Organization schema and how machines recognise your business

Organization schema and how machines recognise your business

Organization schema and how machines recognise your business

THE SHORT ANSWER

Organization markup exists to convert a string of characters into something a machine can resolve to a specific real company. The properties that do the work are sameAs, which links you to profiles a system already trusts, a stable @id referenced from every page, and verifiable identifiers such as a registration or VAT number. Description, logo and address matter too, but only if they match exactly what every other source says about you.

Two roofing firms in Antwerp can share a trading name, a service area and a colour scheme. To a human that is a mild nuisance. To a system trying to attribute a claim to a company, it is a resolution failure, and the safe behaviour when resolution fails is to say nothing specific about either firm.

Organization markup is how you make that resolution succeed. It matters less for the tags themselves than for what they let you do, which is give a machine several independent routes to the same conclusion about who you are and where you trade. Agreement between those routes is what a confident machine description actually gets built from.

The numbers, at a glance

  • The property that carries most weight: sameAs, because it links your node to profiles the system already has confidence in

  • The property most often omitted: a stable @id, which lets every page on the site point at one entity rather than declaring a new one

  • Strongest single identifier: a company registration or VAT number, since it is checkable against a public register

  • Most damaging inconsistency: a legal name in the markup that differs from the trading name used everywhere else

The properties that earn their place

  • name and legalName. The trading name people use and the registered entity behind it. Give both when they differ; give one consistently when they do not.

  • @id. A canonical URI for the organisation node, typically your homepage URL with a fragment. Reference the same value from every page so the site describes one entity, not forty.

  • sameAs. An array of URLs to profiles that already represent you: LinkedIn, a Wikidata item if one exists, a trade register listing, industry association pages, review platforms. This is the corroboration layer.

  • address with PostalAddress. Street, locality, postal code and country as separate fields rather than one blob, matching the format on your business profile character for character.

  • identifier, vatID or taxID. A number that can be checked against a public register turns a claim into a verifiable fact.

  • areaServed and knowsAbout. Where you operate and what you do, expressed as data rather than left to be inferred from marketing copy.

Why sameAs does more than the rest combined

A knowledge system builds confidence in an entity through agreement between independent sources. Your own markup is a claim; a LinkedIn company page with matching details is a second observation; a Wikidata item with a matching registration number is a third. Agreement across three sources is what converts a probable match into a confident one.

This is also why the property fails when treated as a link-building slot. Twenty sameAs entries pointing at low-quality directory profiles that spell your address three different ways actively reduce confidence, because now the system has contradictory evidence about a single entity. Six accurate, corroborating profiles beat twenty sloppy ones by a distance.

Choosing between Organization and LocalBusiness

LocalBusiness is a subtype of Organization, and its more specific descendants such as HomeAndConstructionBusiness, RoofingContractor, HVACBusiness, Electrician or Plumber exist precisely for the trades. Using the narrowest accurate type tells a system what kind of thing you are, not merely that you are a company, and that classification is what lets you be surfaced for an intent rather than a name.

The practical pattern for a contractor with one office and a service radius: the most specific LocalBusiness subtype on the homepage, with openingHoursSpecification, areaServed and geo details, and a plain Organization reference by @id on every other page. Sites with several branches want one node per physical location plus a parent organisation tying them together.

What breaks entity recognition in practice

Almost every failure traces to inconsistency rather than absence. A company that markets itself under a short trading name, invoices under a long legal one, lists a home address on the register and a workshop address on the website has given the world four partially-overlapping descriptions and no way to reconcile them. No amount of schema fixes a disagreement between the sources it is meant to summarise.

The fix is dull and effective. Pick the canonical form of the name, address, phone number and description, write it down, and propagate it everywhere before touching the markup. Then the schema becomes what it should be: a summary of an already-consistent footprint, rather than an assertion nobody else corroborates.

Making your business resolvable

  1. Write the canonical name, legal name, address, phone number and one-line description in a single reference document.

  2. Correct every profile that disagrees with it before you edit a line of markup.

  3. Publish an Organization or LocalBusiness node with a stable @id and reference it from every page template.

  4. Populate sameAs with six to ten accurate profiles rather than every directory you can find.

  5. Add a registration or VAT identifier so at least one property is checkable against a public register.

Want leads like this in your pipeline?

Flock runs the campaigns, screens the enquiries and hands you only the ones that match your service area, job size and capacity. You pay per lead, not per month.

Book a 15-minute fit check  |  See lead package pricing

Related answers

Frequently asked questions

How many sameAs links should I include?

Six to ten accurate ones. The criterion is corroboration, not volume: every entry should be a profile that actually represents your business and repeats your canonical details. A directory listing with a wrong address does more harm than an omitted link ever could.

Do I need a Wikidata item?

It is not required, and creating one for a business with no independent coverage tends to end in deletion. If your company genuinely has verifiable third-party sources, an accurate item is a strong corroborating node. If it does not, spend the effort on the profiles you can legitimately control.

Should every page carry the full Organization block?

No. Publish the full node once, typically on the homepage or an about page, then reference it by @id everywhere else. Repeating the whole block on every page invites drift, because sooner or later one template gets updated and the others do not.

Does this markup affect AI answers directly?

Indirectly. It rarely supplies the sentence an assistant quotes. What it does is make the system confident that the sentence belongs to your company rather than a similarly-named one, which is the difference between being named and being described vaguely as a local provider.

NEED A CLEARER PLAN?

Let’s turn your next move into momentum.

Talk to us →

contact

hello@flockleads.com

Reply within 24 hours

REMOTE

Remote-first

Serving clients worldwide

All meetings via Teams or Google Meet

contact

hello@flockleads.com

Reply within 24 hours

REMOTE

Remote-first

Serving clients worldwide

All meetings via Teams or Google Meet

contact

hello@flockleads.com

Reply within 24 hours

REMOTE

Remote-first

Serving clients worldwide

All meetings via Teams or Google Meet