How Should IR Teams Use Earnings Release Structured Data for AI Search?

Earnings release structured data for AI search is machine-readable Article markup that identifies an investor relations results page. Schema.org, Google Search Central, ChatGPT, Google AI Overviews, Google AI Mode, Microsoft Copilot, and SEC EDGAR form parts of the publishing and verification path. The markup clarifies page identity, but crawlable HTML and dated checks remain necessary.

TL;DR

• Use Schema.org Article as the primary type for each earnings release page.
• Match the markup to the visible headline, dates, author, publisher, and canonical URL.
• Use HTML links with an href attribute so crawlers can reach the release.
• Add the canonical release URL to the XML sitemap.
• Test the live HTML, structured data, crawl status, canonical URL, and filing figures.
• Structured data does not guarantee that an AI engine will select or cite the page.

What structured data should an earnings release use?

An earnings release should use Article markup that describes one visible release page. An IR team may also use FAQPage when readers can see the same questions and answers.

Which Schema.org type fits an earnings release?

Schema.org Article is a general type that can describe an earnings release page. Schema.org defines Article as an article, such as a news article or investigative report. Schema.org also lists more specific types, including NewsArticle and Report.

The selected type should describe the visible content. It should not imply that the page contains material that readers cannot see. Schema.org definitions were last checked on 2026-10-24.

Should an IR team use Article or NewsArticle?

Article is a clear default when an IR system publishes company announcements under a general content template. NewsArticle can fit a dated release that the company presents as news.

The type should match the visible page and the publisher’s content model. Do not select a type only because it appears more specific. Google’s Article documentation supports Article, NewsArticle, and BlogPosting markup. Last checked: 2026-10-24.

How should an IR page use Article structured data?

An Article structured data IR page should identify the visible release and its publication details. Google documents author, datePublished, dateModified, headline, and image for Article pages. Schema.org also defines publisher and mainEntityOfPage.

Use one primary Article node for each release page. Point mainEntityOfPage to the canonical page URL. Identify the issuing organization as the publisher. Use an organization as the author when that choice matches the visible page. Documentation last checked: 2026-10-24.

What must match the visible earnings release?

Structured data must represent content that readers can see on the page. The headline, publication date, update date, author, and publisher should not conflict with the visible release.

Google Search Central’s structured data policies prohibit misleading markup and markup for content that readers cannot see. A policy violation can affect rich result eligibility or lead to a structured data manual action. Last checked: 2026-10-24.

Should an earnings release use FAQPage markup?

FAQPage markup should appear only when the page includes visible questions with company-provided answers. The structured questions and answers must match the displayed content.

Google limits FAQ rich results to certain authoritative government and health websites. An IR team should not expect an earnings release to receive that search treatment. FAQPage can still describe visible content, but it does not force a rich result or AI citation. Google’s FAQ documentation was last checked on 2026-10-24.

How should an earnings release format financial figures for AI engines?

An earnings release should place key figures in visible HTML under clear headings. The reporting period should appear beside each figure so readers and retrieval systems can interpret it correctly.

Do not place a key number only in structured data, a chart interaction, or a document viewer. Article markup identifies the page, but it does not replace visible financial text. The related earnings release PDF and HTML guide explains the format choice without replacing the structured data work.

Why does visible page text still matter?

Visible page text gives readers and retrieval systems the financial facts they need. Structured data adds page metadata, but it should not be the only place where a date or figure appears.

State each key result near a clear heading. Keep the reporting period beside the figure. A reader should not need to open an interactive chart or document viewer to find the number.

Page-level Article structured data identifies the release page. Clear financial table markup labels each figure by row, column, period, and unit. The earnings table formatting guide covers labels, periods, units, footnotes, and GAAP reconciliations.

Does structured data help ChatGPT and Google AI Mode read financial pages?

Structured data gives systems explicit metadata about a financial page. Article markup can identify the title, dates, publisher, and canonical URL. It does not guarantee retrieval or citation by ChatGPT, Google AI Mode, Google AI Overviews, or Microsoft Copilot.

What does Google say about structured data and AI features?

Google says pages must meet its normal Search technical requirements to appear in AI features. Google does not require special schema markup or a special AI text file for eligibility.

Structured data should still match the visible page when a site uses it. Google also says that Search preview controls apply to its AI features. Google’s AI features guidance was last checked on 2026-10-24.

Which Google Search Central documents apply to financial release pages?

Google Search Central provides separate guidance for Article structured data, structured data policies, crawlable links, sitemaps, and AI features. These documents apply to publication and discovery tasks on an investor relations website.

