Is Structured Data Still Worth It After Google Retired FAQ Rich Results?
Summary: Yes—but the rationale has shifted: from competing for a spot on the search results page to telling machines "who you are, when this was written, and why you should be trusted." Google retired FAQ rich results on 2026-5-7, but FAQPage itself is still parsed; Article, Product, Review and other types keep producing rich results.
"Now that FAQ rich results are gone, do we still need structured data?" is one of the most asked technical questions of 2026. The answer depends on what you treat structured data as: if you see it as a way to grab a display position on the search results page, its payoff is indeed shrinking; if you see it as a fact sheet for machines, its weight is actually rising. This article lays out exactly what Google retired this time, whether FAQPage still has value, which types still produce visible results, and what to do next.
Key Takeaways
- Worth doing, but the reason changed: structured data stopped being a way to win a display slot on the results page and became a fact sheet that tells machines who you are, when you wrote it, and why it should be trusted.
- What disappeared is the display style, not the parsing: Google retired the FAQ rich result on 2026-5-7, but FAQPage markup itself is still valid and parseable, and carries no penalty.
- The timeline matters: Search Console's FAQ report and rich-result test support were removed in 2026-6, and related API support in 2026-8 — dashboards and BI connectors must be reworked before then.
- For most commercial sites the impact is close to zero: since 2023-8 FAQ rich results have been broadly limited to authoritative government and health sites, so ordinary commercial sites stopped receiving that display long ago.
- Types still producing rich results include Article, Product, Review and AggregateRating, Video, Event, Recipe, JobPosting, Course, LocalBusiness, and BreadcrumbList; HowTo is effectively retired.
- Organization and Article are the highest-value pair in the AI era: the former supports entity consolidation, the latter states freshness and authorship through dateModified and author, which AI systems commonly use as evidence when screening sources.
- FAQPage should be kept, not deleted: Google has stated plainly that unused structured data causes no problem, and SearchPilot's controlled test showed no traffic gain from removing FAQ markup.
- The biggest misunderstanding is treating markup as a shortcut: Google's AI features documentation states that AI Overviews and AI Mode require no special structured data, and requires markup to match visible page content.
1. The Verdict First: Still Worth It, but the Goal Shifts from "Display" to "Being Understood"
Structured data is still worth investing in during 2026, for three reasons.
First, what disappeared was the display style, not the parsing ability. Google removed the expandable FAQ rich result from the search results page, but the FAQPage markup itself remains valid, is still parsed, and carries no penalty whatsoever. Second, a large set of types still produces rich results. Article, Product, Review, Video, Event, Recipe, LocalBusiness and others still deliver enhanced presentation in search, while Organization and BreadcrumbList continue to support brand-entity and breadcrumb-path signals. Third, machine demand for fact sheets is rising. When an AI system judges "whose claim to trust," it pulls evidence from fields like dateModified, author, Organization and sameAs—and those are exactly the job of structured data.
In other words, structured data's role has changed from "billboard on the search results page" to "barcode on the product"—it no longer grabs attention, it conveys facts.
2. What Was Actually Removed, and What Wasn't
This change is precise, and drawing the boundary correctly matters; otherwise it is easy to make the mistake of over-deleting.
| Item | Status | Note |
|---|---|---|
| FAQ rich result expandable display on SERP | Removed (2026-05-07) | SERP no longer expands Q&A format |
| FAQPage markup itself | Still valid and parsed | No penalty; still readable by AI and crawlers |
| Search Console FAQ report | Removed from 2026-06 | Related API extracts start returning empty |
| Rich Results Test support for FAQ | Removed from 2026-06 | Use a general Schema validator instead |
| Search Console API FAQ data | Removed from 2026-08 | Dashboards and connectors must update before then |
| Article / QAPage / DiscussionForum | Google's suggested alternatives | Cover similar Q&A content scenarios |
| HowTo rich result | Phased out since 2023 | Limited practical value today |
One easily overlooked operational item: if you have dashboards or automation that pull FAQ rich-result data, you must update those calls before 2026-8, or they will silently return empty; if you need historical data for year-over-year comparison, export it before 2026-6.
Worth emphasizing: for most commercial sites, the real impact of this retirement is close to zero. Since 2023-8, FAQ rich results have been largely limited to authoritative government and health sites, and ordinary commercial sites had long stopped receiving that display. Claims of "huge impact" mostly conflate this with the actual structural shift in traffic.
