· 8 min read
How Fast Should a Small Business Website Be?
The three speed numbers Google publishes, what each means in plain terms, how to check your own site in minutes, and what usually slows it down.
By Tanner Goins
“Fast” is one of those words everyone uses about websites and nobody defines. Your web developer says your site is fast. A marketing agency’s audit says it’s slow. Both might be right, because they’re measuring different things.
There is a useful definition, though, and it comes from Google. It has three numbers, each with a published target, and you can check your own site against them for free. Here’s what they mean, how to read them, and what to do when your site misses.
The short answer: under 2.5 seconds, on a phone
If you only remember one number, make it this one: the main content of your page should appear within 2.5 seconds of someone tapping your link. That’s Google’s threshold for a “good” loading experience, and it’s measured on real visitors’ devices, including the cheap phone on a weak signal in a parking lot.
That’s the first of three measurements Google calls Core Web Vitals. Together they describe what a visitor actually feels: how long they wait, whether the page responds when they tap, and whether things jump around while they’re trying to read.
The three numbers, in plain terms
Largest Contentful Paint (LCP): how long until the page shows up
LCP measures how long it takes for the biggest thing on the screen to appear. On most small business sites that’s the large photo at the top, or the headline next to it. It’s the moment a visitor stops looking at a blank or half-drawn screen and sees your page.
- Good: 2.5 seconds or less
- Poor: more than 4 seconds
Interaction to Next Paint (INP): how quickly the page reacts
INP measures the delay between a visitor doing something (tapping your menu, pressing “Call now”, opening a photo) and the page visibly responding. You’ve felt a bad INP: you tap a button, nothing happens, so you tap again, and then both taps land at once.
- Good: 200 milliseconds (a fifth of a second) or less
- Poor: more than half a second
INP replaced an older responsiveness measure in March 2024, so an audit from before then may show a different number here.
Cumulative Layout Shift (CLS): whether the page stays still
CLS measures how much the page jumps around while it loads. The classic example: you go to tap “Directions”, a banner loads above it, everything shifts down, and you tap “Order online” instead. CLS is a score rather than a time, where zero means nothing moved.
- Good: 0.1 or less
- Poor: more than 0.25
Why “75th percentile” matters
Google doesn’t grade your site on its best visit or its average one. A page passes when at least 75% of visits meet each target, measured separately for phones and computers. That’s deliberate: a site that’s quick on your office Wi-Fi but slow for a quarter of your customers doesn’t pass.
This is also why your own experience of your site is a poor guide. You visit it often, so your phone has most of it saved already, and you’re probably on a good connection. Your customers, arriving from a search on the way somewhere, often aren’t.
What a slow site costs
Speed isn’t a vanity score. When a page is slow, some visitors don’t wait for it. They go back to the search results and tap the next business, and you never know they were there. That’s the cost I wrote about in five signs your website is costing you customers: it doesn’t show up as a complaint, only as a phone that doesn’t ring.
The best evidence is from businesses that changed only their speed and measured what happened. Vodafone ran a test in Italy where half its visitors got a faster version of the same page: a 31% improvement in LCP led to 8% more sales. Google keeps a collection of case studies like it, from retailers, publishers, and travel sites.
Those are large companies, and your numbers won’t match theirs. But the direction holds for a plumber or a bakery too. A visitor who leaves before the page loads can’t call you.
Speed also plays a part in search. Google uses Core Web Vitals as part of how it evaluates page experience, alongside many other signals. Being fast won’t make up for a page that doesn’t answer the searcher’s question, and relevance matters much more. But between two local businesses with equally good pages, the one that loads and behaves well isn’t giving anything away. If local search is where your customers come from, pair this with the basics in local SEO for small businesses.
How to check your own site
You don’t need to hire anyone to find out where you stand.
- Go to PageSpeed Insights and enter your home page’s address.
- Look at the top section first, “Discover what your real users are experiencing.” This is data from real Chrome visitors over the last 28 days, and it’s what the thresholds above apply to. It shows LCP, INP, and CLS, each marked green, orange, or red, and whether the page passes overall.
- Switch between the Mobile and Desktop tabs. Mobile is usually worse, and for most local businesses it’s where most visitors are.
- Scroll down to the “Diagnose performance issues” section. That part is a simulated test of a single load, scored out of 100. It’s useful for finding causes, but don’t panic over the score alone; the real-visitor numbers are what count.
- Repeat for the one or two pages that matter most after the home page, like your services or contact page.
If the top section says there isn’t enough data, that’s normal for a small site: Google only reports real-visitor numbers once a page has enough traffic. Use the simulated test as your guide instead, and if you’ve set up Google Search Console, its Core Web Vitals report will show field data as it builds up.
For a sense of what’s possible: after I moved this site to static hosting, real visitors’ LCP at the 75th percentile was 292 milliseconds over a day of visits, and every page measured had a good LCP, INP, and CLS. I wrote up how that move worked. A brochure site for a small business can be in that range too, because it’s mostly text and photos that don’t change from visit to visit.
What usually slows a small business site down
When a site misses these targets, the cause is almost always one of a handful of things.
Photos that are far bigger than they need to be. A photo straight off a modern phone can be several megabytes. Shown at the top of a page on a phone screen, it needs a fraction of that. Oversized images are the most common reason for a poor LCP, and also the easiest to fix: resize them, and save them in a modern format like WebP or AVIF.
Too many add-ons. Every chat bubble, social media feed, review widget, popup, and tracking tag is extra code the visitor’s phone has to download and run. Each one seems small. Together they’re often the reason a page feels sluggish when you tap it, which is what INP measures. Go through what’s installed and remove anything you aren’t actively using. If you can’t say what a tag tells you, you probably don’t need it; my post on the analytics that actually matter can help you decide what to keep.
Slow hosting. Before your page can start loading at all, the server has to answer. On cheap shared hosting, or on a site that builds every page from scratch on each visit, that wait alone can eat most of the 2.5 seconds. Pages that don’t change for each visitor can be built ahead of time and served as plain files, which is about as fast as hosting gets.
Things that load late and push the page around. Images without their size set, cookie banners, and promotional bars that appear after the page has drawn are the usual causes of a poor CLS. The fix is to reserve space for them up front, so the page doesn’t move when they arrive.
Heavy themes and page builders. Some templates load code for every feature they offer, whether your site uses it or not. That’s one of the hidden tradeoffs I covered in custom websites vs. templates. A theme that looks great in the demo can be slow with your content in it.
Most of these aren’t one-time fixes. Sites slow down gradually: a plugin here, a bigger photo there, a widget a vendor asked you to add. Checking PageSpeed Insights every few months, or after any big change, catches it early.
A realistic goal
Aim to pass all three Core Web Vitals on mobile for your most important pages. That’s a clear, published bar, and it reflects what your customers actually experience. Once you pass, a faster score is nice but has diminishing returns; your time is better spent on what the page says.
If your site misses and you’d rather not dig into image formats and hosting yourself, this is exactly the kind of upkeep a managed website covers: someone else watches the numbers, fixes what slips, and keeps the site fast as it changes. Either way, run the test this week. If you’d like a second opinion on what it shows, send me your results.