Website builders, and why they rarely stretch to a restaurant

There is no best website builder

You spend an evening looking for the best website builder for restaurants. Ten listicles, a handful of forum threads, three comparisons that look suspiciously alike, and you are no further along. That is not because you search badly. It is because the question is wrong.

They all make a website. What they do not all do is what your place needs. A business that lives on takeaway wants ordering to run smoothly on its own address and Google to read the menu properly. A business where atmosphere is half the product wants big photographs and a site that feels like the room.

Two goals, so two different builders. Mix them up and you pay the difference six months later: a beautiful site with an order button that does not work, or an ordering site that looks like the one three streets away.

So do not start with the builder. Start with one sentence: what does this site have to have delivered by next month. Everything you compare after that, you hold up against that sentence.

What your site has to do before it looks good

A restaurant site has a short list of jobs that never comes up in a demo. They are dull and they decide almost everything.

Nearly all your traffic comes from a phone, often from someone hungry standing in the street. That person is not looking at your typeface. They want the menu, the price and the button, in that order, and they are gone in seconds if one of the three is missing.

A menu as a PDF is therefore dearer than it looks. Google reads it poorly, your guest has to pinch and zoom, and you export a fresh file with every price change.

Run this list over the site you already have before you pick a new builder. Sometimes the builder is not the problem, the menu is.

Three questions to ask every builder

Owners starting on a new site tend to ask the same three questions. Usually in this order.

How much of the design can I change myself?

This weighs heavily when the image of your place is part of the product. Think of an omakase bar, a chef’s table, a cocktail bar with a handwritten list. There the site cannot look like ten others, because the first thing you sell is an expectation.

If it weighs less with you, freedom is mostly a risk. Every button you can move, you can move to the wrong place. Freedom costs evenings too, and those appear nowhere in the price.

Will it bring me more orders?

Design sells nothing on its own. What does sell: being found on your own name and on your dish, a menu that loads at once, and a checkout with no detour. So do not ask whether the builder is pretty. Ask what happens between the first tap and the payment, and how many screens sit in between.

Ask straight after that where the payment happens. If a different web address appears halfway through, your guest is checking out with somebody else, and that somebody keeps their details.

What do other owners make of it?

Software marketing sounds the same everywhere, and the screenshots in a demo are never of a place on a Friday night. Trustpilot, Capterra and G2 are the counterweight: review sites for software, the way Google and TripAdvisor are for restaurants.

Read them as an operator rather than as a consumer, though. Filter for places your size, look only at recent reviews, and pay most attention to what people write on their way out. That is where the things the seller left out get written down.

The kinds of builder, side by side

There are roughly four kinds of builder a restaurant chooses between. They differ less in price than the sales pitches suggest, and more in who does the work and where your menu lives.

No row wins everywhere, so read the table left to right rather than top to bottom. Look for the row whose last two columns keep you awake least. That is nearly always the row that fits you.

The general builders: Squarespace, Wix and WordPress

These are the names you meet everywhere, and not without reason. All three work in the Netherlands, all three are well documented, and for all three you can find somebody who knows them.

Squarespace

Strong on templates that look tidy straight away, and kind to people who are not technical. Weak on everything specifically restaurant: menu management, variants, allergens, ordering. A sound choice for a site whose job is to say who you are, where you sit and when you are open.

Wix

More freedom than Squarespace, a large stock of templates and a shop full of add-ons you buy alongside. The flip side is the learning curve and the temptation to keep nudging things. It is easy to build a handsome site here on which the order button only turns up after three screens.

WordPress

Not a drag and drop builder but a content system, and the freest option on the list. Anything is possible, with hospitality themes and plugins for the parts you are not going to build yourself.

• a plugin for your menu, with prices and allergens
• a plugin for ordering and payment
• a plugin for bookings
• a plugin for speed and backups

Every plugin is another subscription and another risk. Different people maintain them, they update at different moments, and one day two of them fall out. You usually notice through the order button rather than through an error message.

Pick from these three when design weighs more than orders, or when you have somebody who genuinely maintains the site. Count that person’s hours in the price, including when that person is you.

The site builder that comes with your till

Plenty of POS suppliers now also ship a website with an ordering page. For a place that has the till anyway, that is the shortest route: one contract, one menu, one phone number when something breaks.

You pay for it in distinctiveness. These sites resemble each other, because they come out of the same generator, and there is little you can change beyond colours and photographs. For a chip shop with a fixed menu that is no problem. For a place trading on atmosphere, it is.

The knot you are tying matters more. From that moment your site hangs off your POS contract, so changing till means moving your site as well. Before you switch it on, ask which domain it runs on and what happens to your pages on the day you give notice.

Restaurant platforms, where the site is one part

The last kind is not really selling you a website. They sell a way of selling more directly, and the site is the visible part of that. The menu, the checkout, the repeat order and the messages to your guests sit in one system, because they act on each other.

The trade is clear. You get a template instead of a blank sheet. Colours, photographs, copy and order you can change, the layout you cannot. Against that, the layout has already been tried on other people’s menus, and an improvement that comes out of it is switched on for you too, without you doing anything.

It fits badly with a place where the site is part of the experience and people come for the evening rather than the dish. It fits well with pizza, sushi, poke, kebab and sandwiches, places where takeaway and delivery do most of the work.

Portano sits in this category: your own site and ordering page on your own domain, menu, variants and extras in one place, and your guests’ details in your system rather than in a platform’s.

What to test during the trial

Almost every builder gives you a trial or a demo. Use it as a test rather than a guided tour, because a guided tour always goes well. Take one dish with variants and build the path around it.

1. Put one dish on the menu, with two variants, an extra and an allergen note.
2. Then change the price and count how many places you have to change it in.
3. Order it from yourself on your phone and actually pay, with iDEAL.
4. Look at the address bar at the moment you pay and write down what it says.
5. Set your kitchen hours and try to order at a time when you are shut.
6. Time how long the menu takes in the street, without wifi, on one bar of signal.

Then do exactly the same six steps with the second builder, in the same order. Comparing only works if you try the same thing everywhere. Six steps cost you an hour per builder and save you a contract.

Finally, ask who you ring on a Saturday night when the order button is dead, and what response time comes with that. With most general builders the answer is a web form.

Before you sign and before you move

Starting a website is easy. Getting back out of one is the expensive part. So you agree the things that make a move expensive before you begin, not afterwards.

That last question is the dearest of the five. Move without your old addresses redirecting to the new ones and Google starts again with a site it does not know. Everything you built up drops away, in precisely the week you expected more orders.

Put three lines on paper this afternoon: what your site has to deliver, which two builders you are testing this month, and when the old site is allowed to go off. Once those three lines exist the choice makes itself, and you never have to read another listicle.