Get suggested when diners ask where to eat.

Diners ask assistants for the best tacos nearby, somewhere quiet for a date, or a kitchen still open after ten. Speak Local shows which places get named, and what the assistant reads before it picks.

Free, in under a minute, no email or card. It grades up to 25 pages across Local, SEO, AEO and GEO.

The short answer

How does a restaurant get recommended by AI assistants and Google Maps?

By giving machines the facts diners ask about, in text they can read: cuisine, dishes, prices, hours, reservations and dietary options, on your website and your Google Business Profile, kept current. Reviews that name dishes and occasions do the rest, because assistants summarize them when they recommend a place. Mark it up as a Restaurant, or the closer fit such as CafeOrCoffeeShop or BarOrPub, with a link to the menu.

Updated 9 October 2026

What do diners ask before they choose?

Diners ask about occasions, cravings and logistics. These six are worth tracking:

  • Best brunch near me
  • Quiet restaurant in your city for a date night
  • Gluten-free pizza in your city
  • Restaurants open after 10pm near me
  • Where do locals eat in your neighborhood?
  • Is your restaurant good for large groups?

These are examples to adapt. Put the ones that bring you business to ChatGPT, Perplexity and Gemini on a weekly or monthly round, and each answer is recorded as named, cited or absent, with whoever was named in your place. How AI answers are tracked

What do Google and AI assistants weigh for a restaurant?

They weigh what you serve, when you’re open and what people say about it.

A menu machines can read

A menu that exists only as a photo or a PDF is hard to quote. A plain HTML menu with dishes, short descriptions and prices is something an assistant can repeat when someone asks where to get a particular dish.

Hours that are right, especially on holidays

Wrong hours send diners to a closed door. Keep regular, holiday and special hours current on your profile and your site, and make sure the two agree.

Reviews that name the dish

Assistants summarize what reviewers say. “The birria tacos are the best in town” is a sentence that can end up in an answer; a star rating on its own isn’t.

Attributes and occasions

Outdoor seating, vegan options, reservations, private dining, live music. Set them on your profile and say them on your site, because they’re the words diners use in their questions.

Which schema should a restaurant use?

Restaurant, or the closer fit: CafeOrCoffeeShop, BarOrPub, Bakery, Brewery, Winery and IceCreamShop are all schema.org food establishments. Add what diners filter on: cuisine, price range, reservations and the menu.

Speak Local’s scanner reads every one of them as entity schema on your homepage.

Worth checking: that hasMenu points at an HTML page rather than a PDF, and that the hours in your schema change when your holiday hours do. A marked-up time that disagrees with your profile gives an assistant two answers to pick from.

Worth adding
  • servesCuisine in the words diners use
  • hasMenu pointing at an HTML menu page
  • acceptsReservations, or a link to where they’re taken
  • openingHoursSpecification, updated before every holiday

Check it with Google’s Rich Results Test or the Schema Markup Validator before it goes live.

Restaurant · one locationJSON-LD · example values

  "@context": "https://schema.org"
  "@type": "Restaurant"
  "@id": "https://www.example.com/#restaurant"
  "name": "Example Taqueria"
  "url": "https://www.example.com/"
  "telephone": "+1-555-0100"
  "address": 
    "@type": "PostalAddress"
    "streetAddress": "100 Example Street"
    "addressLocality": "Your City"
    "addressRegion": "ST"
    "postalCode": "00000"
    "addressCountry": "US"
  
  "servesCuisine": "Mexican" "Tacos"
  "priceRange": "$$"
  "hasMenu": "https://www.example.com/menu"
  "acceptsReservations": true
  "openingHoursSpecification": 
    
      "@type": "OpeningHoursSpecification"
      "dayOfWeek": "Tuesday" "Wednesday" "Thursday" "Sunday"
      "opens": "11:00"
      "closes": "22:00"
    
    
      "@type": "OpeningHoursSpecification"
      "dayOfWeek": "Friday" "Saturday"
      "opens": "11:00"
      "closes": "23:30"
    
  

Where should a restaurant start?

