Schema Markup for AI Search: FAQPage, Speakable & HowTo in WordPress

Schema markup has always helped search engines understand your content. In 2026, it does the same for AI engines – but the stakes are different. AI systems like Google AI Overviews, ChatGPT, and Perplexity use structured data to decide what to extract, what to cite, and what to trust.

Pages with proper schema are 2-4x more likely to appear in AI-generated responses. Yet only about 12% of websites implement structured data at all. That gap is an opportunity.

This guide covers the schema types that matter most for AI search visibility, how to implement them in WordPress, and which ones you can skip. It’s part of my AEO checklist for WordPress.

Skip to the implementation guide

How AI Engines Use Schema Differently Than Google

Traditional SEO schema is about earning rich snippets – star ratings, FAQ dropdowns, recipe cards in search results. AI engines don’t care about any of that.

When ChatGPT, Perplexity, or Google AI Overviews process your page, they use structured data as a semantic layer. Schema tells them what type of content the page contains, who wrote it, when it was last updated, and which sections answer specific questions.

This distinction matters for your implementation strategy. You’re not marking up content for a rich result card. You’re marking it up so an AI model can parse it accurately during retrieval.

The 5 Schema Types That Matter for AI Search

Not all schema types carry equal weight for AI visibility. Based on current citation data and how AI engines process structured data, these five deliver the most value:

Let’s look at each one and how to implement it in WordPress.

1. FAQPage – The Highest-Impact Schema for AI

FAQPage schema pre-formats your content in the exact question-answer structure that AI systems need. When an AI engine encounters FAQPage markup, it can extract individual Q&A pairs without parsing surrounding text.

Google restricted FAQ rich snippets to government and health sites in 2023. That change made many site owners stop using FAQ schema entirely. That was a mistake.

AI platforms like ChatGPT, Perplexity, and Claude still rely heavily on FAQPage markup as their primary extraction format. Pages with FAQ schema see 28% higher citation rates than those without, regardless of whether Google shows the rich snippet.

When to use it: On any page that answers multiple questions about a topic – blog posts, product pages, service pages, landing pages. The questions should be real questions users ask, not filler.

WordPress implementation: Use the Yoast FAQ block or Rank Math‘s built-in FAQ block in the block editor. Both generate the correct JSON-LD automatically. For manual implementation, add the JSON-LD directly:

{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [
    {
      "@type": "Question",
      "name": "Your question here?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "Your answer here."
      }
    }
  ]
}

2. Article – The Foundation Schema

Article schema (including BlogPosting and NewsArticle) is the baseline. It tells AI engines what type of content the page is, who wrote it, and when it was published and last updated.

Three fields matter most for AI:

• author – Link to a Person schema with a stable @id and url. AI systems use author information to verify expertise. A named author with a profile page carries more weight than “admin.”
• datePublished – When the content was first published. AI engines factor freshness into citation decisions.
• dateModified – When the content was last updated. This is the field AI engines check most. A page updated this month is more likely to be cited than one untouched for two years.

WordPress implementation: Yoast SEO and Rank Math both generate Article schema automatically. Make sure your author profiles are complete (full name, bio, profile image) and that dateModified updates when you revise content. Some themes don’t update this field on minor edits – verify in the Rich Results Test.

3. HowTo – For Procedural Content

HowTo schema marks up step-by-step instructions. AI engines use it to identify procedural content and extract clear, bounded steps for queries like “how to” and “how do I.”

The key is tightly scoped steps. Each step should describe one action, not a paragraph of context. AI engines extract individual steps, so vague or compound steps produce poor citations.

When to use it: Tutorial posts, setup guides, installation instructions – any content where the reader follows sequential steps to achieve a specific outcome.

WordPress implementation: Rank Math includes a HowTo block that generates the JSON-LD. You can also add it manually:

{
  "@context": "https://schema.org",
  "@type": "HowTo",
  "name": "How to Do Something",
  "step": [
    {
      "@type": "HowToStep",
      "name": "Step 1 title",
      "text": "What to do in this step."
    },
    {
      "@type": "HowToStep",
      "name": "Step 2 title",
      "text": "What to do in this step."
    }
  ]
}

4. Speakable – Marking Content for Voice

Speakable is a schema.org property that identifies which sections of your content are best suited for text-to-speech playback. Voice assistants and AI search engines use it to decide what to read aloud when answering a query.

