Hotel website speed: how to test it and fix it

A fast hotel website loads its main content in 2.5 seconds or less on a phone, responds to a tap in 200 milliseconds or less, and does not jump around while it loads. Those are Google’s published Core Web Vitals targets. You can test yours for free in about two minutes, and for most independent hotels the biggest wins come from images, video, and third-party scripts, not from a new server.

This guide is for owners and general managers who run their own site or manage whoever builds it. It covers what to measure, what the numbers mean, what usually slows boutique sites down, and how to fix it in a sensible order.

StayStrategy

Help Travelers Find Your Property

StayStrategy guides hospitality operators on getting named when travelers ask AI where to stay, with a free course, visibility check, and practical guides.

Explore StayStrategy

TL;DR

Google’s Core Web Vitals set the targets: Largest Contentful Paint within 2.5 seconds, Interaction to Next Paint of 200 milliseconds or less, and Cumulative Layout Shift of 0.1 or less.
Test your homepage, a room page, and your booking page on mobile with PageSpeed Insights. The booking page matters most.
The usual culprits are oversized photos, hero videos, marketing scripts, and a slow booking engine widget.
Speed affects revenue: a Deloitte analysis tied a 0.1 second mobile gain to a 10.1% lift in travel conversion rates.
Speed is not a one-time fix. Recheck after every redesign, plugin, or new tracking tag.

Why speed matters for a direct booking site
The three numbers Google measures
How hotel sites compare
How to test your hotel website speed
What usually slows boutique hotels and rentals
A fix order that works
Prioritize what guests use
Keep it fast over time
Where to learn more
FAQ
Sources

Why speed matters for a direct booking site

A guest who lands on your site from Google, a map listing, or an AI answer has already narrowed the choice. The page has to confirm that choice quickly. If it stalls, the guest goes back to the search results or to an online travel agency (OTA) listing, and you pay commission on the booking you could have had directly.

The numbers behind this are consistent. A 2023 Deloitte study of mobile site data covering retail, travel, luxury and lead generation brands found that a 0.1 second improvement in mobile speed was linked to a 10.1% increase in conversion rates for travel. The same report found travel sites saw a 6.5% improvement in homepage bounce rates after performance work. The report also found that travel sites were especially sensitive to speed in the steps before checkout. That is exactly where a hotel guest is choosing dates and rooms.

Two cautions. These are correlations across many brands, not a promise for your property. And the percentages describe improvements measured in tenths of a second, so do not read them as “cut two seconds, gain 200%.” Use them as a reason to measure, not as a forecast.

There is also a plain usability finding. A 2025 study by Rita and colleagues on website quality and hotel choice found that functionality, meaning reliable and fast access to accurate booking information, influenced booking intentions more strongly than usability alone. For you, that means a working availability calendar and a rate display that loads quickly count for more than a beautiful animation.

Speed also touches how you are found. Google says it recommends that site owners achieve good Core Web Vitals for success in Search. If you are working on being named by AI tools too, a clean, fast, readable site is part of the foundation. The free AI visibility course covers what those systems read on your pages.

The three numbers Google measures

Core Web Vitals are three metrics, each measuring a different part of the guest’s experience. Google’s documentation and the web.dev explanation of how the thresholds were defined describe them.

Metric	What it measures	Good result
Largest Contentful Paint (LCP)	How long until the biggest visible element, usually your hero photo or headline, appears	2.5 seconds or less
Interaction to Next Paint (INP)	How quickly the page responds when a guest taps or clicks	200 milliseconds or less
Cumulative Layout Shift (CLS)	How much the page layout jumps while loading	0.1 or less

A page passes when the 75th percentile of real visits meets the target on each metric, according to Google. In plain terms, three out of four visits have to be good. If you miss the threshold, at least one visit in four is a poor experience.

What each one looks like on a hotel site

LCP is almost always the hero image or the hero video poster on a hotel homepage. If that one file is 4 MB, nothing else you do will fix the score.

INP suffers when the browser is busy running scripts: chat widgets, tracking tags, a heavy slider. The guest taps “Check availability” and nothing happens for half a second.

CLS shows up when a cookie banner, a late-loading booking bar, or a photo without set dimensions pushes content down just as the guest goes to tap. They hit the wrong button, or lose their place.

