Restaurant Website Grader: Free SEO & Guest Experience Audit

A restaurant website has a simple job: help a nearby guest decide, then make the next action easy. Search visibility matters, but a technically indexable page that hides the menu, hours or ordering path can still fail the restaurant. The grader above therefore uses four groups instead of pretending SEO is one score: 1. Search discovery: can search engines understand and index the basic page? 2. Guest utility: can a hungry person find the menu, hours and next action? 3. Local trust: does the page expose the location and useful restaurant structured data? 4. Technical foundation: does the initial page use HTTPS, mobile viewport basics and reasonable image accessibility? The score is a diagnostic, not a Google ranking score. ## What the grader actually does Kitch fetches the public HTML returned by the URL you enter and checks a defined set of signals. It does not: - log into the restaurant website

• crawl private networks
• run JavaScript like a full browser
• claim to measure Google's Core Web Vitals
• know whether a restaurant that does not take reservations should have a reservation button
• know whether the site's copy is persuasive simply because a keyword exists That limitation is intentional. A trustworthy audit says what it measured. ## The 100-point restaurant website framework ### Search discovery: 30 points The grader checks: - whether a noindex robots directive is present
• whether the page has a useful HTML title
• whether it has a meta description
• whether there is one clear H1 heading
• whether a canonical URL is declared These are not magic ranking switches. They are basic communication between the site and search engines. Google's own documentation recommends descriptive, concise title text and warns against vague or boilerplate titles. It also explains that Google may use a page's meta description when that description better explains the page than other available content. Sources: - Google Search Central: Influencing your title links in search results
• Google Search Central: Control your snippets in search results ### Guest utility: 30 points A restaurant page is not a generic brochure. Most visits have high-intent questions behind them: - What do you serve?
• Are you open?
• Where are you?
• Can I order?
• Can I reserve?
• Can I call you? The grader checks whether the initial HTML exposes signals for: - menu
• online ordering / pickup / delivery
• reservations
• tap-to-call phone
• hours A missing reservation path is not automatically a problem for a counter-service concept. That is why the result explains the check instead of silently treating every failed item as mandatory. ## Menu accessibility is an SEO and conversion issue A restaurant's menu is often its most commercially important content. A guest searching a cuisine, dish or restaurant name may be trying to answer a menu question before anything else. If the menu is only available as an old PDF, embedded in an inaccessible widget, hidden behind a click that a crawler cannot interpret, or split across conflicting pages, both people and search engines have less useful information to work with. A strong menu experience should usually be: - readable on mobile
• represented as real page content where practical
• easy to reach from primary navigation
• current
• clear about price and availability where relevant
• connected to an obvious ordering or reservation action when the restaurant offers one The grader only checks the first HTML for menu signals. A human audit should go further and inspect the actual menu experience. Use Kitch's restaurant food cost calculator when the problem is menu economics rather than menu discoverability. ## Local trust: structured data and location clarity Google supports structured data for local businesses, including restaurants. Correct structured data can help search engines understand business details, although structured data never guarantees a rich result or higher ranking. The grader checks for JSON-LD identifying the page as a Restaurant, FoodEstablishment or LocalBusiness, and separately looks for address/location signals. Official reference: Google Search Central: Local business structured data. For a restaurant, useful structured information can include: - business name
• address
• telephone
• URL
• opening hours
• price range
• cuisine where supported in the broader schema model
• menu URL The important operational principle is consistency. Structured data should describe the same real restaurant information guests see on the page. Do not use schema markup to invent information or hide contradictory hours. ## Opening hours have to be operationally trustworthy A technically perfect restaurant website with wrong hours is worse than a simple page with correct hours. Hours can change for: - holidays
• private events
• weather
• seasonal schedules
• kitchen close vs. dining-room close
• temporary staffing situations
• location-specific changes The website, Google Business Profile, ordering systems and social announcements can drift apart. That is not only a technical SEO problem. It creates calls, refunds, disappointed arrivals and bad reviews. The highest-value system is not one that merely has an hours field. It is one where the operator can keep the restaurant's public truth synchronized without hunting through several dashboards. ## Technical foundation: what this audit does and does not measure The grader checks four lightweight technical signals: - HTTPS
• width=device-width mobile viewport
• server-side HTML response time as observed by Kitch
• whether most initial <img> tags have non-empty alt attributes The response-time check is deliberately labelled HTML response, not “site speed.” A proper performance audit needs browser rendering and field or lab data. For that, use Google PageSpeed Insights and review Core Web Vitals separately. Google's current Core Web Vitals are: - Largest Contentful Paint (LCP)
• Interaction to Next Paint (INP)
• Cumulative Layout Shift (CLS) Those metrics cannot be responsibly inferred from a basic HTML fetch. ## Mobile matters because restaurant intent is often immediate Restaurant traffic is unusually tied to real-world moments: someone in a car, walking nearby, deciding where to eat, checking a menu at a table or ordering from a phone. A mobile restaurant audit should manually test: 1. Load the homepage on a normal cellular connection.

