Published March 19, 2026
Service businesses — contractors, agencies, law firms, accountants — depend on being found by the right person at the right time. That has always been true for search. It is now equally true for AI.
When someone asks an AI assistant to recommend a roofer, find a marketing agency, or compare local law firms, the AI needs structured data to understand what each business does and where it operates. Schema is how a service business provides those answers directly to AI.
Most service businesses either have minimal schema on their homepage or none at all. That gap is the single biggest reason they get skipped by AI — not bad schema, but missing schema. If your brand is invisible to ChatGPT, this is likely why.
The core findings on which schema fields AI systems read apply across every industry. But service businesses experience them differently because the signals that drive traditional SEO and LLM visibility are not the same — and service businesses feel that gap more than most.
A service business lives and dies by location and availability. A roofing contractor who serves a 30-mile radius needs AI to understand that boundary. If the contractor's schema does not include service area, AI cannot confirm the match — and it will recommend a competitor whose data is clearer.
This distinction runs through every schema field. Hours tell AI whether a contractor is available for an emergency call right now. Contact information tells AI whether a client can reach the law firm. Service descriptions tell AI whether the agency handles the specific type of work the user is asking about. Each field answers a question AI needs resolved before recommending the business.
The schema fields that matter for service businesses map to the questions AI asks before citing any business — but service businesses need to answer them with more specificity than most.
AI uses schema type to categorize what a business does. The AIFDS blueprint library maps each service business to its correct schema type — a roofer falls under GeneralContractor, a law firm under LegalService, an agency under ProfessionalService. Getting this right means AI categorizes the business correctly from the first interaction. Leaving it out means AI has to guess from page content.
For service businesses that operate within a geographic radius, this is critical. A contractor who serves Minneapolis but has no service area in schema is invisible to a user asking AI for a contractor in Minneapolis. The business exists — AI just does not know it serves that area.
AI treats contact data as both a trust signal and as functional data it can provide directly to users. Research has shown that AI systems parse contact information from JSON-LD schema and serve it to users who ask. If the contact data is not in schema, AI cannot complete that loop.
A user searching at 3 PM for an emergency plumber needs a business that is open now. Without hours in schema or in up-to-date directories, AI cannot confirm availability and will default to a business whose hours are structured.
A law firm that handles personal injury, estate planning, and business litigation needs each service represented — not just a generic description. AI matches user queries to specific services. If someone asks for a business litigation attorney and the firm's schema only says "legal services," the match is weaker than a competitor who lists business litigation explicitly.
Transparent pricing lets AI pre-qualify users. A controlled experiment showed AI-referred traffic converting at 24.9% — roughly 6–10x the industry benchmark. Fewer leads, but leads that arrive already know the price range. For service businesses where pricing varies by project, even a starting price gives AI something to work with.
Across the service businesses tracked in this research, the most common issue was not incorrect schema. It was schema that simply did not exist — no service area, no hours, no individual services, no contact details in structured format.
This is an awareness gap, not a technical failure. Most service business owners do not know schema exists, and most web developers treat it as an optional SEO checkbox rather than a visibility requirement for AI. Schema is eligibility. A service business without it is not being evaluated and rejected by AI — it is not being evaluated at all.
The AIFDS service blueprint library contains 120+ blueprints organized by business type, then by page type. Each type maps to a specific schema.org type:
Within each business type, blueprints cover every page — homepage, service pages, about, contact, FAQ, blog, and more. Fields overlap where AI's questions are the same (contact info, hours, service area) but diverge where the business context differs (service descriptions, pricing structures, team credentials).
The priority order follows the same logic as the general findings, applied to how service businesses are found:
Start with the homepage. The homepage blueprint includes business classification, contact information, service area, hours, and a high-level description. Without this, nothing else matters.
Add individual service pages. Each service should have its own page with schema describing that specific offering. A user asking for "commercial roof repair" should find a business whose schema describes commercial roof repair — not just "roofing services."
Ensure third-party directories match. Google Business Profile, Yelp, Bing Places — AI cross-references these against schema. If your schema says you close at 6 PM but Google says 5 PM, AI confidence drops. Alignment is not optional.
Keep content and schema aligned. If schema lists a 30-mile service area but the page content only mentions one city, the mismatch weakens AI confidence. Context consistency is foundational — schema and content must tell the same story.
David Valencia writes about how AI systems find, parse, and cite websites.
Overlapping findings for businesses with physical locations.
Read →