Published March 19, 2026
Local businesses — restaurants, salons, gyms, hotels, auto shops — operate differently from service businesses in one fundamental way: the customer comes to you. That distinction changes which schema fields matter most and why.
When someone asks an AI assistant to find a restaurant nearby or compare gyms in their area, AI needs to confirm basic facts: Is this place close enough? Is it open right now? Schema is how a local business provides those answers directly to AI.
The AIFDS local business library contains 130+ blueprints across restaurants, salons, gyms, hotels, auto shops, and more — all built from research into which schema fields AI systems actually read.
The core findings apply across every business type, but local businesses experience them through the lens of physical presence. For a service business, service area tells AI the radius it covers. For a local business, physical address tells AI where the customer needs to go. Both are location signals — the difference is direction.
Hours of operation carry more urgency here than almost any other type. A user asking AI for a lunch spot at 11:45 AM needs a place that is open now. If those hours are not in schema or in up-to-date directories, AI cannot confirm availability — and recommending a closed business is worse than no recommendation at all.
LLM visibility starts with being findable. For local businesses, that means AI knows where you are, when you are open, and what the user will experience when they arrive.
The fields that matter map to AI's standard decision chain, but shift toward physical experience and real-time availability.
The most critical field for local businesses. AI matches location queries by proximity. Without a structured address and coordinates, AI cannot calculate distance or confirm the business is near the user.
AI checks hours before recommending. A gym that closes at 8 PM will not be recommended to a user asking at 8:30 PM. Hours in schema give AI a direct answer; hours missing force AI to check directories — or skip the business entirely.
Phone and email serve as both trust signals and functional data. If contact data is not in schema, AI either cannot provide it or has to extract it from page content — which is less reliable. AI systems parse contact information directly from JSON-LD and serve it to users who ask.
A restaurant is not a bar. A salon is not a spa. The AIFDS blueprints map each local business to its correct schema.org type — Restaurant, HealthAndBeautyBusiness, SportsActivityLocation, LodgingBusiness, AutoRepair, and others. Correct classification means accurate matching.
These are the fields that differentiate local businesses from each other. A restaurant's menu link tells AI what kind of food is served. A hotel's amenities tell AI whether it has a pool or parking. A gym's class schedule tells AI what is available. These fields move a business from "exists in this category" to "matches what the user is specifically looking for."
Pricing helps AI pre-qualify users. A controlled experiment showed AI-referred traffic converting at 24.9% — roughly 6–10x the industry benchmark. Even a price range signal ($ vs $$ vs $$$) gives AI enough to match budget expectations.
The pattern matches every other industry in this research. The issue is not incorrect schema — it is missing schema. Most local businesses have no structured data at all, or a minimal implementation covering only the business name. The consequence is invisibility to AI systems.
Schema is eligibility. If AI cannot confirm where you are, when you are open, and what you offer, you are not in the consideration set. The business down the street with structured data is.
Local businesses depend on directories more than any other type. Google Business Profile, Yelp, Bing Places, TripAdvisor — AI cross-references these against schema to verify a business.
Context consistency is how AI builds confidence. If your schema says you close at 9 PM but Google says 8 PM, AI confidence drops. Inconsistency between sources is a reason to recommend a competitor whose data aligns. Schema provides the structured data on your site; directories provide the third-party confirmation. Both need to tell the same story.
The AIFDS local business blueprint library contains 130+ blueprints organized by business type, then by page type. Each type maps to a specific schema.org type — Restaurant, HealthAndBeautyBusiness, SportsActivityLocation, LodgingBusiness, AutoRepair — and carries its own relevant fields. A restaurant needs menu links; a hotel needs amenity lists; a gym needs class or membership details.
Within each type, blueprints cover every page — homepage, location pages, service or menu pages, about, contact, FAQ, and more. Every blueprint contains the exact JSON-LD fields AI needs for that combination.
Start with the homepage. Include the full address, geo coordinates, hours, contact information, business type, and a clear description.
Ensure directory alignment. Before adding schema to secondary pages, make sure Google Business Profile, Yelp, and Bing Places match your homepage schema. Inconsistencies undermine AI confidence in everything else.
Add location-specific pages. If the business has multiple locations, each one needs its own page with its own schema — separate address, hours, and contact data. AI treats each location as a distinct entity.
Structure offerings. Menus, amenities, and class schedules should be on dedicated pages with schema. Content structure matters for AI — this is how AI matches the business to specific queries rather than just category-level ones.
David Valencia writes about how AI systems find, parse, and cite websites.
Overlapping findings for businesses that go to the customer.
Read →