Before a parent pays for the first lesson, an assistant may already have named a few schools.

Two buyers and the questions that decide the booking

A driving school sells to two people at once. The learner, often a teenager and sometimes an adult new to the country, asks about the lessons: automatic or manual, how an instructor handles nerves, how soon the test can be. The parent, often the one paying, asks about the total cost, what it includes and whether the instructor picks up from school.

They ask in three shapes: for names ("which driving school should my daughter use"), about one school ("is [school name] good with nervous drivers") and about lessons ("how many lessons does a beginner need"). With web search on, an assistant answers from pages it fetches; otherwise, from what it learned in training. The first two shapes are where a school is named or left out. In the third, its site can be the page quoted.

Eight facts to state in words, and the markup that repeats them

Each is asked about by name, and only the school can supply it. Write it as text where the question belongs, the same way everywhere.

Reviews belong beside these facts as quoted text, which a parent can read and a crawler can find. The article on pass rates and reviews covers quoting them and publishing a pass rate honestly.

There is no DrivingSchool type in schema.org

Schema.org, the shared vocabulary structured data is written in, has no type for a driving school. We checked the full vocabulary while writing this page: schema.org/DrivingSchool does not exist, and the only driving-school term in it is DrivingSchoolVehicleUsage, a value describing how a car is used. Markup that declares "@type": "DrivingSchool" names a type that is not there.

Use LocalBusiness, with the properties Google's local business documentation reads: name, address, telephone, opening hours, geo and priceRange. In schema.org a LocalBusiness is also an Organization, so areaServed (the pickup area), knowsLanguage (the languages taught) and makesOffer (the packages, each an Offer with a price and a currency) apply too. If a driving-school type is added later, switch to it.

{
  "@context": "https://schema.org",
  "@type": "LocalBusiness",
  "name": "[School name]",
  "url": "https://www.example.com/",
  "telephone": "+1-555-0100",
  "address": {
    "@type": "PostalAddress",
    "streetAddress": "[Street and number]",
    "addressLocality": "[Town]",
    "addressRegion": "[State or province]",
    "postalCode": "[Postal code]",
    "addressCountry": "[Country code]"
  },
  "areaServed": ["[Town]", "[Neighborhood]"],
  "knowsLanguage": ["en", "ar"],
  "openingHoursSpecification": [{
    "@type": "OpeningHoursSpecification",
    "dayOfWeek": ["Monday", "Tuesday", "Wednesday", "Thursday", "Friday"],
    "opens": "08:00",
    "closes": "19:00"
  }],
  "makesOffer": [{
    "@type": "Offer",
    "name": "[Package name]: [n] lessons of [length] minutes",
    "price": "[price]",
    "priceCurrency": "[currency code]"
  }]
}

That block is JSON-LD, and everything in it must also be visible on the page. Google's introduction to structured data asks for that, and its general guidelines are clear that correct markup still guarantees no rich result.

What our crawl found, across all 40 trades

Between late August and early September 2026, our crawler fetched the homepage, contact page and about page of 11,060 small business websites across 40 trades and 90 metros in the US, Canada, Europe and the UAE, with no scripts run and robots.txt obeyed. 9,483 answered. Driving schools were one of the 40 trades. We publish no driving-school-only cut, so these figures describe the whole sample.

For 9,240 of the businesses we also put one buyer question from their trade and metro to Claude, with web retrieval on: 14.9% were named, and 17.7% among the 5,928 answers that gave a clearly differentiated list. Each business got one question, asked once, of one assistant, so read the figure as a direction and as Claude's alone. On inspection, the named businesses were the ones whose sites had answer-shaped content and a consistent identity.

Where the work on a school's site starts

1. Write the name, phone number and address in one form and use it on every page and listing. With several locations, give each its own page and state its address and number the same way everywhere.
2. Put the prices on a page of their own, as text. The prices page article has a template.
3. State the pickup area, the cars, the instructors and their certification, the languages and the road test, each where a learner would look for it.
4. Add LocalBusiness markup that repeats those facts, with one Offer per package.
5. Quote reviews as text, with the reviewer's first name, the date and the platform.
6. Run the crawler access check on your robots.txt and read your pages with scripts switched off. Which crawlers fetch for ChatGPT and Gemini has an article of its own.

Then measure again with the same questions once the fixes are live, and every month with monitoring, as our methodology sets out. None of this puts a school into an answer. It gives the answer something true to say.

Sources: our audit of 9,483 small-business sites, with its method · schema.org, LocalBusiness · Google, local business structured data · Google, introduction to structured data · Google Business Profile guidelines · Google, general structured data guidelines