with name and url
  • publisher — Organization with logo

Article schema unlocks author biopics, publication dates, and article carousels in search. For blog content, BlogPosting is semantically most accurate. NewsArticle fits time-sensitive news.

Product Schema

Product schema is highest-value for eCommerce. Fully implemented with Offer and AggregateRating, it generates rich results showing price, availability, and star ratings.

Required fields for Product rich results:

FieldRequirement
nameRequired
imageRequired
offers.priceRequired
offers.priceCurrencyRequired
offers.availabilityRequired
aggregateRatingRequired for stars

Critical detail: availability must use the full Schema.org URL (https://schema.org/InStock), not a plain string. Same for itemCondition.

BreadcrumbList is highest-ROI for content sites. It transforms raw URLs into human-readable path hierarchies: Home › Blog › SEO Guide.

Two requirements:

  1. Minimum two ListItem objects
  2. Each needs position (integer) and name

The last ListItem should omit the item URL. Google recommends this.

Organization Schema

Organization schema on your homepage helps Google understand your brand and can contribute to sitelinks. But the real value is entity disambiguation.

The sameAs property is critical. Link to:

  • Wikidata (most powerful for Knowledge Graph)
  • LinkedIn company page
  • Crunchbase profile
  • Official social profiles

Each SameAs identifier is an external authoritative source Google cross-references to verify your entity. More identifiers = higher entity confidence score.

LocalBusiness Schema

For location-dependent businesses, LocalBusiness schema is non-negotiable. It provides Google your address, phone, hours, and service area—critical for map pack and local AI answers.

Schema.org v30.0 (released March 19, 2026) added floorLevel for multi-floor businesses. If you have one, add it.

VideoObject Schema

Video content with VideoObject schema remains active. Required fields:

  • name
  • thumbnailUrl
  • uploadDate

Important: use interactionStatistic for view counts—not interactionCount, which is deprecated as of February 2026.

Schema Types to FIX in 2026

These still work, but implementations are commonly broken.

Fix Mismatched Schema and Content

Google now classifies mismatched schema—applying a type that doesn’t match primary content—as misleading markup. Examples:

  • Marking a landing page as Article when it’s a sales page
  • Adding Product schema to pages without purchasable products
  • Wrapping editorial ratings in Review schema

The fix: audit every page with schema. The markup must describe the primary content purpose.

Fix Stale AggregateRating Data

Inflated or unverifiable aggregate ratings face enhanced scrutiny. Practical threshold: minimum 5 genuine reviews with accessible sources.

If your reviews come from closed systems Google can’t verify, remove the aggregateRating until you have verifiable data.

Fix Missing Date Fields

Article schema with missing datePublished won’t display author attribution or freshness signals. Update dateModified only for substantive changes—artificially inflating modification dates triggers manual review flags.

Fix Video View Counts

If you’re still using deprecated interactionCount on VideoObject, switch to interactionStatistic:

"interactionStatistic": {
  "@type": "InteractionCounter",
  "interactionType": { "@type": "WatchAction" },
  "userInteractionCount": 0
}

Schema Types to REMOVE in 2026

Some schema types are retired. Remove or stop implementing these.

Remove FAQPage Schema (Don’t Rush)

FAQPage rich results are dead as of May 7, 2026. But here’s what most guides get wrong: the schema type itself isn’t deprecated. FAQPage markup is still valid at schema.org—it just produces zero SERP lift in Google.

Google’s exact words: “Structured data that’s not being used does not cause problems for Search.”

What this means practically:

  • Don’t rush to remove existing FAQPage markup—it’s harmless
  • Stop adding it to new pages for SEO purposes
  • Keep it on genuinely Q&A-shaped content (support docs, help articles) if the markup accurately describes the page

The danger: treating FAQ schema as an AI search shortcut. Templated FAQ markup across 168,000 articles—much of it not describing genuine content—is likely what killed the feature.

Remove HowTo Schema

HowTo rich results were retired September 2023. No surface, desktop or mobile. The schema is valid but produces nothing.

Stop implementing HowTo schema entirely. Focus on Article and VideoObject for instructional content.

Remove Deprecated Types

These seven types lost rich result support. Remove them:

  • Book Actions
  • Course Info
  • Claim Review
  • Estimated Salary
  • Learning Video
  • Special Announcement
  • Vehicle Listing

Removing them doesn’t hurt rankings—Google ignores unused schema. But cleaning them reduces code bloat and future audit confusion.

The AI Search Shift

Here’s what changed that most SEOs are missing: AI Mode uses schema differently than traditional search.

AI Mode—powered by Gemini—reads structured data as a trust signal and entity verification input, not a display trigger. The distinction matters:

Traditional SearchAI Mode
Schema triggers visual SERP featuresSchema verifies claims and entity identity
Rich result = direct CTR liftSchema = citation probability increase
Display eligibility is binaryTrust scoring is continuous

Pages with clean entity schema—Organization with SameAs, Person with knowsAbout—see measurably improved citation rates in AI Mode. This isn’t a Google-confirmed ranking factor, but observational studies show 2.5x-3.4x higher AI citation rates for pages with comprehensive structured data.

The strategic implication: entity disambiguation schema is now the highest-leverage implementation available. SameAs identifiers, knowsAbout properties, and Organization-Person relationships help AI systems resolve who published the content and why they’re credible.

Your 2026 Schema Implementation Priority

If you’re starting from scratch or rebuilding after audits, here’s the priority order:

  1. BreadcrumbList — Every interior page. Simple, high CTR impact.
  2. Organization — Homepage only. Brand signals, sitelinks support.
  3. Article/BlogPosting — All editorial content. Author authority, freshness.
  4. Product — eCommerce product pages. Price and rating rich results.
  5. LocalBusiness — Physical locations. Map pack, knowledge panel.
  6. VideoObject — Video content. Discovery and Key Moments.

Validation: Don’t Skip This Step

Schema implementation without validation is guesswork. Use both tools:

ToolPurpose
Google Rich Results TestWhich rich results your page qualifies for
Schema.org ValidatorVocabulary completeness, property validity

Before deploying: validate JSON syntax, run Rich Results Test on staging, confirm eligibility.

After deploying: monitor Search Console Enhancements weekly for the first month. Catch regressions before they compound.

Quick Reference: 2026 Schema Status

Schema TypeRich Result StatusAction
Article/BlogPostingActiveKeep, implement fully
Product + OfferActiveKeep, complete implementation
BreadcrumbListActiveKeep, every page
OrganizationActive (no SERP display)Keep, add SameAs
LocalBusinessActiveKeep if applicable
VideoObjectActiveKeep, fix interactionStatistic
FAQPageRetired May 7, 2026Stop adding, keep existing
HowToRetired Sep 2023Remove from priorities
Book Actions, Course Info, etc.RetiredRemove or ignore

Sources

schema markup strategy 2026 schema SEO structured data strategy schema fix SEO remove schema SEO
WORK WITH US

Ready to scale your B2B SaaS?

Build a growth engine that delivers qualified demos, pipeline, and predictable revenue.

BOOK A STRATEGY CALL