A one-page trades website has no fixed limit on photos, but the median mobile page already spends 911 KB on images alone, out of 2,164 KB total. Google's own performance guidance treats the single largest photo on a page, not the total count, as the one that decides how fast a page feels. The real question is never how many photos a gallery holds. It is how heavy the biggest one is, and how much sits behind it waiting to load next.
A homeowner standing next to a leaking water heater is not opening a trades website twice. One slow load on two bars of signal is the whole chance, and a gallery of ten photos uploaded straight from a camera roll can burn through it before the phone number ever renders. The fix has nothing to do with cutting photos a business needs to show. It is about what those photos cost in bytes, and which one has to load first.
- Median mobile page weight: 2,164 KB
- Images' share of that: 911 KB, about 42%
- Google's LCP target: 2.5 seconds for most visits
- Bighorn Roofing demo, gallery-led: 78 KB total
What a Photo Costs a Page
The median mobile web page weighs 2,164 KB before a visitor sees anything, and images account for 911 KB of that, close to 42 percent, per HTTP Archive's 2025 Web Almanac. That figure covers the whole web, not trades sites specifically, and a one-page site built around before-and-after photos or a storm-damage gallery tends to run heavier than the median rather than lighter.
Every one of those kilobytes has to cross the same connection the phone number does. A camera phone's default photo commonly runs several megabytes at full resolution, and uploading four or five of them straight into a gallery can outweigh the rest of the page combined. Those figures measure compressed delivery size, not the uncompressed original sitting in a phone's camera roll before it gets squeezed down for upload.
Why the Biggest Photo Decides the Whole Page's Speed
Google measures a page's loading experience with Largest Contentful Paint, the point where the largest image or block of text inside the viewport finishes rendering. Its own performance guidance sets the target at 2.5 seconds or less for most visits. For a one-page trades site, that largest element is almost always a photo: the hero shot behind the headline, or the first tile in a before-and-after set.
Compress every other image on the page and leave that one alone, and the page still measures as slow, because Largest Contentful Paint is timed against the biggest element, not the average one. For that one image, the same guidance recommends against lazy loading and against pulling it in later with a script, so the browser can start fetching it as soon as the page begins loading.
Is There a Photo Limit?
There is no fixed number. A page with fifteen compressed, correctly sized photos can load faster than one with four uploaded straight from a camera roll. What decides the speed is the total weight in the section a visitor sees first, not the count of images sitting in a folder.
One roofing demo's gallery stays large on purpose, because in that trade the photographs are the argument, and the whole build still measures 78 KB in total. What changes is that the hero loads first and the rest of the gallery loads only as a visitor scrolls to it, so the photo count never competes with the page's speed.
Where the Extra Weight Hides
The heaviest photo on most trades pages is rarely the one anyone notices. It is the hero image saved once at full camera resolution and never resized for the screen it fills, or a background slider that loads every slide's image on the first visit instead of only the one showing.
A gallery plugin often ships its own script and stylesheet alongside the photos, adding weight that has nothing to do with the images themselves. None of that shows up by looking at the page. It shows up in a page-weight report, and that report is the only way to know for certain where the kilobytes went.
What to Cut Before the Gallery Goes Up
The steps below move the most weight for the least effort, in the order a one-page trades site should take them. The sample trades builds each handle photos differently by trade, since a roofer's photos and a plumber's photos do different jobs.
- Resize every photo to the width it will display at before uploading it, not after.
- Keep the hero the lightest photo on the page, since it is also the one timing Largest Contentful Paint.
- Load a gallery's later photos only as a visitor scrolls to them, never all at once on page load.
- Run the finished page through a page-weight report and fix whichever single image is heaviest first.
Questions Trades Owners Ask About Photos and Page Speed
Each answer below assumes the hero image has already been identified as the page's single heaviest element.
Do before-and-after photos need to be full resolution?
No. A photo only needs to be as wide as the space it fills on screen. For most trades galleries that is under 1,000 pixels on a phone. Resizing before upload cuts the file size without any visible loss, since nobody zooms into a before-and-after shot.
Does a photo slider or carousel add more weight than a static gallery?
Usually, yes. Many sliders load every slide's image on the first visit rather than only the one showing. A five-slide hero built that way carries five times the weight of the one currently on screen. A static gallery that loads photos as a visitor scrolls avoids that entirely.
Should a trades site skip photos to stay fast?
No. Photos are what sell a roof replacement or a finished bathroom, and cutting them trades one problem for another. The fix is controlling what each photo weighs, not how many of them the page shows.
Does a heavy photo gallery affect search rankings as well as visitor patience?
Largest Contentful Paint is one of the Core Web Vitals Google documents as part of page experience, so a heavy gallery is a speed problem before it becomes a ranking one. Fixing the weight fixes both at once.
A page-weight report takes under a minute to run against a live page, and it names the single heaviest image instead of leaving an owner guessing which of ten photos to fix first. Running one after every new gallery catches the problem before a customer does. Getting phone number placement right still means nothing if the images loading above it eat the visitor's patience first.
The next photo a trades business adds to its site is worth resizing before it gets uploaded, not after a customer standing on two bars of signal has already given up waiting for it.