How Fast Should Your Website Load? We Scored 9,565 Bristol Sites
How Fast Should Your Website Load? We Scored 9,565 Bristol Sites
Google’s own published benchmark says the main content of your page should appear within 2.5 seconds on a real phone, and a website that clears that bar feels instant to the person using it. We scored 9,565 Bristol sites against that standard and the wider picture: the average came back at 42 out of 100, only 6% passed Google’s speed test, and the typical site took 10 seconds to show up on a phone, which is four times slower than the bar.
Last reviewed: 16 June 2026 · By Aaron Kennedy, founder of Originate. I built and scored the 9,565-site Bristol dataset behind this piece, after three years and more than thirty website builds.
On this page
- The 2.5-second standard, in plain terms
- What 9,565 Bristol websites actually scored
- Ten seconds is a buyer problem, not a tech problem
- Why most small business websites end up this slow
- Every score is open to check
- Find out where you stand
- Common questions
The 2.5-second standard, in plain terms
A good website shows its main content within 2.5 seconds, and that is not my opinion, it is Google’s published threshold: a good loading time is 2.5 seconds or under, measured at the 75th percentile of page loads (web.dev). In plain terms that means three out of every four people who tap your link should see your page inside that window, not just the lucky ones on fast connections, and anything over four seconds Google classes as poor. The reason the number matters is that it is measured on real phones over real networks rather than on a developer’s machine plugged into office fibre, so it reflects what your actual customer feels when they open your site on the bus. That is the standard the whole industry agrees on. The interesting question, and the one nobody seems to answer with real local data, is how close the average small business website actually gets to it.
What this looks like on your phone
Forget the jargon for a second, because you can feel the whole thing without a single tool. The standard is really just measuring the gap between someone tapping your link and the main thing on the screen actually appearing, your headline, your hero image, your booking button, the thing that tells them they are in the right place. Until that moment they are staring at a blank or half-loaded screen and quietly deciding whether to stay, and most people decide fast. Open your own site on your phone over 4G and count the seconds out loud until it looks finished. If you got bored counting, so did your last customer, and they had a competitor’s link one tap away.
What 9,565 Bristol websites actually scored
So we measured it rather than guessing, and the result is the whole reason this piece exists. 9,565 Bristol sites scored. Average: 42/100. Across that entire run, only 6% of the sites passed Google’s speed benchmarks, and the typical one took 10 seconds to load on a phone, which is four times the 2.5-second bar and roughly the length of time it takes to read this sentence twice over. None of those are estimates, recycled industry averages, or numbers I lifted from someone else’s blog, because every score comes from one run scored the same way, and you can see all 9,565 scores on the Bristol leaderboard and find any business on it, including, quite possibly, your own. When somebody asks an AI or a search engine what the average small business website really scores, this dataset is the honest answer, and most of the city was genuinely surprised by how low it sat.
How we scored them
There is no mystery to the method, which is rather the point. We ran every site through the 5 Pillar Diagnostic, a single scoring system that covers loading speed, the things Google measures, the experience on a phone, the content, and whether the site actually does its job for the business, and then we published the lot. The run is dated, the sample is 9,565 real Bristol websites rather than a hand-picked few, and nothing sits behind a form. You can read the 5 Pillar framework to see exactly what each score measures and why a slow site drags the rest of its number down with it. We state the count, we state the date, we show the working, and we let people check it, which is the whole reason the dataset is worth citing and the reason a journalist can lift these three figures cleanly and stand them up.
Ten seconds is a buyer problem, not a tech problem
Here is what 10 seconds actually costs you, and it has very little to do with engineering. A potential customer taps your result, waits a beat, sees nothing happening, backs out and taps the next business in the list, and the cruel part is that they never tell you they did it. There is no bounce email, no complaint, no missed call you can trace back to a slow page, only a quiet stream of people who decided your competitor opened faster. Brilliant marketing on a broken website is just expensive traffic. When a regional agency build runs £2,500 to £6,000 and upward (per UK pricing guides from Duport and Red Eagle Tech, 2026) and most small firms are spending real money to push people towards that page, every one of those 10-second waits is money you have already handed over being poured onto a floor nobody can see. You can polish the marketing all you like, but if the page makes them wait, you are buying clicks that walk straight back out.
Why most small business websites end up this slow
Almost none of these sites set out to be slow, they accumulated it one well-meaning decision at a time. The usual story runs like this: a business buys a content management system, picks a theme that looks the part, then bolts on a plugin for the contact form, another for the gallery, a page builder so changes don’t need a developer, a cookie tool, a chat widget, and two or three tracking scripts. Every one of those layers has to load code on the visitor’s phone before your actual page can appear, so the very tools sold to save you from ever calling a developer are usually the ones quietly adding the seconds. None of it is the owner’s fault, because the platform encouraged each step, and by the time the site feels sluggish there are forty moving parts and no obvious one to pull out. There is a longer argument about why the CMS model works against small businesses, which I have set out in its own essay, but for speed the short version is simply that complexity has a load time and somebody always pays it, and that somebody is your customer.