How hotel sites compare

Hotel sites have historically been slow. A 2019 benchmarking report, Don’t Make Your Guests Wait on Line, compared hotel reservation website load times with best practice and found averages of 12 seconds on desktop and 15 seconds on mobile, far over the 3-second mark cited in many consumer studies. That report is several years old and tools and phones have improved, so do not treat those figures as today’s averages. The lesson that holds up is that perception misleads. An owner on fast office Wi-Fi and a new laptop sees a different site than a guest on a phone with two bars.

The same research is a useful corrective to a common excuse. Owners often blame the guest’s connection. The report points instead to heavy site construction as the reason many hotel sites fail to load fast whatever connection the guest has.

How to test your hotel website speed

You need no paid tool. Use PageSpeed Insights, Google’s free tester. It reports on both mobile and desktop and combines two kinds of data: real-world measurements from the Chrome User Experience Report and a controlled lab test with suggestions for fixes.

Open PageSpeed Insights and enter your homepage URL. Wait for the report. Look at the mobile tab first. Most of your guests are on phones.
Read the field data at the top, if it appears. This is what real visitors experienced over recent weeks. A small property may not have enough traffic to show it. If so, rely on the lab data and repeat tests over several days.
Check the three Core Web Vitals. Compare each with the table above.
Scroll to the opportunities and diagnostics. These name the specific files and scripts that cost time.
Repeat for a room or unit page and your booking page. The homepage is rarely your worst page. The booking step often is.
Save the scores in a simple sheet with the date. You will need them to see whether a change worked.

Run each test two or three times. Lab scores move a little from run to run, so look at the pattern, not one number. Then test on a real phone on mobile data, not just Wi-Fi, and tap through to the booking page the way a guest would. A tool score cannot tell you that the date picker feels sticky.

GTmetrix is another free option some operators use for a second opinion. Whatever tool you pick, keep using the same one so your comparisons stay fair.

What usually slows boutique hotels and rentals

Boutique properties sell on atmosphere, so their sites carry big photos, full-width video, and animated effects. Those choices are the most common source of slowness. In my experience of how these sites are built, and in line with the performance guidance in the Deloitte report, the heaviest costs tend to come from a short list of causes.

Oversized images

A photographer delivers 6000-pixel files at 8 MB each, and they go straight into the gallery. The browser downloads all of it, then shrinks it on screen. Fix this first, because it usually pays the most.

Resize images to the largest size they actually display, often about 2000 pixels wide for a full-width hero and far less for thumbnails.
Use modern formats such as WebP or AVIF instead of large JPEGs.
Set width and height on every image so the layout does not shift.
Load images below the first screen lazily, so they download only when the guest scrolls near them.
Do not lazy-load the hero image. It is your LCP element and should load first.

Hero video and sliders

A looping background video looks lovely and can weigh more than the rest of the page combined. Options, from lightest to heaviest: a single still photo, a short compressed clip under a few megabytes that starts after the page is usable, or a full video. If you keep video, serve a still poster image first. Replace auto-rotating sliders with one strong image. Guests rarely wait for the third slide.

Third-party scripts

Every tag you add is code your guest’s phone must download and run: analytics, Meta pixel, Google Ads tags, chat, reviews widgets, heatmaps, pop-ups, translation tools. The research on hotel sites flags marketing and tracking scripts that load before the page content as a common cause of delayed interactivity. Audit them once a quarter.

List every script PageSpeed Insights shows under third-party code.
Remove any you cannot name a use for.
Ask whoever manages your site to load the rest after the main content, or on interaction, where possible.
Do not delete your analytics. Fix tracking first, then trim. Our guide to setting up GA4 for hotels and testing one booking shows how to keep measurement honest while you slim the site down.

Code and plugins

On WordPress and similar builders, themes and plugin stacks add stylesheets and scripts to every page, whether needed or not. Minifying CSS and JavaScript, removing unused plugins, and turning on caching all help. They are usually settings, not rebuilds.