The documents do not provide special financial schema for an earnings release. IR teams should use the general guidance, match markup to visible content, and keep links crawlable. Each cited Google document on this page was last checked on 2026-10-24.

What does structured data not do?

Structured data does not force a crawler to visit a page. It also does not guarantee indexing, ranking, answer selection, or citation.

This limit matters when an AI answer repeats an older figure. Accurate Article dates can clarify when a release appeared. They cannot force an engine to reject another source. The IR team must publish a clear page, permit crawling, and test the resulting answers separately.

Which AI engine gives the most accurate public company financial answers?

This page does not provide evidence that one engine is always more accurate than another. Structured data alone cannot rank ChatGPT, Google AI Overviews, Google AI Mode, or Microsoft Copilot by financial accuracy.

An IR team should capture each answer on a stated date and compare it with the same filing evidence. Engine results may then be reviewed using one consistent method. A general claim about the most accurate engine would require separate dated research.

How should an IR team make earnings release links crawlable?

An IR team should link each earnings release through a standard HTML anchor with an href attribute. The destination should load the release at a stable URL.

What does a crawlable results link look like?

A crawlable link uses an <a> element and a resolvable URL in href. Google documents this pattern in its crawlable link guidance.

Use descriptive anchor text, such as the visible announcement title. Avoid vague labels when several documents appear together. Google also advises against links that depend only on scripts or click handlers. Last checked: 2026-10-24.

<a href="/investors/results/example-earnings-release/">
  Example earnings release
</a>

Which IR page patterns can hide a release?

A results archive can hide a release when the link appears only after JavaScript runs. A button with an onclick action but no href also lacks the standard link pattern that Google documents.

An embedded document viewer may show a release without giving the archive a direct HTML link to it. A filter can create the same problem when no server-rendered link exists before interaction. Test the raw HTML instead of treating the browser view as proof of crawlability.

Should an IR website allow AI crawlers to reach earnings releases?

An investor relations website should not block approved crawlers from public earnings releases. Check robots.txt and edge controls for GPTBot, ClaudeBot, PerplexityBot, Googlebot, and Bingbot.

Crawler access does not guarantee indexing, retrieval, or citation. It only removes one possible access barrier. Test the public URL after every crawler or edge configuration change because a permissive robots.txt file does not prove that the page loads for each user agent.

Should an IR team check crawler access before changing structured data?

Yes. Check crawler access first. Valid structured data cannot help a crawler that cannot fetch the canonical HTML page.

Confirm that the release URL returns HTTP 200 for OAI-SearchBot. Check robots.txt and edge controls before changing the markup. The OAI-SearchBot access guide explains how to test access and record the results.

Should every release have a stable URL?

Each earnings release should have its own stable, canonical URL. A stable URL connects one release page with its Article node, internal links, sitemap entry, and verification record.

A recurring path such as /latest-results/ can point readers to the newest announcement. The IR site should also preserve a separate URL for each historical release. The broader IR website guide for Google AI Overviews covers site structure beyond this page-level task.

How should an IR team use an XML sitemap?

Add each canonical release URL to the XML sitemap. Use lastmod only when the page receives a meaningful update.

Google describes sitemaps as discovery signals, not guarantees of crawling or indexing. Google may ignore an inaccurate lastmod value. The value should reflect a significant content update, not an unrelated template change. Google’s sitemap documentation was last checked on 2026-10-24.

Can structured data make Google AI Overviews use the latest figures?

No. Structured data can describe a new release, but it cannot force Google AI Overviews to use its figures. Google controls crawling, indexing, retrieval, and answer generation.

What helps Google AI Overviews identify the latest quarter?

A distinct canonical URL, accurate visible dates, consistent Article markup, and crawlable internal links can clarify which release covers the latest quarter. An accurate sitemap entry can help Google discover the URL.

The page should state the reporting period and each key figure in visible HTML. Change dateModified only after a meaningful update. These steps reduce ambiguity, but none promises selection. Google says that meeting Search requirements does not guarantee appearance in its AI features. Last checked: 2026-10-24.

Can an IR team correct stale figures with schema alone?

No. Schema cannot correct stale visible content or an outdated third-party source. First confirm that the current IR page and SEC filing exhibit show the intended figure.

Next, verify that crawlers can reach the page and that Google can inspect it. Then record what each engine says on a stated date. The guide to financial data sources in Google AI Overviews explains why an answer may use a source other than the issuer.

How long will an updated figure take to appear?