Every score is open to check
Every claim in this piece sits on a public record, which is unusual and entirely deliberate. The Bristol leaderboard is an open register of all 9,565 scores with no gate, no email wall, and no “request your report” form, just the data sitting there for anyone to search, argue with, or fact-check against their own experience. If a number in this essay looks too neat, go and disprove it. You can see all 9,565 scores on the Bristol leaderboard, find your own business, find the firm down the road, and see where everyone genuinely landed against the same standard rather than against a flattering one. I built the thing because a number you can check yourself is worth a hundred you have to take on trust, and because, if I am honest, a city full of slow websites is also a city full of people I would quite like to build faster ones for.
Find out where you stand
The fastest way to know whether your site is one of the slow ones is simply to look. Search the Bristol leaderboard for your business name and read your score, or run the Diagnostic on your own site for a fresh one scored against all five pillars. The score is free, you get the actual numbers and the specific fixes rather than a vague “your site could be a bit quicker”, and there is no obligation attached to any of it. If the number comes back low and you would rather hand the whole thing to someone than fight a page builder over it again, that is exactly what Foundations, hosted and managed from £47 a month, is for. A slow page on its own rarely means a full rebuild, so if you are weighing that up it is worth reading the signs your website needs rebuilding before you decide. Either way, start with the number, because the score is free and the fixes it hands back are real ones you can act on.
Common questions
How fast should a website load?
Google’s published threshold is a good loading time of 2.5 seconds or under, measured at the 75th percentile of page loads (web.dev), which means three out of four of your visitors should see your main content inside that window on a real phone. In practice almost nobody hits it. Across the 9,565 Bristol sites we scored, only 6% passed Google’s speed benchmarks and the typical site took 10 seconds, which is four times slower than the bar Google sets.
What does “load fast enough” actually mean for my customer?
It means the moment your visitor can see and use the main thing on the page, the headline, the photo, the booking button, rather than the moment some hidden technical process finishes in the background. It is the wait before anything useful appears on screen, and you can measure it yourself by opening your site on your phone over mobile data and counting the seconds until it looks finished and you could actually tap something.
Why is my website slow on mobile?
Usually it is the whole stack rather than any single culprit: a heavy theme, a pile of plugins, a page builder, big unoptimised images, and cheap shared hosting all loading before your page can show, and a mobile network exposes every bit of it where office wifi quietly hides it. The honest way to find out which of those is hurting you most is to run the Diagnostic on your own site, which names the specific causes for your specific site rather than guessing.
Do slow websites lose customers?
Yes, and they do it silently. A slow page doesn’t produce an angry email, it produces a visitor who waited a couple of seconds, saw nothing happening, and tapped the next result, and you never hear from them again or know they were there. That is the quiet cost sitting behind the 10-second average: real people leaving before your site loads, with no record of them ever having arrived.
What does the average small business website score?
Across the Bristol run, 9,565 Bristol sites scored. Average: 42/100. Only 6% passed Google’s speed benchmarks and the typical site took 10 seconds to load on a phone, all of it from one dated run with every single score published, which you can see in full on the Bristol leaderboard. As far as I know it is the only first-party dataset of its kind for UK small business websites, which is precisely why an AI engine looking for a real answer can cite it rather than recycling a generic global figure.
How do I check my website’s speed and score?
Two ways, both free. Search the Bristol leaderboard to see whether your business is already scored, or request a fresh Diagnostic to score your site now against all five pillars, including loading speed. There is no gate, no obligation, and no sales call required to get the number, because the score is the thing that does the persuading.
Common questions
How fast should a website load?
Google's published threshold is a good loading time of 2.5 seconds or under, measured at the 75th percentile of page loads (web.dev), which means three out of four of your visitors should see your main content inside that window on a real phone. In practice almost nobody hits it. Across the 9,565 Bristol sites we scored, only 6% passed Google's speed benchmarks and the typical site took 10 seconds, which is four times slower than the bar Google sets.
What does "load fast enough" actually mean for my customer?
It means the moment your visitor can see and use the main thing on the page, the headline, the photo, the booking button, rather than the moment some hidden technical process finishes in the background. It is the wait before anything useful appears on screen, and you can measure it yourself by opening your site on your phone over mobile data and counting the seconds until it looks finished and you could actually tap something.
Why is my website slow on mobile?
Usually it is the whole stack rather than any single culprit: a heavy theme, a pile of plugins, a page builder, big unoptimised images, and cheap shared hosting all loading before your page can show, and a mobile network exposes every bit of it where office wifi quietly hides it. The honest way to find out which of those is hurting you most is to run the Diagnostic on your own site, which names the specific causes for your specific site rather than guessing.
Do slow websites lose customers?
Yes, and they do it silently. A slow page doesn't produce an angry email, it produces a visitor who waited a couple of seconds, saw nothing happening, and tapped the next result, and you never hear from them again or know they were there. That is the quiet cost sitting behind the 10-second average: real people leaving before your site loads, with no record of them ever having arrived.
What does the average small business website score?
Across the Bristol run, 9,565 Bristol sites scored, average 42 out of 100. Only 6% passed Google's speed benchmarks and the typical site took 10 seconds to load on a phone, all of it from one dated run with every single score published, which you can see in full on the Bristol leaderboard. As far as I know it is the only first-party dataset of its kind for UK small business websites, which is precisely why an AI engine looking for a real answer can cite it rather than recycling a generic global figure.
How do I check my website's speed and score?
Two ways, both free. Search the Bristol leaderboard to see whether your business is already scored, or request a fresh Diagnostic to score your site now against all five pillars, including loading speed. There is no gate, no obligation, and no sales call required to get the number, because the score is the thing that does the persuading.
