Problem
A well-reviewed restaurant that ranks for its own name — and nothing else. Every dish, occasion, and "near me" search goes to aggregators and competitors.
Intervention
Rebuild the site around the searches diners actually make — menu pages machines can read, occasion pages with substance, local schema, and speed that survives a phone on cellular data.
Evidence used
Query research on real dining-search patterns; before/after Core Web Vitals; structured-data validation; rank tracking on named dish and occasion queries.
End state
The demonstration shows the structural end-state — every menu item indexable, occasion pages answering real queries, and measurement wired to reservations and calls. As a labeled demonstration, it reports no client metrics.
This is a demonstration study. It applies our method to a representative scenario — it is not a client engagement, and it claims no client results. We publish these so you can judge how we think before any money changes hands.
The situation
A typical case on this coastline: a restaurant with a strong local reputation, full weekend covers, and a website that is effectively a business card — a homepage, a PDF menu, an Instagram link. Search for the restaurant’s name and it appears. Search for anything it sells — “birthday dinner San Clemente,” “fresh fish tacos near the pier,” “private dining south OC” — and the results belong to aggregators, listicles, and better-structured competitors.
The business is excellent. The signal is absent.
The survey
Our audit for a scenario like this asks three questions:
- What do diners actually type? Not “cuisine-forward coastal dining” — real queries: dish names, occasions, “open now,” “near me,” “with a view.”
- Who wins those queries today, and why? Usually: pages that load fast, name the dish in a heading, mark up their menu, and answer the occasion directly.
- What can this restaurant honestly claim? The dishes it genuinely serves, the occasions it genuinely hosts — matched to the queries it deserves to win.
The build
- A menu that machines can read. The PDF becomes structured pages — every section and signature dish in real HTML with Menu schema, so a search engine (or an AI assistant asked “where should we eat tonight”) can quote it.
- Occasion pages with substance. One page each for the occasions the restaurant actually serves well — not thin keyword pages, but honest pages with the private-room photos, the group-menu details, the booking path.
- Local structure. LocalBusiness and Restaurant schema from verified facts; hours and phone consistent with the Google Business Profile; directions and parking answered on-page.
- Speed for phones on cellular. Diners search on the sidewalk. The build targets sub-2.5-second LCP on a mid-range phone — image discipline does most of that work.
- Measurement on reservations. Calls tapped, reservation clicks, direction requests — wired from launch day, so the owner sees search turn into covers.
What this demonstrates
The method is the same one we apply to every build: find the real queries, earn them structurally, claim only what’s true, and measure what matters. When we publish verified client studies, they will replace demonstrations like this one.
Your scenario next
Every study starts the same way — with the survey. Yours is free.
Request a visibility audit