Your hotel website isn't getting bookings: seven checks before you decide to rebuild it

A hotel owner writes to say the website isn't bringing in bookings and asks for a quote to rebuild it. The first thing I answer, almost every time, is: wait. That sentence — "the website isn't bringing in bookings" — hides two completely different ones, and until you know which is yours, any quote is a gamble.

This article won't sell you a new website. It gives you seven checks you can run yourself in half an hour, without buying anything, to work out where the problem really is. For each one I tell you a single but important thing: whether it can be fixed on what you have, or whether it's a signal the site needs rebuilding.

First: two different problems with the same symptom

"Nobody comes" and "they come and don't book" look like the same problem seen from the till. They aren't, and they're repaired in opposite ways.

Telling them apart is free. Open Google Search Console and Google Analytics (GA4): they're free, and if you haven't installed them, that's the first job to do, before anything else. Look at how many people reach the site in a month and how often you show up in Google's results. Then compare that with the bookings and enquiries that actually come in.

If the visits are few, the problem is visibility: they don't find you. If the visits are many and the bookings zero, the problem is conversion: they find you, they come in, and they leave.

These are two different jobs. Rebuilding a beautiful site when the problem was that nobody visits it is the most expensive way to change nothing; buying ads to send people to a site where they can't book is the same mistake in reverse. The seven checks that follow start here: some are for people who have traffic and don't convert, one is for people who have no traffic. The first job is to know which half you're in.

The seven checks

They're in order of impact, and the first ones cost nothing to fix. Do them in order.

1. The price, for the same dates

Pick two dates at random. Look up your property on Booking, then on your own site, and compare the total for the same room and the same nights.

If your own site costs the same or more than the portal, nobody will ever book direct — and they'd be right: at the same price they pick the one they already know and that feels safer. It's probably the number-one reason for zero direct bookings, and it's also the one almost nobody checks. Since 2017 Italy lets you offer a better condition on your own channel; the full reasoning is in another article, here the check is enough.

This isn't a website problem. It's a decision of yours about rates. Don't rebuild anything: align the price, or add a concrete reason to book direct (check 5).

2. The book button, from a phone

Take your own phone and try to book a room in your property the way a guest would, starting from the home page. Count the taps it takes to see an available date with its price, and to confirm. Stop where you would stop.

If there's no real way to book — only a "request a quote" form — or if you get lost in the last few steps, the guest does exactly the same and goes back to the portal, where one tap is enough. A request form isn't a booking: you're asking them to wait for your reply while Booking gives them one immediately. Research on online shopping has confirmed it for years, and it's common sense before it's data: the more steps there are, the more people fall off along the way.

Almost always this can be fixed: the booking engine is added or replaced, and it lives on the site you already have. It becomes a rebuild signal only if the site is too old to host one, or if the journey breaks because the site was never built for phones (which is the next check).

3. Speed, on a phone

Put your site's address into PageSpeed Insights (free, from Google) and look at the mobile score, not desktop.

Google considers a site "good" if it loads within 2.5 seconds, responds to taps within 200 milliseconds, and doesn't jump around while it loads. Above 4 seconds to load it rates it as "poor". These aren't my thresholds: they're the official numbers Google measures every site against, and it's the same yardstick it uses to decide how to show you in results. A site that's slow on a phone loses people before they even see the rooms, and today most people searching are on a phone.

Here the answer depends on the cause. Enormous photos, poor hosting, too many plugins: fixable, without rebuilding anything. But if it's slow because it's built on an old base that doesn't adapt to phones, and every fix breaks something else, that's a rebuild signal.

4. The photos, next to the portal's

Open your Booking listing and your site side by side. Are the photos on your own site better or worse than the ones the portal cropped for you?

If your property looks worse on your own site than on the portal, the person goes back to where it looks better — and you've handed them the reason. The site should be where you show yourself at your best, not the fallback with the old photos.

Almost always this can be fixed: it's new photos and a decent gallery, not a rebuild. It's rarely a rebuild signal.

5. A reason to book direct

On your site, at the same price, what do you offer that the portal doesn't? A flexible check-in, the better room at the same rate, an easier cancellation, a small extra on arrival.

If your site and the portal are identical in everything but the logo, the guest picks the one they know. It doesn't take a discount: it takes a reason, said clearly where the person is deciding. The concrete ways to build one — without underselling yourself — are in the guide to direct bookings, which is the natural sequel to this check.

Fixable: it's copy and commercial policy, not structure.

6. The right languages

Look at where your guests come from — you know it from your property management system — and at which languages your site is in.

If you get Germans and French guests and the site is only in Italian, you're leaving those bookings to the portal, which does speak their language. It isn't a matter of elegant translation: someone who can't read the page won't book on it.

Fixable if the site is built to handle several languages cleanly. It's a rebuild signal if it was born single-language and adding one means keeping a parallel copy aligned by hand every time — because that's exactly where languages stop being updated, and a site with a two-year-old English page does more harm than one language fewer.

7. The shop window before the site: your Google profile

Search your property's name on Google from a phone and look at the panel that appears: photos, up-to-date information, and above all whether there's a direct link to your site.

For a great many guests that panel is the first place they see you, before the site itself. If it's incomplete, or if it only points to the portals, direct traffic never starts: you lose the booking one step before your own front door.

This isn't the site, and it's the quickest of all: the profile is free and fixable in half an hour. Do it anyway, before you spend a euro on the rest.

Rebuild or fix: how to read the answers

Read back over the seven checks and count. If the problems you found are price, photos, a reason to book, the Google profile — or a booking engine to add — the site doesn't need rebuilding: it needs fixing, and often for far less than a new site.

The signal that a rebuild is genuinely worth it comes when the checks pointing the same way are the structural ones: the site can't handle phones, can't handle more languages without becoming unmanageable, can't host a modern booking engine. When two or three of those coincide, patching costs more than starting again, and every repair opens another.

And if the verdict is "rebuild", it's worth saying it plainly: a ten-page website doesn't describe a job, it describes a container. Whoever built it years ago was answering a different need — being there — not this one, which is selling. Nobody got it wrong: what a site is asked to do has changed.

What you're leaving to the portals in the meantime

While you decide, one number helps put things in proportion: what the portal channel costs you today, and what you'd recover by shifting a share of your bookings to direct. Put your figures in below — the tool does the sum, without asking for your email.

That number is also the right way to read the cost of a site. For a place to stay the reference is the two-language version, around €790 – €1,090 a month; the booking engine and the channel manager carry their own subscriptions to their own suppliers, to be budgeted separately. The full reasoning on prices is in how much a website costs, and two real examples of how it went are in the Hotel Punta Scario and Residence Celeste case studies.

Where to start

Run the seven checks. If nearly all of them land in the "fixable" column, you've saved the cost of a new site by finding out you didn't need one. If they land in the "rebuild" column, now you know why — and it's no longer a gamble.

If you'd rather have an answer before deciding, there's the digital check-up: it costs €1,250, I go through the site you have now, your numbers and the properties you compete with, and leave you a document listing what to do in order of impact. If we then go ahead with the work, 50% comes off the quote.

Otherwise write to me and tell me which half you've landed in: nobody comes, or they come and don't book. From there, almost everything follows.