12-table bistro, 5 months: calls from Google 18 → 96/mo
“The menu finally reads like a menu online. We stopped emailing a PDF to people who asked what we serve.”
Neighbourhood bistro · single site
Menu Board · Local Search
SEO for Restaurants That Fills Tables, Not Just Rankings
SEO for restaurants is about being the obvious answer when someone standing four blocks away types “ramen near me” at 7:40pm. We rebuild the parts of your site and Google profile that decide that moment: crawlable menus, correct categories, fresh photos, honest reviews, and a booking path that works on a phone one-handed.
Single sites · Multi-location groups · Delivery-first kitchensToday’s Service BoardPrix Fixe
To Start — Discoverability
Crawlable menu pagesHTML
Every dish as its own URL, with price text Google can read, served fast on mobile data.
Profile optimisationGBP
Primary and secondary categories, service attributes, holiday hours, geo-tagged photo cadence.
Main Course — Conversion
Near-me keyword mapLOCAL
Cuisine plus neighbourhood plus daypart, mapped to the page that can actually rank for it.
Review engineTRUST
A quiet, repeated ask at the right moment, plus replies that mention the dish and the district.
To Finish — Retention
Booking and direct order pathsCLICKS
Reserve, call, or order direct in two taps, with the third-party apps demoted to a fallback.
The empty tables problem
Why well-reviewed restaurants still lose the “near me” search
Good food is not a ranking factor. Six specific things on your site and profile decide whether a hungry searcher ever reaches the door.
01The menu is a PDFOne file, no headings, no dish URLs. Google indexes a placeholder and you lose every long-tail query for the plates you are actually known for.
02Wrong profile categoriesTagged as “American restaurant” when the searches are for “brunch near me” or “steak frites”. Categories drive the map pack more than most owners realise.
03Photos from 2021Stale imagery kills click-through on the profile. Diners judge the room, the plating, and the light before they judge the menu.
04Four locations, one pageA single contact page for a group means Google cannot tell which branch serves which neighbourhood, so none of them rank strongly.
05The delivery app owns the trafficMarketplace listings bring orders but not a customer list, and the commission is paid on every repeat visit you could have owned.
06No dish-level contentNobody writes about your signature dish, your sourcing, or your wine list, so there is nothing for an informational search to land on.
What you get
A restaurant site built for the way people actually choose where to eat
Most diners decide in under four minutes, on a phone, between two or three tabs. We build for that minute, not for a brochure.
- Dish and cuisine keyword researchSignature plates, dietary terms, daypart modifiers, and the neighbourhood vocabulary your guests use instead of your brand name.
- Menus that are part of the siteStructured headings per course, real prices as text, allergy notes, and a separate URL per menu so seasonal changes do not overwrite last year’s rankings.
- Local SEO For Small BusinessCategories, attributes, hours including holiday exceptions, reservation links, and a photo publishing rhythm that keeps the listing warm.
- Location pages that behave like separate restaurantsEach branch gets its own address, menu notes, parking detail, and staff-written copy instead of a template duplicated four times.
- Technical SEO and Core Web VitalsCompressed hero imagery, no render-blocking fonts, clean internal linking between cuisine pages, and schema that ties the site to the physical address.
- Review and reputation workflowA repeatable ask after the meal, replies that name the dish, and a process for handling the one-star week without panicking.
- Direct booking and orderingReserve, call, and order buttons above the fold on mobile, with tracking so you can see which path actually produces covers.
Daypart search
The same kitchen, three different searches
Lunch, dinner, and weekend brunch are not one audience. The queries, the device, and the patience level all change, so the pages and the profile notes should change too.
Daypart
What searchers type
What we publish
Where it lands
Lunch
“lunch near me”, “sandwich takeaway [district]”
Speed-first menu, pickup timing, pre-order link
Menu page plus profile order button
Dinner
“best [cuisine] [city]”, date-night phrasing
Dish pages, wine notes, booking widget above the fold
Cuisine hub plus reservation page
Weekend
“brunch [district]”, “bottomless”, “walk-in”
Wait-time expectations, family notes, parking detail
Branch page plus review replies
If your kitchen closes between services, say so in the profile and on the site. Nothing damages a map-pack conversion faster than a family arriving at 3:15pm to a locked door they were told was open.