2. Find the menu with one thumb.
3. Confirm today's hours.
4. Tap the address and make sure it opens the correct map location.
5. Tap the phone number.
6. Start an order or reservation.
7. Verify the navigation does not cover the primary action.
8. Make sure text does not require pinch zoom.
9. Check that pop-ups do not block the whole screen.
10. Confirm the menu does not force a large PDF download. A page can pass the viewport check and still have a poor mobile experience. The automated grade is the start of the audit. ## What a restaurant title tag should do There is no universal perfect title length. Search engines may rewrite titles based on the query and page content. A useful restaurant homepage title normally communicates the identity and, where helpful, the local context. Examples: - Bar Isabel | Spanish Restaurant in Toronto

• North Shore Pizza | Wood-Fired Pizza in Burlington
• Mina's Bakery | Persian Bakery in North Vancouver Avoid stuffing every suburb, cuisine synonym and service into one title. The page should sound like a real business, not a keyword spreadsheet. ## What a restaurant meta description should do The meta description is an opportunity to make the search result useful and specific. It can include: - what the restaurant is
• neighbourhood/city
• what distinguishes it
• dine-in, pickup or delivery availability
• reservation or ordering cue Do not use one identical description across every location or page. Google may choose another snippet when page content better answers the query. A meta description is guidance, not a guaranteed display string. ## One H1 is a clarity test, not an SEO superstition Modern search engines can interpret pages with more than one H1. The grader still awards the heading check when there is one clear H1 because it is a useful editorial discipline for a restaurant homepage. The H1 should answer: what is this page about? If the site has zero H1s, the primary message may only be visual styling with no semantic heading. If it has ten, the hierarchy may be unclear. Treat the check as a content-structure prompt, not a penalty myth. ## Canonical URLs help when a restaurant site has duplicate paths A canonical link tells search engines which URL the site considers the preferred version of substantially similar content. This matters when: - tracking parameters create many URL versions
• HTTP/HTTPS or www/non-www variants are not perfectly redirected
• multiple platform URLs expose the same restaurant page
• ordering or menu content appears at more than one path The canonical should normally resolve to the real preferred public URL. ## Restaurant structured data: common mistakes ### Marking up information that is not visible or true Structured data should represent the business accurately. ### Using the wrong location Multi-location groups need location-specific addresses, phone numbers and hours. ### Leaving hours stale Machine-readable wrong information is still wrong information. ### Adding schema and assuming local SEO is solved Local search also depends on relevance, prominence, reviews, links, Google Business Profile quality, content and many other signals. ## What to fix first after a website grade Do not chase points in order of easiest implementation. Fix based on guest and revenue impact. A practical priority order is: ### P0: wrong or missing restaurant truth Fix immediately: - wrong hours
• wrong address
• broken phone
• broken ordering
• stale menu
• noindex accidentally blocking the site ### P1: the guest cannot complete the main job Examples: - menu buried
• no order/reserve CTA
• mobile navigation broken
• ordering opens the wrong location ### P2: search communication is weak Examples: - generic title
• no meta description
• no canonical
• unclear heading structure
• missing local structured data ### P3: refinement Examples: - better Open Graph image
• improved alt text
• richer internal linking
• content for cuisine/location/dish intent
• performance work after measuring it properly ## A website can pass the audit and still fail commercially The grader cannot know: - whether photography creates appetite
• whether the menu pricing is compelling
• whether the brand feels credible
• whether the booking system has poor availability
• whether the restaurant answers reviews
