A website for a restaurant: what it is really for

Someone is standing on a corner at 7pm deciding where to eat. They have your name from a friend and a phone in their hand. In the next forty seconds they want three things: are you open, where exactly are you, and what do you serve. A restaurant website that answers those three on the first screen has done most of its job — and most restaurant websites do not.

The first screen, and nothing clever on it

• Today’s hours, as text. Not an image, not "see our Facebook", and correct on holidays — a wrong closing time costs a table and a review.
• Address that can be tapped to open maps, plus one line about parking or the nearest transit stop.
• Phone number that dials when tapped.
• The two actions most people came for: see the menu, and book or order.

Everything else — the story of the chef, the interior photography, the press quotes — is worth having and belongs below that. It sells the second visit, not the first one.

The menu is a web page, not a PDF

A PDF menu is the single most common mistake in the category, and it fails in a specific way: it opens as a zoomed-out document on a phone, it cannot be searched, screen readers struggle with it, and search engines cannot read your dishes as dishes. Someone looking for "vegan ramen near me" never finds the ramen inside your PDF.

The fix is unglamorous: menu items as text on a normal page, grouped by section, with prices. Add dietary markers, because they are the filter people actually apply, and photographs of the eight dishes you want to sell rather than all forty. And keep it current — a menu with last season’s prices generates arguments at the table, not sales.

Direct ordering versus the delivery apps

This is the part with real money in it. Marketplace delivery commissions run roughly 15% to 30% depending on the tier a restaurant is on, and operators consistently report the effective cost landing nearer 30–40% once fees, promotions and marked-up pricing are counted. Against a typical independent margin of 3–5%, an order through an app is close to margin-free.

Meanwhile, roughly two thirds of customers say they would rather order from a restaurant’s own site or app — many of them explicitly to support the restaurant. That preference is the asset. The apps are a discovery channel; your website is where the repeat order should happen.

• Put your own ordering link where the app buttons usually go, and make it the visually obvious choice.
• Give people a reason to switch: a dish only available direct, no service fee, loyalty points. It does not need to be a discount.
• Print the direct link on the takeaway packaging and the receipt. The second order is the one worth competing for.
• Keep the checkout short. A direct order that takes longer than the app just proves the app’s point.

Do not delete the apps. Use them for discovery, and treat every delivery bag as an advertisement for ordering direct next time.

Bookings

Two workable options. A booking platform brings its own audience and costs a monthly fee or a fee per cover; a simple request form on your own site costs nothing and requires someone to confirm. For a small place with predictable service the form is genuinely fine — as long as confirmations go out fast, and the form asks for four fields, not ten.

Whichever you choose, put the booking option on every page, not only the homepage. Most visitors land somewhere else entirely — usually the menu.

Google is half of your website

For a restaurant, the business listing often gets more views than the site. Treat it as part of the same job: hours that match the site, menu link, photos updated this year, and replies to reviews. Your website’s role is to be the source of truth behind it — the place where the menu is complete and the hours are right — and to catch the people who tap through.

Mark the practical facts up in structured data so search results can show your hours, price range and menu directly. It is invisible work with a visible result.

Phones and photographs

Assume the phone. Most restaurant traffic is mobile, often on a poor connection outside your door, and the two things that lose those visitors are a page that takes six seconds to load and text you have to pinch to read.

Photography is the one budget line worth defending. Stock images of anonymous food are actively harmful in this category — people are trying to see your room and your plates. A half-day with a photographer, including a few shots of the space and the team, outlives three redesigns.

Responsive design: what it means when most visits are on a phone

What it costs

Ranges are the same as for any small business site: a builder subscription and your own evenings, a freelancer in the low thousands, an agency well above that. What is specific to restaurants is where the money should go — photography and a menu that is easy to keep current, ahead of animation and clever layout.

How much does a website cost, honestly

The mistakes worth naming

• A PDF menu, and a menu that has not matched the kitchen since spring.
• Music that starts on its own. It is 2026 and it is still happening.
• Hours only on social media, so anyone who finds the site first gets the wrong answer.
• An address as an image, which cannot be copied into a maps app.
• Reservation and order buttons that appear only on the homepage.
• A gallery of forty photos and no view of the menu without scrolling past them.

If you are deciding what a single page for one offer should contain — a set menu, an event, a new location — the structure question is a different one, and there is a separate piece on it.

Landing pages by niche: what changes and what never does