There is no single published interval for updated figures to appear across AI engines. Publication, crawling, indexing, retrieval, and citation happen as separate steps.

Do not promise management a fixed correction date. Report the publication time, crawl evidence, index status, and dated answer checks instead. This creates a clear record even when an engine has not changed its response.

How should an IR team test the page after publication?

IR page verification after publication should test the public URL, structured data, crawl access, raw HTML, canonical URL, and filing figures. The test should use the deployed page rather than a preview.

What should the first live-page test include?

The first live-page test should confirm that the release loads without an HTTP error and appears in the initial HTML. View the source and search for the headline, publication date, reporting period, and key figures.

Confirm that the canonical tag points to the intended release URL. Check that archive and results pages use links with an href attribute. A rendered browser view is not enough because scripts can add content after the initial response.

How should the structured data be tested?

Run the live URL through Google’s Rich Results Test. Record the detected type, errors, warnings, test date, and tested URL.

The Rich Results Test checks whether Google can process supported structured data on that page. It does not prove that Google indexed the URL or will display a rich result. It also does not show whether another engine uses the markup. Tool behavior last checked: 2026-10-24.

How should crawl and index status be tested?

Use Google Search Console’s URL Inspection tool to inspect the canonical release URL. Record whether Google knows the URL, which canonical it selected, and any crawl or indexing issue shown.

A request for indexing is a request, not a guarantee. Keep the inspection result separate from the Rich Results Test result. URL Inspection reports Google’s view of a URL. The Rich Results Test checks supported structured data. Documentation last checked: 2026-10-24.

How should the filing comparison work?

The filing comparison should match visible release figures with the exhibit available through SEC EDGAR. For results reported under Form 8-K Item 2.02, use the related announcement or release exhibit.

Record the filing URL, exhibit, page location, and exact supporting text. If the IR page and exhibit differ, investigate the difference before using either source for verification. SEC Form 8-K describes Item 2.02 and its exhibit requirement. Last checked: 2026-10-24.

What should the verification log contain?

A useful verification log should preserve evidence instead of showing only a pass label. Keep a separate row for each release URL and test event.

1. Record the public URL and test time.
2. Save the canonical URL and HTTP result.
3. Record the detected Article type and validation messages.
4. Note the crawl and index status shown by the inspection tool.
5. Cite the SEC filing and exhibit used for comparison.
6. Capture each engine answer with its date and cited source.
7. Repeat the answer check after a correction.

The MS House methodology explains how answer checks are compared with filing evidence.

How does SEC Form 8-K Item 2.02 relate to the IR page?

Form 8-K Item 2.02 connects a completed-period results announcement with filing evidence. When the company provides the related release as an exhibit, the IR team can compare the public page with that primary document.

Which source should an IR team use for verification?

An IR team should use the company’s SEC filing and official IR page as verification sources. A news article, search snippet, or AI answer is not financial ground truth.

The comparison should preserve the exact filing text and its location. A reviewer should be able to move from the disputed answer to the release page and then to the SEC exhibit. This evidence trail helps the IR team explain a finding to finance, legal, and communications staff.

Does page verification prove that AI answers changed?

No. Page verification can show that the source page is published, crawlable, and consistent with the selected filing evidence. It does not prove that an AI engine changed its answer.

Proof of a changed answer requires another dated check. Use the same investor question, engine, and evaluation method when possible. Compare the new answer with the earlier capture and filing text. The public sample Investor Answer Audit shows how source evidence and action ownership can appear together.

FAQ

Does structured data guarantee that ChatGPT cites my earnings release?

No. Structured data cannot guarantee crawling, selection, or citation in ChatGPT, Google AI Overviews, Google AI Mode, or Microsoft Copilot. Each engine controls how it finds, weighs, and cites sources.

Which schema type should an earnings release page use?

Use Schema.org Article as the main type. The markup should identify the headline, publication dates, author, publisher, and main page URL. Use FAQPage only when the same questions and answers appear on the page.

Can I add FAQPage schema for questions that are not visible on the page?

No. Google Search Central requires structured data to represent visible page content. Markup that misrepresents a page can lose rich result eligibility or receive a structured data manual action.

How long does it take for new results to appear in AI answers?

There is no fixed time. Publication, crawling, indexing, retrieval, and answer selection happen on separate schedules. Test the page first, then record dated answers from each engine.

Does each earnings release need its own Article node?

An IR team should give each release a stable URL and one primary Article node that describes that release. This keeps the page identity, dates, publisher, and canonical URL clear.

See the sample Investor Answer Audit to review how an engine answer appears beside filing evidence and an assigned fix.