Do these in order. Most of the work takes an evening.

  1. Put the menu in HTML

    Keep the PDF for printing, but publish the menu as a page too, with dishes, descriptions and prices.

  2. Check hours everywhere before every holiday

    Your profile, your site, and any directory or booking app that shows them.

  3. Fill in your profile’s attributes

    Dine-in, takeout, outdoor seating, dietary options, accessibility, reservations: whatever’s true.

  4. Reply to reviews, especially the critical ones

    Briefly and graciously. Future diners read the reply as closely as the complaint.

  5. Track the occasion questions

    “Best brunch”, “date night” and “open late” in your neighborhood, on ChatGPT, Perplexity and Gemini.

How does Speak Local help a restaurant?

Every plan includes all of it. See the full platform for how the pillars combine into one score.

AI answers: diners’ questions put to ChatGPT, Perplexity and Gemini on a schedule, with the places named instead of you and the sites each answer cited
A map grid: “near me” searches run from up to 49 points within a mile or more of your door
Reviews: every Google review in one inbox, replied to from here, with a notification when one lands
Review widgets: your real Google rating on your own site, never an average of only the good ones
Heatmaps on the menu page: clicks and scroll depth drawn over a real screenshot at phone, tablet and desktop width
Thin-page checks: pages with too few words to quote get flagged, which is where an image-only menu shows up

What it looks like for a restaurant

Six screens from the workspace of an example restaurant, as they look in the app. Choose one to see it.

Overview Every tool for one business on one screen: the audit score, traffic, rankings, backlinks and AI answers, each with its trend.
The Overview screen in Speak Local, for an example restaurant
Site Audit 32 checks across Local, SEO, AEO and GEO. Issues are sorted worst first, and each one shows how many points its fix adds to your score.
The Site Audit screen in Speak Local, for an example restaurant
AI Answers Your customers’ questions put to ChatGPT, Perplexity and Gemini. Every answer is kept word for word, with who was named and which pages were cited.
The AI Answers screen in Speak Local, for an example restaurant
Rankings The searches you choose to watch, with your position, the page that ranks and the last 30 days of movement.
The Rankings screen in Speak Local, for an example restaurant
Map grid One search checked from up to 49 points around the business, so you can see where you rank across town and not only at your own address.
The Map grid screen in Speak Local, for an example restaurant
Analytics Visits by source, with AI assistants counted on their own line, and the calls and form fills those visits turn into.
The Analytics screen in Speak Local, for an example restaurant

An example workspace for a fictional restaurant. The site audit and the competitors’ scores are real scans of test sites; the rankings, answers and traffic are sample data. Swipe the screen to see all of it.

Common questions

Is a PDF menu bad for SEO?
It isn’t invisible, since search engines can read many PDFs, but it’s a weaker page: harder to quote, awkward on a phone, and easy to let go stale. Keep the PDF for printing and put the menu on a web page too.
Which reviews do AI assistants read for restaurants?
Each assistant reads from different places. Gemini can draw on Google Maps; others lean on web pages, review sites and local articles. That’s why the same question can name different places in each, and why Speak Local asks all three.
How do we get into the “best of” lists assistants quote?
Local guides and best-of articles are among what assistants cite for restaurant questions. Speak Local records which sites each answer cited, so you can see which guides matter in your market before you pitch any of them.
Should we respond to bad reviews?
Yes, briefly. Thank them, acknowledge the problem without arguing, and invite them to get in touch. Future diners read the reply as closely as the review.
Does this work for a bar, brewery or coffee shop?
Yes. They’re all schema.org food establishments, and the same audit, map grid, reviews and AI answers apply. Track the questions your own customers ask, like “best patio bar” or “coffee shop with wifi”.

See where a restaurant like yours stands.

Audit, score, fix and monitor how you show up in Google and in AI answers.

Free, in under a minute. No account.  Or start a plan from $49/mo

Google SearchChatGPTPerplexityGeminiAI Overviews