When auditing structured data, first check the markup types and validity on important pages. Then list the affected pages, fields, and causes of errors to identify what needs fixing. Set priorities according to the audit’s goals and scope: the tool’s errors, warnings, and notices are a starting point, not a rigid order to follow. Recheck after making fixes and schedule regular audits to track whether issues have been resolved and whether new ones have appeared.
First, identify which pages and fields have issues
Check not only whether markup is present, but also whether it is valid. Important pages should also provide additional information, such as star ratings, to give search engines the context they need to display rich results. Completing these checks does not guarantee that a page will appear as a rich result.
Semrush’s 2023 article describes using Google’s Rich Results Test to check markup types and validity for a single URL. For larger sets of pages, it describes using a site audit tool to gather issues in one place. The workflow starts with viewing invalid items in the markup report, then expanding them to see the affected pages and fields. These interface names come from the article and may differ from those in the current version.
- Record the affected page URLs rather than keeping only a sitewide error count.
- Record the specific fields and issue descriptions so it is clear what needs fixing.
- Collect the results in a tracking sheet to make it easier to prioritize fixes and compare results afterward.
Start with severity, then check against the audit’s goals
The same Semrush article emphasizes that priorities depend on the audit’s goals and scope. It uses the tool’s three-level classification as a quick starting point:
- Errors: The most severe issues; address these first.
- Warnings: Less urgent; address these after resolving errors.
- Notices: Lower priority; address these when time allows.
This classification belongs to the tool; it is not a universal rule from Google for prioritizing fixes. For example, if the audit focuses on structured data on important pages, first list the issues within that scope rather than combining every technical issue across the site into a single list.
For syntax errors, awoo’s 2023 article gives examples involving Search Console’s “Unparsable structured data” report. Finding missing commas, brackets, or quotation marks may require checking the code section by section. The article suggests using ChatGPT to help locate and fix issues. This is the author’s suggestion for working more efficiently, not an official validation procedure. A revision generated by ChatGPT should not be treated as having passed validation.
Recheck after fixes and plan for ongoing monitoring
After making fixes, check whether the issues have been resolved. Semrush’s 2023 article mentions scheduling automated audits to track progress after changes. Its 2024 article on review structured data also notes that requirements may change, so it recommends regularly checking the implementation and scheduling technical SEO audits.
To track progress, compare new reports with the original list to see whether invalid items remain, then adjust the priorities of outstanding tasks. These articles do not specify an audit frequency suitable for all websites, so weekly or monthly audits should not be treated as mandatory.
Structured data audits and fix prioritization: perspectives from across the web
After a structured data guideline violation, will fixing the issues automatically lift a manual action?
You should not assume that fixing the issues will automatically lift the manual action. Google’s guidelines for local business rich results state that violations may lead to a manual action. After fixing the relevant issues, a site can submit a reconsideration request. The guidelines do not say that the action will be lifted automatically or guarantee that a reconsideration request will be approved.
Does valid structured data improve visibility in large language models?
There is no guarantee. Semrush states that markup may help AI systems understand content and may affect visibility in large language models. This claim carries uncertainty: it does not establish that valid markup will necessarily increase visibility, nor does it provide a basis for determining specific results.
References
- 當地商家 (LocalBusiness) 結構化資料 | Google 搜尋中心 | Documentation | Google for Developersdevelopers.google.com — Discusses: After a structured data guideline violation, will fixing the issues automatically lift a manual action?
- How to do a website audit in 2026 (+ free tracker)www.semrush.com · 2023-12-28T11:37:00Z — Discusses: Does valid structured data improve visibility in large language models?
- ChatGPT x SEO 13 個應用案例大公開,用 AI 輕鬆提升網站 SEO 執行效率!www.awoo.ai · 2023-03-14T06:49:06+00:00 — Source used in the article above.
- What Is Review Schema? & How to Implement Itwww.semrush.com · 2024-01-19T16:19:00Z — Source used in the article above.

