CrawlCheck

Guides · 2026-10-02 · By · 1 view

LocalBusiness schema for service-area businesses: example

Google requires two properties. The rest is where local records break: coordinates with the wrong sign, a name that differs from the listings, two nodes claiming one phone. A node to copy, and the four checks that read it.

Google's LocalBusiness documentation requires only name and address; geo, telephone, url, openingHoursSpecification and priceRange are recommended, and coordinates need at least five decimal places. For a service-area business, use the most specific subtype, one stable @id, the exact name and phone your listings use, a real PostalAddress (never an invented storefront), coordinates for that address, and areaServed for the cities you actually serve. Then check for an address with no coordinates, coordinates in the wrong country, two nodes claiming one identity, and a name that drifts from your listings.

Check your own site free

About a minute. First scan free with your email, nothing installed.

Google requires only two LocalBusiness properties, and a node can pass its Rich Results Test while carrying a phone no listing uses or coordinates on another continent. The failures are in the recommended properties, where the node contradicts itself or the business’s other records. This guide gives a complete node to copy for a service-area business such as a contractor, explains each property, and covers the four checks that catch a node disagreeing with the world.

LocalBusiness node@type (specific subtype)@id …/#businessname · telephoneaddress (PostalAddress)geo (latitude, longitude)areaServedopeningHoursSpecificationurl · sameAs · imageName drift: name or phone differs from listingsNo coordinates: an address with no geoWrong country: geo outside the address's countryCollision: two nodes claim one identity
Google requires only name and address. The other properties are where records disagree with each other.

What Google requires, and what it only recommends #

Google's LocalBusiness documentation lists two required properties: name and address. Everything else is recommended: geo, telephone, url, openingHoursSpecification, priceRange, department and, for sites that review other businesses, aggregateRating and review. Two details in that documentation matter more than they look. Coordinates must carry at least five decimal places, which is roughly one metre. And review markup is meant for sites that collect reviews of other businesses, not a business marking up praise for itself.

Passing Google's Rich Results Test only proves the required properties are present and well formed. It does not compare the node with anything, so a node can pass while carrying a phone number no listing uses or coordinates in another hemisphere.

A complete node to copy #

Replace every value in angle brackets with your own and delete any property you cannot fill truthfully. Use the exact name and phone number your business profile and listings carry.

{
  "@context": "https://schema.org",
  "@type": "HomeAndConstructionBusiness",
  "@id": "https://example.com/#business",
  "name": "Example Fence Co",
  "url": "https://example.com/",
  "telephone": "+1-303-555-0100",
  "address": {
    "@type": "PostalAddress",
    "streetAddress": "<the street address you operate from>",
    "addressLocality": "Denver",
    "addressRegion": "CO",
    "postalCode": "<postal code>",
    "addressCountry": "US"
  },
  "geo": {
    "@type": "GeoCoordinates",
    "latitude": "<latitude of that address, five or more decimals>",
    "longitude": "<longitude, negative west of Greenwich>"
  },
  "areaServed": [
    {"@type": "City", "name": "Denver"},
    {"@type": "City", "name": "Lakewood"},
    {"@type": "City", "name": "Arvada"}
  ],
  "openingHoursSpecification": [{
    "@type": "OpeningHoursSpecification",
    "dayOfWeek": ["Monday", "Tuesday", "Wednesday", "Thursday", "Friday"],
    "opens": "<HH:MM>",
    "closes": "<HH:MM>"
  }],
  "sameAs": [
    "<your Google Business Profile URL>",
    "<a directory listing that links back to this site>"
  ],
  "image": "https://example.com/<a real photo of your crew or work>.jpg"
}

What goes in each property? #

PropertyWhat to put in itWhy it matters
@typethe most specific subtype, such as HomeAndConstructionBusiness or Plumbera category signal; LocalBusiness alone says nothing about the trade
@idone stable IRI, such as https://example.com/#business, reused on every pagelets every page and every other node refer to one entity
namethe exact name used on the business profile and listingsa different spelling reads as a different business
telephonethe number the listings carry, in one formatthe strongest join key between records
addressa full PostalAddress, even if the address is not shown to visitorsrequired by Google
geolatitude and longitude of that address, five or more decimalsplaces the record on a map without geocoding
areaServedthe cities or regions actually served, as City nodesthe only way to say where a service-area business works
sameAsthe business profile, listings and social profiles that link backcorroboration; see entity SEO

Declare the node once in the site template, so every page emits the identical record, and refer to it from page-level nodes by @id rather than redeclaring it. A node redeclared per page with small differences is the usual source of the drift described below.

What address should a business with no storefront use? #

A real one, never an invented one. Google lists address as required, and it is also an identity key, but nothing obliges you to claim premises customers can visit. Use the address the business genuinely operates from, keep it in the markup, and hide it from visitors if you like; Google’s business profile lets a service-area business hide its address and list service areas instead. If you will not publish a street address at all, a PostalAddress with locality, region, postal code and country is still truthful. Say plainly in visible text that customers are served at their own location, so the page and the markup agree. What you must not do is rent a virtual office or borrow a storefront address to fill the field: that is a false fact a record-matching system will try to reconcile with every listing.

Where the business works is what areaServed is for. List the cities you actually serve, not every town within a hundred miles. A long list of places with no matching pages reads as a doorway pattern; we measured the gap between declared areas and published area pages in schema that lists service areas the sitemap barely covers.

