To test structured data on your own website, start with Schema Markup Validator to check the markup, then use Google’s Rich Results Test to see which rich results the markup might produce. After deployment, also confirm that Google can access and render the pages. Look beyond whether the markup passes: check which markup types the tool detects, which fields have issues, and whether your changes resolve the errors. Passing validation does not guarantee that your pages will appear as rich results.
Are you testing a draft or a published page?
Before deployment, you can test the code directly. Once a page is live, you can enter its URL to check the markup on the page. Semrush’s instructions introduce the following tools, both of which can test code or URLs:
- Schema Markup Validator: Check a URL or code snippet, then use the issues flagged by the tool to locate what needs fixing.
- Rich Results Test: Validate published or draft markup and see which rich results the existing markup might produce.
If you use a WordPress plugin to generate markup, do not skip testing. Plugins support different markup types, so first confirm that the plugin meets the page’s needs, then check the resulting configuration. Installing a plugin does not mean the markup has passed validation.
Understand errors and warnings, not just the green checkmark
When reviewing results, first verify the markup type, then look for missing information or formatting issues. Different types require different content, so you cannot apply the same field checklist to every page. The guidance for Schema Markup Validator describes zero errors and zero warnings as the ideal result. If issues remain, fix them and test again.
In its 2019 analysis of car insurance search results, Screaming Frog found that pages using Product markup were missing images required at the time, along with several recommended properties. In that example, missing required fields were classified as errors, while missing recommended fields were classified as warnings.
The same analysis also noted that another page used UK for addressCountry rather than GB, as required by that check. Google’s testing tool at the time did not detect the issue. This is a historical example, not a list of current field requirements. It reminds us that even if one tool reports no errors, you still need to check each item against the applicable requirements.
After deployment, check access and rendering
Google’s documentation for carousel rich results (beta) instructs users to deploy a few pages, use the URL Inspection tool to check rendering, and check whether robots.txt, noindex, or login requirements prevent access or indexing. This guidance applies to a specific feature; its other eligibility requirements should not be extended to all structured data.
You can also monitor the relevant reports in Search Console. In 2023, awoo described how to check markup types and the causes of errors under “Shopping” and “Enhancements.” That description reflects the interface at the time; the names and locations may differ today. After fixing the reported issues, test the updated pages again to confirm that the problems are resolved.
Structured data testing: perspectives from across the web
How can you identify markup that needs checking when your website has many pages?
You can use Screaming Frog's Structured Data tab to filter and sort pages, identify URLs with missing or incorrectly implemented markup, and then consult the detailed reports to see which elements or properties need adjustment. When making fixes, refer to the schema.org documentation and the page's existing content. Do not enter information that does not match the content merely to clear a warning.
How can you test whether AI citations change after adding structured data?
Ahrefs recommends selecting 5 to 10 test pages and an equal number of control pages with similar citation counts. Record baseline citation counts for both groups in AI Overview, AI Mode, and ChatGPT. Then add markup only to the test pages and record the date. Make no further changes to those pages during the test, and compare changes in the two groups after at least 30 days. Ideally, the test pages should already have some citations; otherwise, it is difficult to determine why they have none.
If AI citations to the test pages increase, can you conclude that structured data caused the increase?
Not directly. If the test and control groups show similar changes, this is more likely to reflect platform-wide trends. If the test group performs better, that is only an indication that the markup may have had a positive effect, not sufficient evidence on its own to establish causation. Ahrefs' research also observed a decline in test pages relative to control pages, but the cause was unclear.
References
- 輪轉介面 (Beta 版) 結構化資料 | Google 搜尋中心 | Documentation | Google for Developersdevelopers.google.com — Source used in the article above.
- Screaming Frog SEO Spider Update – Version 11.0www.screamingfrog.co.uk · 2019-03-05T09:52:50+00:00 — Source used in the article above.
- SEO必學:Google結構化資料標記(Schema)是什麼?www.yesharris.com — Source used in the article above.
- 五分鐘了解結構化資料是什麼,與搜尋引擎進行有效溝通!www.awoo.ai · 2023-09-13T07:15:35+00:00 — Source used in the article above.
- Organization Schema: What It Is & How to Implement Itwww.semrush.com · 2017-07-26T08:00:00Z — Source used in the article above.
- We Tracked 1,885 Pages Adding Schema. AI Citations Barely Moved.ahrefs.com · 2026-05-11T16:36:19+00:00 — Discusses: How can you test whether AI citations change after adding structured data?; If AI citations to the test pages increase, can you conclude that structured data caused the increase?
- Mastering SEO Audits: Unlocking Insights with the Screaming Frog SEO Spiderwww.screamingfrog.co.uk · 2023-10-18T10:42:49+00:00 — Discusses: How can you identify markup that needs checking when your website has many pages?

