Bathroom Website Performance and Core Web Vitals

Bathroom websites ask a lot of a phone. A shower page may carry a gallery of polished chrome shot from every angle, a spec tab, three PDFs, a fitting video and a “complete the look” carousel. Category pages for taps or vanity units can run to hundreds of products with filters for finish, tap holes and mounting. Some sites that serve both trade and retail buyers swap prices after the page loads. Each of these can have a cost that Google’s speed measures can see.

This page covers how those measures work and where bathroom pages typically lose ground.

How Google scores a page

Google uses three Core Web Vitals, each taken from real Chrome visits at the 75th percentile. Phones and desktops are judged on their own.

Measure	In plain terms	Good	Poor
LCP	when the biggest image or text block in view has been drawn	up to 2.5 s	above 4 s
INP	how long a click, tap or key press waits for the screen to update	up to 200 ms	above 500 ms
CLS	how far things jump around without being asked	up to 0.1	above 0.25

Search Console and PageSpeed Insights report the field version of these numbers. Lighthouse is a lab test that helps you find causes. It cannot measure INP, so it reports Total Blocking Time as a proxy.

What I measured on UK bathroom product pages (October 2026)

I tested five UK bathroom product pages once each with PageSpeed Insights on 23 October 2026, in its mobile mode: a simulated mid-range phone on slow 4G.

Page	Performance score	Lab LCP	Total Blocking Time	Lab CLS
Retailer A	38	11.7 s	900 ms	0
Retailer B	60	9.0 s	310 ms	0
Retailer C	30	10.9 s	1,840 ms	0.001
Retailer D	61	7.0 s	310 ms	0

The four valid pages had a median score of 49 and a median lab LCP of about 10 seconds. Three had images without set dimensions and three were flagged for how the LCP image was discovered, which points at the main product photo. Lab layout shift was near zero everywhere, yet Retailer B’s site-wide field CLS was 0.23, a reminder that one logged-out test cannot see later shifts such as a trade price swap.

What these figures cannot tell you:

Throttled lab runs are slower than most real shopping visits, and Google’s assessment rests on field data, not lab LCP.
Only one of the four URLs had its own field data in the Chrome UX Report, and it passed Core Web Vitals. The other three showed only site-wide data, and one of those sites passed.
Four pages are far too few to rank anyone.
A fifth page was tested, but the run did not complete, so it is left out.

The 2024 picture, for the whole web

The 2024 Web Almanac performance chapter is based on mid-2024 data from all kinds of websites. It is not a bathroom study, but several findings are relevant to bathroom templates:

43% of sites had good scores on all three measures on mobile, and 54% on desktop.
CLS was good on 79% of mobile sites and 72% of desktop sites.
66% of mobile pages had at least one image without set dimensions, with a median of two per page.
35% of mobile LCP elements could not be found in the initial HTML.

Layout shift and the trade price swap

Many bathroom businesses sell to installers and to the public from one site. One pattern sends the retail price in the HTML, then a script checks the account and swaps in a trade price, an ex-VAT price or a stock message. When the new text is longer, wraps onto a second line or adds a “trade saving” badge, everything below it moves.

Google’s CLS guidance lists five causes, without ranking them: unsized images and video, web fonts that render at a different size from the fallback, third-party widgets that resize, content inserted above existing content, and content loaded asynchronously. A trade price swap is asynchronous content, and any badge it adds is inserted above what follows.

Render the right price on the server for logged-in trade accounts, rather than correcting it in the browser. Make sure those personalised pages are not cached and served to retail visitors.
Where a swap is unavoidable, reserve the space for the longer version, with the same font size and line count.
Hold space for delivery messages, finance offers and stock badges, which also tend to arrive late.
Give every product image width and height attributes, or a CSS aspect-ratio.

Shifts within 500 ms of a tap, click or key press are excluded, so opening a spec accordion is fine. A trade banner that appears two seconds after load is not.

The first photo of the tap (LCP)

On a product page the LCP element is typically the main product photo, often a tap or shower on white. On category pages it may be a lifestyle banner above the grid. Google’s LCP optimisation guide gives the fixes.

Keep the main photo in the HTML as an <img> with src and srcset, and never lazy-load it.
Watch category banners. A banner set as a CSS background is only found once the stylesheet has loaded. Use an <img>, or preload it with fetchpriority="high".
Use fetchpriority="high" on one image only. Mark offscreen gallery slides fetchpriority="low".
Size images to the layout and serve AVIF or WebP. A tap shown 400 pixels wide on a phone does not need the 3,000-pixel studio file.
Serve from your own origin, with a CDN in front if you use one, and set long cache lifetimes.

Big categories and filter taps (INP)

Tap and shower listings can run long, and bathroom filters are technical: finish, tap holes, mounting, pressure, valve type, width. Each filter tap is an interaction that INP measures, and it usually reports the slowest one in a visit.

Acknowledge the tap straight away, with a ticked box and a loading state, before fetching results.
Keep the DOM lean by paginating or loading more on request. A page that draws every product in the category gives the browser more to recalculate on each change.
If tiles show finish dots that swap the tile photo, make that swap cheap: change one image source and nothing else.
Fetch filter counts with the results, rather than recomputing them in the browser after every tap.

Spec tabs, PDFs and fitting videos

Bathroom product pages often carry more reference material than many other product pages. It does not have to slow the page.

Spec tabs should be plain HTML that is present on load. Hidden text costs little, and it also helps search.
PDFs are better linked than embedded. An inline PDF viewer can load a large script before anyone asks for it.
Fitting videos from a hosted player can bring a lot of script with them. Show a poster image with a play button, and load the player only when it is tapped. If the video sits above the fold, a <video> poster can become the LCP element, so keep that poster light.

Carousels, bundle builders and planners

“Complete the look” carousels can download images for products that many buyers never scroll to. Lazy-load them and keep them below the first screen.

Bundle builders, where the buyer picks a pan, seat and cistern, are interactive by nature. Keep each choice’s work small: update the price and summary first, and defer anything else. The same applies to bathroom planners and tile visualisers. Load them only when the buyer opens them, behind a still image. I describe one approach in how a 3D product configurator is built. For examples, see the bathroom configurator demos.

Testing a bathroom site

Log in as a trade customer on a phone and reload a product page. Watch the price area for movement.
Run a tap product page and a shower category through PageSpeed Insights, and read the field data first.
Open the Performance panel in Chrome DevTools and look for Layout Shift entries during load.
Check whether the category banner is an <img> or a CSS background.
On a mid-range Android phone, tap the finish, tap-hole and mounting filters one after another. Note any lag.
Confirm that fitting videos load the player only after a tap.
Search the product template for images without width and height.

Most of this work is invisible to customers apart from the speed. If you would like it done for you, see Core Web Vitals, Shopify speed optimisation or WooCommerce speed, caching and scripts. Related pages: bathroom ecommerce SEO and bathroom ecommerce.

Sources

web.dev, checked October 2026: Web Vitals and LCP for thresholds, Optimise LCP for product photos and banners, INP for filter taps, CLS for trade price swaps.
HTTP Archive, Web Almanac 2024: Performance, November 2024. Mid-2024 data for all websites, with mobile and desktop labelled; not specific to bathroom retail.
My own tests: PageSpeed Insights, 5 UK product pages per industry, 23 October 2026, for the bathroom results table above.