3. Does the Retired Markup Still Have Value? Yes—but Not That Value
FAQPage markup has not lost all meaning; it has only lost the "guaranteed expandable display" aspect.
First, it is still a readable fact. Crawlers and AI systems can read the JSON-LD on the page, and a clear Q&A structure does lower machine-parsing cost. Second, it is a byproduct of content quality. If the page genuinely contains both question and answer and the answer stands on its own, marking it simply describes the page truthfully; if the page content is thin, no amount of markup will earn a citation. Third, Google has explicitly stated that not using it does no harm. Official guidance says unused structured data causes no problems; SearchPilot's A/B test also showed removing FAQ markup brought no traffic lift.
So the correct move is: keep existing markup unless it is fabricated, keyword-stuffed, or added purely to chase the old display format. Removing it just to remove it is wasted effort. What truly deserves vigilance is something else—Google's AI Features documentation explicitly states that AI Overviews and AI Mode need no special structured data, and requires structured data to match the page's visible content. Treating markup as a shortcut into AI answers is one of the most widespread and most harmful misconceptions today.
4. Types That Still Produce Visible Results in 2026
Ranked by return on effort, the table below lists the types that remain effective.
| Type | Status | Effect produced |
|---|---|---|
| Article | Active | Enhanced headline display; supports Top Stories and similar |
| Product | Active | Price, stock, rating shown directly; supports AI shopping |
| Review / AggregateRating | Active | Star ratings appear in search results |
| Video | Active | Video carousel and thumbnail enhancement |
| Event | Active | Structured display of event time and location |
| Recipe / JobPosting / Course | Active | Their respective vertical cards |
| LocalBusiness | Active | Knowledge-panel data source; essential for local scenarios |
| BreadcrumbList | Active | Breadcrumb path in search results |
| Organization | No direct display | But supports brand-entity recognition—the key foundation in the AI era |
| HowTo | Largely retired | Limited practical value; not recommended |
Among these, the Organization + Article combination has the highest value in the AI era: the former lets machines be certain "which entity this content comes from," while the latter conveys timeliness and authorship through complete headline, datePublished, dateModified and author fields. These two are exactly the common evidence AI systems use when filtering citation sources.
Pay attention to field completeness. Most sites' Article markup has only headline and image; author, datePublished and dateModified are often missing, yet the latter two are the very source of timeliness signals. After updating content, syncing dateModified and ensuring it matches the visible update time on the page is a low-cost, clearly worthwhile step.
5. Three Common Misjudgments
The first misjudgment is "FAQ rich results are gone, so the FAQ content must go too." The opposite is true: Q&A content that resolves real doubts is still effective for conversion; it exists because human readers will read it, not for some search-result field.
The second misjudgment is "add structured data and you'll be cited by AI." Google's official position is that it is not required; direct crawl tests by multiple practitioners also show that if the page body lacks the corresponding content, information hidden in JSON-LD will not be extracted. Claims of citation lift mostly come from tests whose methodology is undisclosed and should not be used as a decision basis.
The third misjudgment is "retirement equals penalty." This confuses the nature of the signal—Google has never said pages with FAQPage markup will be demoted; in fact it has explicitly said non-use has no effect either. What truly affects traffic is the structural shift of search behavior toward AI answers, and the two must not be conflated.
6. What to Do Next: Ranked by Priority
First, fix the data pipeline. Audit every Search Console API call, dashboard and BI connector that depends on FAQ rich-result data, and complete the rework before 2026-8; export any historical data for the record before 2026-6.
Second, keep existing markup but run one honesty screening. Delete markup for which the page has no corresponding content and that was added purely to chase display; keep the parts that truthfully describe the page.
Third, shift the focus of structured data to Article and Organization. Fill in author, dateModified and sameAs external-profile links for the same entity so machines can complete entity consolidation—this is the highest-ROI part in the AI era.
Fourth, if you truly need to replace hand-written Q&A content, consider QAPage or DiscussionForum. The former fits a single Q&A page, the latter fits forum-style discussion content; ordinary articles can just use Article without forcing a Q&A type.
Fifth, establish a periodic verification mechanism. Template changes, plugin upgrades and CMS field failures can all silently break markup. We recommend spot-checking one page with a rich-results testing tool after each release, and auditing a competitor's entity structure during quarterly reviews.
FAQ
Q: What exactly did Google retire?