How the work runs
Five passes, roughly one service cycle each
You will not be asked for a content calendar you cannot maintain. Each pass is small, visible, and tied to a number you already track.
PASS 01
Pre-service audit
Crawl the site, pull the profile data, check menu files, and list every place where a hungry searcher hits a dead end.
PASS 02
The keyword map
Group queries by dish, cuisine, district, and daypart, then assign each group to one page with one job.
PASS 03
Menu rebuild
Move the PDF into structured HTML, add schema, and keep a stable URL for each seasonal menu.
PASS 04
Profile and photo rhythm
Fix categories and attributes, then publish a steady trickle of real room and plate photography instead of one annual dump.
PASS 05
Measure the covers
Calls, direction requests, booking completions, and order starts, reviewed monthly against the dayparts you care about.
Where a group runs several sites across one metro, we also coordinate the branch architecture so the pages reinforce each other. That is standard practice inside Google Business Profile Optimization, and it is the difference between four listings that compete and four that cover a map.
Proof, not promises
What changed for kitchens like yours
Numbers below are monthly aggregates from client reporting, anonymised at the client’s request. No ranking guarantees are offered or implied.
4-branch taqueria, 7 months: map-pack views 2.1k → 9.4k/mo
“Each branch got its own page and its own hours. The late-night location stopped dragging the others down.”
Taqueria group · four locationsBakery cafe, 6 months: direct orders 5 → 23/mo
“Same products, own checkout. The delivery app is still there, it is just no longer the only door in.”
Bakery cafe · pickup plus local delivery
Scope
Three ways to work together, quoted by scope
No price list, because the honest variable is how many menus, branches, and languages you have. Tell us that and we will quote the scope in writing.
Single site
One kitchen, one map
- Menu rebuild and dish page structure
- Profile categories, attributes, hours
- Review reply templates
- Monthly call and direction report
Multi-location
Branch architecture
- All of the single-site scope
- Branch pages with unique local detail
- Profile management across listings
- Internal linking between districts
Direct-order focus
Own the repeat guest
- Ordering path and checkout review
- Dish and bundle content
- Email and profile offer alignment
- Attribution for direct versus marketplace
Questions we get asked
Restaurant SEO, answered plainly
How long before the phone rings more?
Profile fixes and menu rebuilds usually move calls and direction requests inside four to eight weeks, because those changes act on a listing that already has visibility. Competitive dish and cuisine terms take longer, often three to six months, and seasonal menus re-rank faster the second year because the URLs already exist.
Do you use spam tactics or paid review schemes?
No. No purchased reviews, no fake locations, no keyword stuffing in the business name. Those shortcuts get listings suspended, and a suspended listing costs more covers than any ranking gain is worth.
Can you work with our existing website?
Usually yes. If the platform cannot output crawlable menu pages or control URL structure, we will say so early and explain the trade-off before you spend on content that has nowhere to live.
How much of our time does this need?
Roughly two hours in the first month for menu facts, hours, and photo access, then about an hour a month for approvals and review replies. We write the copy; you correct the details that only the kitchen knows.
Should we drop the delivery apps?
Rarely. Keep them as a discovery channel while building a direct path for repeat guests. The goal is not to leave the marketplace, it is to stop paying commission on customers who already know your name.
How do we measure whether it worked?
Four numbers, monthly: calls from the profile, direction requests, booking completions, and direct order starts. Revenue per cover is yours to add. We report the ones we can influence and label the rest as context.
Do you handle menus in more than one language?
Yes, with hreflang set correctly per language and per region so a French menu does not compete with the English version for the same query.
Tell us your busiest night and your emptiest one
That single answer usually tells us which daypart to fix first. Send it over WhatsApp with your website and profile link, and we will reply with the first three things we would change.
WhatsApp +8618465139145 · Telegram @bdtg678