How to Speed Up a Small Business Website: A Practical Fix List

Most people find a local business on their phone, often in a car park, on a smoko break, or standing in a kitchen that has just started leaking. Their patience is measured in seconds.

When the page takes too long, visitors assume the business is as slow and fiddly as the site. Fair or not, speed is a trust signal.

Google also uses page experience signals, including Core Web Vitals, as part of how it evaluates pages. At the same time, Google is clear in its page experience guidance that good scores don't guarantee top rankings. Relevant, helpful content still matters more. So treat speed as a tiebreaker for Google and a dealbreaker for humans.

If your site already shows other warning signs, like a high bounce rate or phone calls drying up, our guide to the signs your website is costing you customers is a useful companion to this one.

The biggest speed mistake is starting in the wrong place: buying a premium plugin, flipping twelve settings, and ending up with a site that's slightly faster and noticeably broken. Work in this order instead.

1. Measure. Get a baseline on mobile, so you know whether anything you do actually helps.
2. Images. On most small business sites, they're the heaviest thing on the page by a long way.
3. Scripts and plugins. Chat widgets, tracking pixels, and forgotten add-ons pile up quietly.
4. Hosting and caching. Cheap hosting and no caching put a ceiling on how fast you can ever be.
5. Fonts, sliders, and layout. The polish layer that often causes slow loads and jumpy pages.
6. Real-world testing. A real phone on mobile data, not your office fibre connection.

You won't always need every step. Going in order simply stops you spending money on step four when the real problem was step two.

Open PageSpeed Insights, paste in your homepage, and look at the mobile results first. Then test your main service page and contact page too, especially if you run ads.

Field data comes from real Chrome users, if you have enough traffic. Lab data is a simulated test. Google explains the difference in its PageSpeed Insights documentation. Smaller sites often won't have field data yet, which is normal. Use the lab results as your guide and don't panic over a single score.

The three numbers that matter most are the Core Web Vitals. Here's what Google's web.dev Web Vitals guide treats as good, and what each one means in plain English.

INP replaced the older First Input Delay metric in 2024, so if you read an older guide that talks about FID, it's out of date. Screenshot your scores so you have a before picture.

If your website was built by uploading photos straight off a phone or a professional photographer's drive, there's a good chance some of them are several megabytes each. That's like posting a customer a brochure printed on concrete.

That hero image point trips up a lot of people. Lazy loading is great for photos at the bottom of the page, but applying it to everything delays the one image that decides your LCP. Google's guide to optimising Largest Contentful Paint goes into the technical detail if you or your developer want it.

Also ask whether the image needs to be there at all. A full-screen background video looks lovely in a design review. On a mid-range phone over mobile data, it often looks like a grey rectangle with ambitions.

Every website collects digital clutter over time. A chat widget from a free trial in 2022. Two analytics tools tracking the same thing. A Facebook pixel, a heatmap tool, a review widget, a cookie banner, and an Instagram feed that shows six photos nobody clicks.

Each one loads its own code, often from other companies' servers. Together they can make your page sluggish when someone taps a button, which is exactly what INP measures.

Run a simple audit. List every third-party tool on your site and ask one question about each: has this helped us win a customer in the last six months? If nobody can answer yes, remove it.

• Keep: analytics you actually check, a working enquiry form, and tracking for ads you're currently running.
• Question: chat widgets nobody monitors, social feeds, and pop-ups that appear before visitors have read a sentence.
• Remove: duplicate tracking, expired trials, and plugins that are installed but inactive.

On WordPress, plugin bloat is especially common, often with several plugins doing overlapping jobs. Fewer plugins also means fewer updates and fewer things to break.

You can optimise images all day, but if your server takes ages to respond, the page can't start loading quickly. In PageSpeed Insights, look for a warning about server response time. That's often a sign of cheap shared hosting, an overloaded server, or no caching.

Wix, Squarespace, Shopify, and Webflow manage servers for you. On self-hosted WordPress, this one is on you.

Moving from bargain hosting to a decent managed host is often one of the best value speed upgrades for a WordPress site.