Failure one: an address with no coordinates #

A PostalAddress without geo forces every consumer to geocode the text itself, and addresses geocode differently in different systems, particularly rural routes and new developments. A business entity with no coordinates turns up on 6.2% of scanned sites. Add the coordinates of the address itself, taken from a map, not of the city centre.

Failure two: coordinates that contradict the address #

The commonest form is a missing minus sign. Western-hemisphere longitudes are negative, and a longitude of 104.9903 instead of -104.9903 moves a Denver address to northern China. It passes every syntax check because a positive longitude is legal; we found a real case in a newsroom that published coordinates in China. Coordinates falling in a different country from the one the address names turn up on 0.5% of scanned sites.

Failure three: two nodes claiming one identity #

When a page carries two LocalBusiness nodes with the same telephone or address but different names and no shared @id, a system resolving the business cannot tell which one is real. This usually happens when a theme and an SEO plugin both emit a node, and it turns up on 2.6% of scanned sites. Before treating it as two businesses, count the distinct @ids: two blocks with an identical @id are one record stated twice, explained in JSON-LD @id errors.

Failure four: a name that drifts from the listings #

The node says one name, the business profile says another, a directory listing a third. Small differences such as “LLC”, “&” versus “and”, or a city appended to the name are enough for record-matching systems to hesitate. A node name that differs from the names on listings found under the same phone number turns up on 0.6% of scanned sites. Pick one form, use it on the profile, the site and every listing, and change the others to match. The full name, address and phone comparison is in the NAP consistency audit.

One business, several locations #

A business with more than one office or yard should declare one node per location, each with its own @id, address, coordinates and telephone, plus one Organization node for the company that every location points to with parentOrganization. Matching systems treat a shared phone as the strongest sign that two records are one business, which is exactly the collision described above, so give two locations the same number only if the business really answers both on one line. Put each location’s node on that location’s page and the organisation node on the homepage.

Hours, price range and images #

openingHoursSpecification should match the hours on the business profile exactly, including closed days, because a mismatch is one more disagreement for a record-matching system to resolve. Use priceRange only if it is honest and stable. Add image with a real photograph of the business, its vehicles or its work, not a logo or a stock photo, and a logo property for the logo itself.

Checking your own node #

Fetch the homepage as a crawler would, extract every application/ld+json block, and read the business node as data rather than in a browser. Confirm one node, one @id, the listing name and number, coordinates with five decimals inside the right country, and the cities you serve under areaServed. Then compare those facts with the visible page text. A free scan at CrawlCheck runs the four checks above on any domain and shows which property each finding came from.

Share of scanned sites with each finding in this articleAddress with no coordinates6.2%Two business records share a phone, @id or location2.6%Directory listing under a different name0.6%Coordinates outside the address's country0.5%Share of scanned sites with eachfinding in this articleAddress with no coordinates6.2%Two business records share a phone, @id or location2.6%Directory listing under a different name0.6%Coordinates outside the address's country0.5%
Read live from the same counters the dataset page uses, at the moment this page was served. Bars are scaled to the largest value shown, not to 100%.

Share this finding:LinkedInXEmailFollow Emmanuel on LinkedIn

Questions this post answers

Which LocalBusiness properties does Google require?

Only name and address. Google recommends geo, telephone, url, openingHoursSpecification, priceRange and others, but a node with a name and a PostalAddress meets the requirement.

What address should a service-area business with no storefront put in its schema?

The real address the business operates from, never a borrowed or virtual storefront. It can stay in the markup while being hidden from visitors, or be reduced to locality, region, postal code and country. Say in visible text that customers are served at their location, and list the cities you work in under areaServed.

How precise must LocalBusiness coordinates be?

Google's documentation says latitude and longitude must have at least five decimal places, which is about one metre.

Why do my coordinates put my business in another country?

Usually a missing minus sign. Longitudes west of Greenwich are negative; a positive value moves a North American address to Asia while still passing syntax checks.

What is an entity collision in structured data?

Two business nodes on one page that share a phone number or address but have different @ids or none, so a system cannot tell which describes the real business. Two blocks with the same @id are one node, not a collision.

Should a business mark up its own reviews with aggregateRating?

Google's LocalBusiness guidance recommends review markup for sites that collect reviews about other businesses, not for a business rating itself.

Related findings

The main product

Found this on your own site? We fix it for $100.

Scan free to see where you stand. The fix is one site, every finding implemented and re-measured, with a sealed before and after.

Free browser extension

Check this on any site you open, without leaving the page.

CrawlCheck for Chrome grades every site you visit A–F and shows, from your own browser, what search engines and AI assistants are actually served. Free, no account.

Write for the library

Measured something yourself? Submit an article.

Send your finding, the data and the method. A person reads every submission and replies by email; accepted articles are published under your name with a link to your study. Free to submit. Articles are accepted on merit; no one can pay to have one published.

Submit an article

How it works · a reply within about a week

Spotted a mistake or a bug on this page? Report an error. Confirmed errors are fixed and logged on the corrections page.

Comments

Comments are read before they appear. Nothing is published automatically, and no account is needed. An AI agent with a published signing key can post without the queue: it signs the request and the signature is checked against its own domain (how). Those appear at once, marked verified.

Writing about this? Facts, live figures and marks — every number on that page is dated and traceable to a scan.

All guides · The dataset · How the dataset works