A: It retired the dropdown display of the FAQ rich result on the results page (from 2026-5-7). The FAQPage type itself is still valid and parseable, and carries no penalty. Three follow-on dates: Search Console’s FAQ report and rich-result test support removed in 2026-6, and API support removed in 2026-8.
Q: How much will this hurt our traffic?
A: For most commercial sites the impact is close to zero. Since 2023-8 FAQ rich results have been broadly limited to authoritative government and health sites, so ordinary commercial sites stopped receiving that display long ago. What actually moves traffic is the structural shift of user search behavior toward AI answers; the two should not be conflated.
Q: Should I delete the FAQ markup on my pages?
A: Keep it. Google has stated that unused structured data causes no problem, and SearchPilot’s controlled test showed no traffic gain from removing FAQ markup. The case for deletion is different: markup stacked purely to chase the old display format with no corresponding content on the page. Also, if an answer stands on its own and is useful to users, keeping it still has value.
Q: If I add structured data, will AI cite my content?
A: No — do not treat it as a shortcut. Google’s AI features documentation states that AI Overviews and AI Mode require no special structured data, and requires structured data to match visible page content. Practitioner crawling tests also show that when the page body has no matching content, information hidden in JSON-LD is not extracted. What actually works is extractability: self-contained answer passages, data with dates and sources, and a clear heading structure.
Q: Which structured data types are still worth doing in 2026?
A: Prioritize Article, Product, Review and AggregateRating, Video, Event, LocalBusiness, and BreadcrumbList; the vertical types Recipe, JobPosting, and Course also still work. Organization produces no visible display directly, but it underpins brand entity recognition and knowledge consolidation, making it critical infrastructure in the AI era. HowTo is effectively retired and not worth further investment.
Q: Why is the Organization plus Article combination the most important?
A: Because together they supply the two kinds of evidence AI systems reach for most when screening citation sources. Organization with sameAs lets machines consolidate brand mentions scattered across the web into one entity, avoiding a name mismatch splitting it into several weak entities; Article states freshness and authorship through complete headline, datePublished, dateModified, and author fields. The former answers who wrote it, the latter how recent it is and who is accountable.
Q: What type should replace FAQPage content?
A: It depends on the content shape. For an ordinary article, Article is enough — there is no need to force a Q&A type. If the page is genuinely a single question and answer, use QAPage; for forum or community discussion content, use DiscussionForum, which is also the direction Google’s documentation suggests.
Q: What order should we work through next?
A: Five items by priority: first, before 2026-8 rework every API call, dashboard, and BI connector that depends on FAQ rich result data, exporting anything needed for year-over-year comparison before 2026-6; second, keep existing markup and run one honesty audit, deleting stacked markup with no matching content; third, shift focus to Article and Organization and complete sameAs; fourth, where Q&A content genuinely needs replacing, switch to QAPage or DiscussionForum; fifth, set up a sampling check after every release.
About Tengence
The structured annotation of topics and entities is one component of the content go-live stage of Tengence GEO; its dashboard covers the Article / Organization field completeness and citation-observation comparison mentioned in this article. Tengence (TENGENCE) is the enterprise growth engine for the AI era, focused on public-traffic acquisition and private-domain conversion; its core products are Tengence GEO (dual-engine GEO + SEO) and Tengence Search.
Related Reading
- What Is GEO (Generative Engine Optimization)? Definition, Principles and How It Differs from SEO
- Can't Get Your Brand Found by AI? A Self-Check List for Brand AI Visibility
- 2026 GEO Trends: 7 Shifts in AI Search and Enterprise Response Priorities
Data Sources
- Google Search Structured Data Documentation (2026 update): FAQ rich results stopped displaying on 2026-5-7; the Search Console FAQ report and rich-results test support were removed in 2026-6, and related API support in 2026-8; Google recommends describing similar content with Article, QAPage or DiscussionForum.
- Google Search AI Features Official Documentation: AI Overviews and AI Mode require no special structured data; structured data must match the page's visible content.
- Google Search Central Blog (2023-08-08): FAQ rich-result display scope narrowed to authoritative government and health sites; HowTo rich results stopped desktop display from 2023-09-13.
- SearchPilot Structured Data A/B Test: Removing FAQ markup from pages showed no traffic lift; consistent with Google's statement that "unused structured data causes no problems."
Get Started
Free trial of Tengence
Book an expert consultation
Learn about Tengence GEO
