How to Add LocalBusiness Schema Markup to Your Website

LocalBusiness schema gives search engines and AI assistants a machine-readable confirmation of your name, address, phone number, hours and service area — the exact facts they need to answer "near me" and local recommendation queries confidently.

About 1 hour 8 steps Last updated: 30 July 2026

1

Gather your exact, consistent business details

Legal business name, full address, phone number, and opening hours exactly as they appear on your Google Business Profile — any mismatch between schema and GBP creates a trust problem.
2

Choose the correct schema subtype

Schema.org offers dozens of LocalBusiness subtypes (e.g. ProfessionalService, Store, Restaurant); pick the most specific one that matches your business rather than the generic LocalBusiness type.
3

Build the core JSON-LD object

Include "@type", "name", "address" (as a nested PostalAddress), "telephone", "openingHoursSpecification", and "url" at minimum.
4

Add geo-coordinates if available

A "geo" property with latitude/longitude removes any ambiguity about your physical location, useful for multi-location businesses in particular.
5

Include "areaServed" for your service region

List the specific cities, regions or countries you serve, rather than relying on your address alone to imply reach.
6

Add the schema as JSON-LD in your page's <head>

It doesn't need to be visible to users, only present in the page's HTML source, typically on your homepage and contact page.
7

Validate with Google's Rich Results Test

Confirm the LocalBusiness type is detected with no errors, and fix any missing required-field warnings.
8

Keep it in sync with your Google Business Profile

Whenever your hours, address or phone number changes, update both at the same time to avoid the NAP inconsistency that undermines local search.

Pro tips

If you have multiple locations, use a separate LocalBusiness object for each, rather than one generic object for the whole business.
Don't guess at opening hours — inaccurate schema is worse than no schema, since it can actively mislead users and AI systems.