8 min read
Rich results disappeared from my pages. What changed?
By CrawlWarden, which builds the monitoring described at the foot of this article.
Published , sources last re-read .
Rich results disappear for one of four things: Google retired the feature (FAQ rich results stopped appearing in Google Search on 7 May 2026), the markup stopped parsing, the page left the index, or Google chose not to show an eligible result. Two of the four start on your own site. CrawlWarden reads the JSON-LD on each watched page, records its set of @type values, and reports when that set changes or empties, once four consecutive runs show the same new value.
What CrawlWarden records, exactly
The check is called "Structured data present, and its @type". On every run it reads each <script type="application/ld+json"> block in the HTML the server returns, collects the @type of every top level object and of every object inside @graph, removes duplicates, sorts them, and stores the result as one value. A page with no block records "no structured data". A block that is not valid JSON is recorded as invalid-json-ld, so a broken block is a change in the value rather than a silent gap.
This page is the worked example. The build of 11 October 2026 produced one JSON-LD block for it, and the value the check records from that block is Article,BreadcrumbList,FAQPage. Every row below describes what happens to a value like that one.
| What happens on the page | What the check records | Incident? |
|---|---|---|
| A template edit drops the JSON-LD block | "no structured data" replaces the type list | Yes, after four consecutive runs |
A plugin update renames the type, say Article to BlogPosting | The new sorted list | Yes, after four consecutive runs |
| A deploy leaves a trailing comma inside the block | invalid-json-ld appears in the list | Yes, after four consecutive runs |
You remove FAQPage on purpose | Article,BreadcrumbList replaces Article,BreadcrumbList,FAQPage | Yes. The check reports changes, not mistakes, so a deliberate edit is an expected incident |
| A required property goes empty and the type stays | Nothing new | No. That is a job for the Rich Results Test |
A nested type changes, such as an author Organization becoming a Person | Nothing new. Only top level and @graph types are read | No |
| A tag manager or script adds the markup after the page loads | Nothing. The check reads the served HTML and runs no JavaScript | No |
| The markup is Microdata or RDFa | Nothing. Only JSON-LD is read, although Google supports all three formats | No |
Four runs, in hours
Four is the confirmation count set for this check in CrawlWarden's check catalogue, the same as for Open Graph tags and the meta description. The indexability checks, such as the HTTP status code and the robots meta tag, speak after two. A value that flips back before its fourth sighting never reaches an inbox.
The fourth sighting comes three check intervals after the first. At the three hour cadence the Agency and Portfolio plans allow, that is about nine hours. At six hours, the fastest Solo allows, about eighteen. On Free, checked once a day, about three days.
Google's side runs on a different clock. Search Console says "Only indexed pages appear in Search Console reports", that "The report only shows a sample of your pages", and that "Newly implemented structured data might not appear immediately, as Google requires time to crawl updated pages" (Debug missing structured data). The CrawlWarden record is dated from your HTML, on every watched URL, whether or not Google has processed the page yet.
This site still emits FAQPage, on purpose
Of the 45 HTML files the 11 October 2026 build produced, 25 carry a FAQPage type: every article, every comparison and the pricing page. Each FAQ block mirrors the questions printed on the same page, which keeps it inside Google's rule that structured data "must be a true representation of the page content" (structured data general guidelines).
In Google Search that markup now produces nothing. Google's changelog records that the FAQ rich result "will no longer appear in Google Search starting May 7, 2026", and on 15 June 2026 that the documentation was removed (documentation updates). The search gallery, last updated the same day, no longer lists FAQ or HowTo (search gallery).
Google's published position on unused markup comes from its 2023 post that first limited FAQ rich results: "there's no need to proactively remove it", and structured data that is not used "does not cause problems for Search, but also has no visible effects in Google Search" (Changes to HowTo and FAQ rich results). CrawlWarden keeps the blocks for search engines and assistants other than Google that may read them. Whether any of them does is not something this project has measured, and the sources below say so.
If the blocks ever go, CrawlWarden's own check would open the incident in the table above on all 25 pages, four runs later. That is the correct outcome: the @type set changed, and the check reports what changed, not whether the change was wanted.
Telling the four apart, in order
Work down the table before editing any markup. Three of the five rows are not changes to your schema, and two of them are not problems at all.
| What you see | What it points to | Where Google shows it | What CrawlWarden has |
|---|---|---|---|
| One feature vanished on every site you manage, on one date | Google retired the feature | The Search Central documentation updates page | Nothing new. Your HTML did not change, so the @type value did not either |
| One template lost it, every page of that template at once | The markup changed or stopped parsing | Unparsable structured data report; Rich Results Test | A dated @type change, or invalid-json-ld, on each watched URL of that template |
| Items fell in step with indexed pages | The pages left the index | Page indexing report; URL Inspection | Google index status, read once a day, and the HTTP status code |
| It comes and goes, on some queries and not others | Google chose not to show an eligible result | Nowhere. The Rich Results Test can still pass | Nothing, and nothing to record |
| The report count moved slightly with no other signal | The report is a sample | Nowhere | Nothing |
What Google documents about each
Retired features. FAQ is the current case, with the dates above. HowTo went first: Google cut it to desktop in August 2023 and stopped showing it on desktop on 13 September 2023, then said it would drop the HowTo search appearance, report and Rich Results Test support in 30 days and Search Console API support in 180 days (Changes to HowTo and FAQ rich results). A rich result is a Google feature, not a property of your page.