The booking engine

Many independent hotels embed a booking engine from an outside vendor, and its speed is partly out of your hands. Still, you control how it is placed. Check three things: does the “Book” button open a new fast page or load a heavy embedded frame on every page; does the availability search respond quickly; and does the engine work well on a phone. If it is slow no matter what you do, raise it with the vendor with your test results in hand. If you are weighing how your Google listing connects to booking, the page on Google hotel booking links, quick link or live rates explains the setups.

Tour and activity operators should pay special attention here. Booking flows are linear and time-sensitive, so a calendar that hangs on a busy Saturday costs real sales. Test your booking step during your busiest hour, not just on a quiet Tuesday.

A fix order that works

Do the work in this order, because each step is cheaper and gives more than the next.

Compress and resize images. Start with the hero and the top of each room page.
Replace or defer the hero video.
Remove unused plugins and tracking tags.
Set dimensions on images and reserve space for banners so the page stops jumping.
Turn on caching and compression at your host or through your site platform.
Review the booking engine placement and test the full path on a phone.
Retest and record the scores.

Change one thing at a time where you can. If you change everything at once and the score drops, you will not know why.

A worked example

Say a 12-room inn tests its homepage on mobile and finds LCP at 5.1 seconds. The diagnostics point to a 6 MB hero photo and a 9 MB background video. The owner replaces the video with a still photo, exports the hero at a sensible width in WebP, and removes two old chat and heatmap tags. The retest shows LCP near 2.3 seconds. These figures are an illustration of the method, not a result from a real property. What matters is the sequence: find the largest file, shrink it, remove what is unused, measure again.

Prioritize what guests use

Do not spend a month polishing the “About us” page. Rank pages by their role in a booking:

The booking engine and availability search.
Room or unit pages, with rates and photos.
The homepage.
Everything else.

This follows what the Rita study suggests about functionality: guests reward fast, accurate booking information. It also matches the Deloitte finding that gains in the stages before checkout are what lift conversion. If you want to see where guests drop out of your path once the pages are fast, read our guide to hotel booking funnel optimization for direct reservations.

Short-term rental hosts who run their own site have the same priorities. A direct site only helps if it opens quickly and takes a booking without friction, which is part of the case in why every short-term rental needs a direct booking site.

Keep it fast over time

Speed slips. Someone uploads a 10 MB photo, marketing adds a pixel, a plugin updates. Deloitte’s analysts describe speed as a business metric that needs continuous monitoring and a mobile-first mindset, not a one-time fix.

Build a small routine:

Monthly: run PageSpeed Insights on your homepage, one room page, and the booking page. Log the scores.
Before any new tag or widget: test the page before and after.
After any redesign or new season’s photos: retest everything.

If you outsource your site, put the Core Web Vitals targets in writing and ask for a mobile report every quarter. You can read the numbers yourself, so you do not have to take anyone’s word for them.

Where to learn more

The full free course on getting your property named by ChatGPT, Gemini and Google’s AI starts at the AI visibility course. Fast, readable pages are one part of it, and it walks through the rest step by step. For Google’s AI answers specifically, see the hotel Google AI Overview visibility guide.

FAQ

What is a good load time for a hotel website?

Aim for your main content to appear within 2.5 seconds on mobile, the Largest Contentful Paint target in Google’s Core Web Vitals guidance. That should hold for at least 75% of visits. Many consumer studies cite 3 seconds as the point where people start leaving, so treat 2.5 as the goal and 3 as the ceiling.

How do I speed up my hotel website performance?

Start with images: resize them, convert them to WebP or AVIF, and lazy-load those below the first screen. Then replace hero video with a still or a short clip, remove tracking scripts and plugins you do not use, and set image dimensions so the page does not jump. Retest with PageSpeed Insights after each change.

Does hotel Wi-Fi track websites?

Generally the network does not log which pages you visit unless you use its sign-in page or other tracking tools, though unencrypted connections can be intercepted by other users. The FTC advises sending personal information only over encrypted HTTPS sites. Most tracking is done by websites themselves through cookies and pixels. This question is about guest Wi-Fi, which is a different matter from your own site speed, but it is a good reason to make sure your booking pages use HTTPS.

Is my site slow because of the guest’s internet connection?

Usually not by itself. Research on hotel reservation sites found that heavy site construction often kept pages from loading quickly regardless of connection. Test on a phone using mobile data and compare it with the PageSpeed Insights report to separate site bloat from a weak signal.