By 
What a restaurant website actually has to do
Most diners decide from the map listing, the photos and the menu, then open the website only to confirm hours, see prices or book. A restaurant website that makes those four things fast on a phone does most of its job.
That makes the menu the centre of the build. It should be real text, not a PDF or an image, so search engines can read dishes and prices and so you can change them in minutes.
- Menu as page text with dish names, descriptions, prices and dietary labels
- Opening hours, address, phone and a map that match your Google Business Profile exactly
- Table booking or enquiry form that sends to your own inbox and phone
- Direct ordering or a WhatsApp order button so you do not depend only on delivery apps
- Photos that are compressed so the page stays fast on mobile data
What to compare between companies
Quotes for restaurant sites look similar on the surface and differ in what sits behind them. Compare the written scope line by line.
| Area | Weak answer | Strong answer |
|---|---|---|
| Menu | Menu uploaded as a PDF | Menu as editable page text with Menu schema |
| Local search | Mentions SEO in general | Sets up and aligns Google Business Profile, citations and hours |
| Ordering | Links out to a third-party app | Direct ordering or WhatsApp ordering you control |
| Speed | No target given | Core Web Vitals checked on a real phone before launch |
| Ownership | Site stays on the agency's account | Domain, hosting and code in your name at handover |
| Changes | Every edit is a paid ticket | You can edit menu and hours yourself, with training |
Local search matters more than rankings for a single outlet
A restaurant is found by people nearby, so the map pack and your Business Profile do more work than a generic keyword ranking. A good partner will talk about categories, photos, attributes, review replies and consistent name, address and phone details before they talk about blog posts.
Ask how they will measure results. Calls, direction requests, bookings and orders are honest measures. Impressions alone are not.
Questions to ask on the first call
These questions separate teams that know restaurants from teams that sell the same package to every business.
- Which restaurant or cafe sites have you built, and can I see them on my phone?
- How will my menu and prices be marked up so they can appear in search?
- Who updates the site when a dish or an opening hour changes?
- What do I own at the end: domain, hosting account, design files and code?
- How do you handle bookings and orders without taking a commission?
- What is included after launch, and what is billed separately?
Red flags
Be careful with any company that guarantees a position on Google, hides the hosting account in its own name, or sells the same template to every restaurant on your street. Also be cautious of a quote with no mention of Google Business Profile, because for a restaurant that is often where the sale is won.
Score your shortlist
Give each company a score from 1 to 5 on every row, multiply by the weight and total the result. A scorecard stops a confident sales call from outweighing the evidence, and it gives you a record of why you chose.
| Criterion | Weight | How to score it from 1 to 5 |
|---|---|---|
| Restaurant work they can show | 20% | 5 = live restaurant sites you can open on your phone; 1 = only screenshots |
| Menu and local search setup | 20% | 5 = text menu, schema and Business Profile alignment in the scope; 1 = PDF menu |
| Direct ordering and booking | 20% | 5 = you keep the customer and the data; 1 = everything goes through a commission app |
| Speed on a real phone | 15% | 5 = tested on mid-range Android before launch; 1 = no test |
| Ownership and handover | 15% | 5 = domain, hosting and code in your name; 1 = agency holds the accounts |
| Support after launch | 10% | 5 = named contact and response times; 1 = pay per ticket |
What drives cost and timeline
Cost in a restaurant build follows four things: how many outlets or locations need their own page and profile, whether ordering and booking are built directly or embedded from a platform, how much photography and menu writing you need, and how often the menu changes.
A single-outlet site with an embedded booking tool is the smallest build. Several outlets, a custom ordering flow and a loyalty or gift-card feature add design, integration and testing time. Photography is often the largest hidden cost, so decide who pays for it before you compare quotes.
Our pricing starts at USD 10–40 per hour. The final charges and timeline are set after we confirm your requirements, and depend on your budget and the team members assigned to the project, so you receive a fixed written quote within 24 hours of a short discovery call rather than an open-ended hourly bill.
Illustrative scenario: a two-outlet cafe moving off delivery apps
This is an illustration of how a project could be scoped, not a client case study, and it states no results. A cafe with two outlets takes most orders through delivery apps and pays commission on each. The owner wants more direct orders without losing the apps.
A sensible scope: one page per outlet matching each Business Profile, a text menu with dish-level markup, a WhatsApp and web ordering flow for pickup and local delivery, and a short guide on keeping hours current. The apps stay as a channel, but regulars are invited to order direct.
Your first 90 days after launch
A launch is the start of the work. This plan keeps the first three months focused on what changes results.
- Week 1: confirm hours, address and phone are identical on the site, Business Profile and every directory.
- Weeks 2 to 4: add fresh photos and reply to every review, good or bad.
- Month 2: review which dishes and pages get visits and update the menu order and descriptions accordingly.
- Month 3: compare direct orders, calls and booking requests with the month before launch, and decide what to improve.
Common mistakes to avoid
These errors come up repeatedly and are cheaper to prevent than to fix.
- Uploading the menu as an image or PDF.
- Letting opening hours differ between the site and Google.
- Paying for ranking promises instead of local profile work.
- Launching without testing the ordering flow on a real phone.
- Leaving the agency in control of the domain.
Sources and further reading
Industry pages
Related services