Markup that stopped parsing. Google: "Invalid or unparsable structured data on previously valid pages can cause items to disappear from reports" (Debug missing structured data). The separate Unparsable structured data report lists markup "that could not be parsed because of a serious syntax error", where "The intended type of structured data (Job, Event, and so on) could not be determined" (Unparsable structured data report). A JSON syntax error is breakage that CrawlWarden's check and that report both see. A missing required property shows only in Google's tools.
Pages that left the index. Only indexed pages appear in the reports, and Google also lists login walls, robots.txt rules and noindex as things that keep it from seeing the markup. If the items fell on the same date as the indexed page count, the page leaving Google is the incident and the schema went with it.
Eligibility. Google: "Google does not guarantee that your structured data will show up in search results" (structured data general guidelines). A competitor showing the enhancement you used to get is not, on its own, evidence that your markup broke.
The checks that report this
CrawlWarden watches a list of URLs you choose and reports what changed. It does not diagnose and it does not score. Severity is a property of the check rather than a setting: the number below is how many consecutive runs must see the new value before it becomes an email, and it is never less than two, because one observation is noise.
Structured data present, and its @typeconfirmed over 4 consecutive runs
Reads every JSON-LD block in the served HTML and records the sorted set of top level and @graph @type values, or invalid-json-ld for a block that does not parse. Reports when that set changes or empties, once four consecutive runs agree. It does not validate properties and does not run JavaScript.
Google index statusconfirmed over 2 consecutive runs
Google's own index status for the page, from URL Inspection once a day. A page that leaves the index takes its rich results with it.
HTTP status codeconfirmed over 2 consecutive runs
The page stopped returning the status it used to. Watched beside Google index status, so both dates sit in one history.
Every check is available on every plan, including the free one. CrawlWarden pricing, per site rather than per URL.
Questions
- Is there any point keeping FAQPage markup now that Google shows nothing for it?
- In Google Search, no. Google's 2023 post said there is "no need to proactively remove it" and that unused structured data has "no visible effects in Google Search". This site keeps it for other engines and assistants that may read it, which it has not measured. Adding new FAQPage markup in the hope that Google brings the feature back has no basis in anything Google has published.
- The Rich Results Test passes but no rich result appears. Is the test wrong?
- Not necessarily. Google states that it "does not guarantee that your structured data will show up in search results", so valid markup with no rich result is consistent with Google's own policy.
- How soon does Search Console reflect a change to the markup?
- Not immediately. Google says new structured data "might not appear immediately, as Google requires time to crawl updated pages", and the report covers only indexed pages and a sample of them. CrawlWarden's record is dated from your HTML instead, about nine hours to three days after the change depending on the check cadence.
- Will CrawlWarden validate my structured data?
- No. It reports that the structured data on a page disappeared, stopped parsing as JSON, or changed its set of
@typevalues, confirmed over four consecutive runs. It does not tell you which property is wrong. The Rich Results Test does that, and it is free. - Will CrawlWarden see schema that my tag manager adds?
- No. The check reads the HTML the server returns and runs no JavaScript, so markup injected after the page loads is invisible to it. Such a page records "no structured data" on every run, which never changes, so it never alerts. If that is your setup, the check has nothing to watch on those pages.
Related
Sources
What was checked, and where. An entry with no link is something that could not be verified, named here rather than left out.
Google Search Central: latest documentation updates
The 8 May 2026 deprecation entry ("will no longer appear in Google Search starting May 7, 2026") and the 15 June 2026 entry removing the FAQ documentation. Re-read 2026-10-11.
That unused FAQ markup need not be removed and has no visible effect in Google Search, and the HowTo timeline (desktop only, then gone on 13 September 2023, reports in 30 days, API in 180). Re-read 2026-10-11.
Google Search Console Help: Debug missing structured data
That unparsable markup on previously valid pages removes items, that only indexed pages are reported, that the report is a sample, the crawl delay, and the access issues list. Re-read 2026-10-11.
Google Search Console Help: Unparsable structured data report
That a separate report lists markup that could not be parsed, whose intended type could not be determined. Re-read 2026-10-11.
Google Search Central: General structured data guidelines (updated 2026-07-10)
The three supported formats, the true representation rule, and that Google does not guarantee structured data will show. Re-read 2026-10-11.
Google Search Central: Structured data markup that Google Search supports (updated 2026-06-15)
That the gallery lists 25 features and neither FAQ nor HowTo. Re-read 2026-10-11.
Not verified: the dates Search Console reporting, the Rich Results Test and API support for FAQ ended
The FAQ documentation page that carried Google's notice now redirects to the changelog (checked 2026-10-11), and the changelog gives no date for these. The only date stated as fact here is 7 May 2026, for the feature leaving Search.
Not verified: whether other search engines or assistants use FAQPage markup
This is the ground for keeping the markup on this site, and it has not been measured. No vendor documentation was checked for this page.
Not verified: search volume for this query
No keyword volume tool was available. Demand was judged from the live US SERP on 2026-10-04, which carried no ads and ranked two Google Help threads, Reddit and Stack Overflow alongside two small sites.
Know the moment this happens to you
Give CrawlWarden a domain, keep the pages it suggests, and the first check runs immediately. Free forever for one site.