It’s currently in beta and primarily supported for Google News publishers. But with 62% of searches now involving voice, marking your key sections as speakable positions your content for a growing channel.

How it works: You add a SpeakableSpecification inside your Article schema, pointing to specific content sections using CSS selectors or XPath.

{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "Your Article Title",
  "speakable": {
    "@type": "SpeakableSpecification",
    "cssSelector": [".article-summary", ".key-takeaway"]
  }
}

What to mark as speakable: Target the first paragraph of a section, a TL;DR block, or a direct answer to a question. Keep speakable sections to 40-50 words – concise enough for a voice assistant to read in one breath.

WordPress implementation: The “Structured Data for WP” plugin has built-in Speakable support. Enable it in the Article schema settings and set conditions for which posts include it. For manual implementation, add the JSON-LD to your theme’s header or use a custom hook.

5. Organization – Entity Establishment

Organization schema establishes your brand as a recognized entity in AI knowledge graphs. It includes your name, logo, social profiles, and contact information.

This schema belongs on your homepage or contact page – Organization schema on every page is a bad implementation pattern. AI engines use it to build an entity profile for your site, which feeds into how they evaluate your content’s authority.

WordPress implementation: Both Yoast and Rank Math handle this automatically through their setup wizards. Verify that your organization name, logo URL, and social profiles are filled in correctly. Check the output in Google’s Rich Results Test.

Implementing Schema for AI in WordPress

Here’s a practical workflow for adding AI-relevant schema to your WordPress site:

1. Audit your current schema. Run your homepage and 2-3 key posts through Google’s Rich Results Test. Note which schema types are already present and which are missing.
2. Verify Article schema basics. Check that your SEO plugin generates Article or BlogPosting with author, datePublished, and dateModified. Fix any missing fields.
3. Add FAQPage schema to your top content. Start with your 5-10 highest-traffic posts. Add genuine Q&A sections and mark them up with FAQ schema. This has the highest impact-to-effort ratio.
4. Add HowTo schema to tutorials. Any post with step-by-step instructions should have HowTo markup. Use your SEO plugin’s built-in block or add JSON-LD manually.
5. Verify Organization schema. Make sure it exists on your homepage with correct name, logo, and social links. Remove it from other pages if it was added everywhere.
6. Consider Speakable for key content. If voice search is relevant to your audience, add SpeakableSpecification to your most important articles, targeting summary paragraphs.

QAPage vs FAQPage: Which One Do You Need?

These two schema types are often confused, but they serve different purposes:

For most WordPress blog and business sites, FAQPage is the right choice. QAPage only makes sense if your site has user-generated Q&A content like a forum or support community.

Common Mistakes to Avoid

• Schema without matching visible content. Every schema property must have corresponding content visible on the page. Hidden or mismatched schema violates Google’s guidelines and AI engines can detect the discrepancy.
• Generic FAQ filler. AI engines evaluate answer quality. Generic questions like “Why should I choose your product?” with marketing-speak answers get ignored. Use real questions that users actually ask.
• Missing dateModified. Many WordPress themes don’t update this field on content revisions. AI engines treat stale dateModified values as a freshness signal – or lack thereof.
• Organization schema on every page. This dilutes the signal. Keep it on the homepage or About/Contact page only. See: Organization schema on every page.
• Skipping validation. Always test with the Rich Results Test before and after changes. Invalid schema is worse than no schema – it can confuse AI engines about your content structure.

Schema markup increases AI citation probability by 30-36%. But schema alone doesn’t guarantee citations – it makes content parseable, not recommendable. Quality content structure and expertise signals are equally important. Schema is one layer of AEO, not the whole strategy.

FAQs

Common questions about schema markup for AI search:

Summary

Schema markup for AI search comes down to five types: FAQPage for the highest citation impact, Article for content identity and freshness, HowTo for procedural queries, Speakable for voice search positioning, and Organization for entity establishment.

Start with FAQPage and Article – they cover the most ground with the least effort. Add HowTo to your tutorials, verify Organization on your homepage, and consider Speakable if voice search matters to your audience.

The window of opportunity is real. With only 12% of sites using structured data, proper schema implementation puts you ahead of most competitors in AI search visibility.