• whether the order checkout converts
• whether a competitor has stronger local authority That is why a 95/100 score should never be read as “your website is done.” It means the initial page contains most of the technical and restaurant-utility signals this particular audit checks. ## Website ownership vs. marketplace dependence A restaurant website has a different job from a marketplace listing. Marketplaces can create discovery and convenience. The restaurant's own site is where it can create an owned, canonical destination for: - current menu
• current hours
• brand story
• direct ordering
• reservations
• catering
• gift cards
• events
• first-party customer relationships If direct ordering economics are part of your website decision, use the DoorDash and Uber Eats commission calculator with the restaurant's actual merchant rates. ## Compare website builders based on the work you need done The wrong question is only “which website builder has the most features?” A better restaurant-specific evaluation asks: - How fast can the operator change hours?
• What happens when an item sells out?
• Does the menu sync with the operating system or need manual editing?
• How does online ordering affect margin?
• Does the site work with the POS already in use?
• Who owns the customer relationship?
• Can the restaurant control design enough to feel like itself?
• What does a change require during Friday dinner? See The Best Restaurant Website Builders (2026) for the full comparison framework. ## How the Kitch grader fetches sites safely The grader accepts public HTTP and HTTPS websites only. It blocks local/private network addresses, validates DNS before connecting, limits redirects, caps the amount of HTML downloaded and times out slow requests. It connects to the validated public IP rather than blindly following arbitrary user-supplied network locations. That matters because a public URL grader should not become a proxy into internal infrastructure. ## Source and methodology notes The grading criteria were designed from two sources: 1. official Google Search documentation for search and structured-data fundamentals

2. restaurant-specific guest tasks such as menu, hours, ordering, reservations, calling and location A passed check receives its listed points. A failed check receives zero. The weighting is Kitch's diagnostic framework, not Google's ranking algorithm. The grader analyses the initial server-returned HTML. Sites that render essential content only after JavaScript may therefore score differently from what a human sees in a full browser. That itself can be useful diagnostic information, but it is not proof the content is invisible to Google. ## FAQ ### What is a good restaurant website score? Use the score as a triage tool, not a benchmark. A lower score can reveal obvious gaps, while a high score only means the site passes the checks in this framework. It does not guarantee search rankings or conversion. ### Does the grader measure Core Web Vitals? No. It reports a basic server-side HTML response time but does not claim to measure LCP, INP or CLS. Use Google PageSpeed Insights and real field data for performance analysis. ### Does a restaurant need structured data? It is useful because it gives search engines explicit machine-readable information about the business. Google supports LocalBusiness and restaurant-relevant structured data, but markup does not guarantee a rich result or higher ranking. ### Should a restaurant menu be a PDF? A PDF can be printable, but it is often a weaker primary mobile menu experience because updates, navigation and accessibility can be harder. A live HTML menu is generally easier to keep current and integrate into the rest of the website. ### Can Kitch grade any website? The tool is optimized for restaurant websites. It can fetch other public HTML pages, but restaurant-specific checks such as menu, hours and reservations make the score less meaningful outside hospitality. ### Why did my JavaScript-heavy site miss something I can see in the browser? The grader analyses the initial HTML response and does not execute a full browser render. Treat the miss as a prompt to verify how critical information is delivered, not as proof that search engines cannot render it.