A customer asks ChatGPT where to find your business and gets your old phone number. Or they ask Google's AI Overview for your hours and it quotes a closing time you stopped using a year ago. Or worse, it recommends a competitor down the street instead of you, even though you outrank them in a normal Google search.
This isn't a website bug, and adding schema markup to your site usually won't fix it, even though schema is the fix people recommend most often. AI tools are more likely to trust and repeat your business information when your name, address, phone number, and hours match everywhere they appear online. AI systems build confidence in a fact by checking it against multiple sources. When those sources disagree, the AI either guesses, picks the wrong one, or leaves you out entirely.
The advice everyone gives, and why it's overrated
If you've asked anyone in marketing how to get an AI tool to mention your business, you've probably heard "add schema markup." Schema is a way of labeling parts of a webpage, your hours, your address, your services, so a computer program can read them without guessing. It sounds like exactly the kind of technical fix that should help an AI understand your business better.
Ahrefs tested that assumption at scale. In May 2026, Ahrefs published a data study which tested whether adding JSON-LD schema markup to a page produced a measurable increase in citations across Google AI Overviews, Google AI Mode, and ChatGPT.
The team identified 1,885 pages that added the markup between August 2025 and March 2026, matched them against 4,000 control pages with similar citation levels, and measured citation changes 30 days before and after.
The AI Mode and ChatGPT figures were close enough to zero that Ahrefs classified them as noise, while the AI Overviews change was actually a statistically significant decline.
A separate rollout presented at a marketing conference found the opposite pattern on some platforms. At BrightonSEO in April 2026, OtterlyAI presented results from a sitewide schema rollout across 2,000+ URLs where Google AI Overviews citations increased by 1,500% and AI Mode increased by 377%, while citations dropped on ChatGPT, Gemini, and Copilot, and Perplexity showed no impact.
That mixed result suggests schema does not affect every platform the same way: Google's ecosystem responded strongly in that rollout, while other AI tools did not appear to weight it the same way.
One technical detail helps explain the inconsistency. A separate experiment from searchVIU tested whether five major AI systems, ChatGPT, Claude, Perplexity, Gemini, and Google AI Mode, actually used schema markup when fetching a page in real time, and found that none of them did: every system extracted only visible HTML content: JSON-LD, hidden Microdata, and hidden RDFa were all ignored.
Schema markup is not worthless. It can still help a page get crawled and understood in the first place. But it does not determine whether an AI tool gets your basic business facts right, and it is not a substitute for consistent business information across the web.
What builds AI's confidence in your facts
Local search research has used the same concept for years, usually shortened to "NAP": your business Name, Address, and Phone number appearing identically everywhere it's listed. NAP consistency refers to your business name, address, and phone number being the same across all of your online citations.
To verify that its impression of a local business is correct, a search engine requires a sufficient pool of data, and many citations with identical information signal that what it knows about a business is correct.
For AI discovery, large language models and answer engines rely on retrieval pipelines, citations, and corroborating sources. A business with consistent NAP data is easier to match, summarize, and recommend, while a business with fractured records is more likely to be omitted or described incorrectly. The mechanism is corroboration, not code. If your website says one address, your Google Business Profile says another, and a directory listing has your old phone number, the AI has no clean way to decide which version to trust, so it either picks wrong or drops you from the answer.
For multi-location businesses, the risk is higher. Search engines and AI engines treat each business location as a separate entity, and when the name, address, or phone number for a location varies across listings, the algorithms hedge by showing competitors instead.
What to do about it
If you want an AI tool to describe your business correctly when a customer asks, the practical work isn't installing a plugin. It's an audit:
- Pick one canonical version of your business name, address, and phone number, and write it down exactly as it should appear everywhere.
- Check your Google Business Profile, your website's contact page and footer, and any directory listings you can find, and fix every place they don't match.
- Update anything tied to a location you've moved from, a number you've disconnected, or hours that changed and were never corrected on a third-party site.
- Re-check this periodically. A single missed update on one directory is enough to reintroduce the inconsistency you just cleaned up.
That audit is less appealing than a technical fix you install once and forget. But it matches the problem the evidence supports, and it is the same foundation that keeps your Google Business Profile accurate for the customers who still search the old-fashioned way.
Key Takeaways
- AI tools like ChatGPT and Google's AI Overview build confidence in facts about a business by cross-checking multiple sources, not by reading a single authoritative page.
- Recent studies on schema markup and AI citations show mixed, platform-dependent results, and one large study found no meaningful citation increase and even a decline on Google AI Overviews.
- Independent testing found that several major AI systems ignore schema markup entirely during real-time page retrieval, reading only the visible text on a page.
- Consistent name, address, phone number, and hours across your website, Google Business Profile, and directory listings is the signal that research ties most directly to accurate AI mentions.
- Fixing this requires an ongoing audit of where your business is listed online, not a one-time technical install.
Getting an AI tool to describe your business accurately starts with the same foundation as ranking locally: information that matches everywhere it appears. Mindstate Strategy's SEO & GEO service audits and corrects that consistency across the web, and you can get in touch to have your current listings reviewed.