Fonts. Many sites load four or five font weights from an external service when they only use two. Stick to one or two families and only the weights you use.

Sliders and carousels. Homepage sliders load several large images, plus the code to rotate them, so visitors can ignore four messages instead of one. Replace them with a single strong hero that says what you do and who it's for. It's faster, and it's usually clearer too.

Layout shift. You know the moment. You go to tap "Call now", an ad or banner loads above it, and you end up tapping something else entirely. That's layout shift, and CLS measures it. Common causes are images without set dimensions, late-loading banners, and embedded widgets that push content down.

A fast website that jumps around while loading still feels broken. Speed is about how quickly a visitor can read, trust, and act, not only how quickly the bytes arrive.

Our guide on what makes a website convert covers why one clear message usually beats a rotating slideshow of five.

Your office has fast fibre. Many of your customers have a three-year-old Android and patchy reception.

Once you've made changes, do a real-world test. Turn off Wi-Fi, use mobile data, and load your site on an average phone, not the newest one in the building. Then try the things a customer would actually do.

• How long until you can read the headline and see what the business does?
• Can you tap the phone number or main button without the page jumping?
• Does the contact form load, submit, and confirm without lag?

Customers won't describe a sticky page as "poor INP". They'll call it annoying. For more on what properly mobile-first builds look like, read our breakdown of responsive web design services.

Sometimes the platform is the problem. More often, it's how the site was built on it.

Switching platforms purely for speed is rarely the first move. If you are weighing options, our comparison of Wix vs Squarespace vs WordPress looks at speed alongside cost and control.

Speed work attracts quick-fix promises. Some are useful. Some are expensive ways to feel productive.

• Installing a speed plugin on top of a bloated site. Optimisation plugins can help, but they can't fully compensate for oversized images and 25 unnecessary scripts. You end up adding software to fix too much software.
• Chasing a perfect 100 score. The overall score swings between tests. Green Core Web Vitals on mobile is the sensible target.
• Only testing the homepage. If your Google Ads land on a service page, that's the page that needs to be fast. Ads send paying traffic, and a slow landing page wastes it.
• Ignoring the copy. A fast page that doesn't explain what you do just loses visitors faster.
• Making changes without a backup. Back up first, change one thing at a time, and test the enquiry form after every change.

A site that loads in 1.8 seconds with a broken contact form is fast in the least useful way possible.

If you've sped things up and Google still isn't showing your pages, speed probably isn't the issue. Our Google not indexing fix checklist walks through what to check next.

Not every slow site needs a new website. Plenty just need a weekend of image compression and a ruthless plugin cull. Others are so shaky that patching them is like repainting a car with no engine. Here's what each option typically costs in New Zealand.

These are typical ranges, not quotes.

If two or three of those sound familiar, paying to speed up a site that also doesn't convert can be throwing good money after bad. Our guide to website redesign cost and timeline covers what a rebuild involves and how to protect your rankings during the switch.

We'd rather build a site that's fast from day one than sell you a speed fix later. Performance is a design decision for us, made at the start. If a design choice slows the page down without helping the visitor trust you or enquire, it doesn't make the cut.

The honest note: we're a young studio, so we won't wave a stack of client speed results at you. What we can do is look at your current site and tell you plainly whether it needs a tune-up, a few hours of fixes, or a rebuild. Get a free website audit and we'll show you what's slowing it down and what to fix first. If you already know you want a fresh start, take a look at Onyxarro's packages and pricing.

Learning how to speed up a small business website mostly comes down to doing the boring things in the right order. Measure on mobile. Shrink your images. Remove what you don't use. Fix your hosting. Simplify the page. Then test it like a real customer would.

You don't need a perfect score. You need a site that shows your offer quickly, responds when someone taps, and stays still long enough for them to call you. Speed buys you a few extra seconds of attention. A clear website is what turns those seconds into an enquiry.

If you'd like a second pair of eyes before you start, request a free audit. You'll get a plain-English list of what's working, what's slow, and what's worth fixing first.