Google PageSpeed Audits: 3-Minute Video Walkthrough

If your website feels a bit sluggish, you are not alone. We see it all the time with small business sites around Leighton Buzzard and Bedfordshire. A page that takes "just a few seconds" to load can quietly cost you enquiries, bookings and sales, because people simply do not wait. They hit back, they try a competitor, and Google notices.

A Google PageSpeed audit is one of the quickest ways to find out what is slowing your pages down and what to fix first. The trick is not just running the test. It is understanding what the results actually mean, and which recommendations will make a real difference for real customers on real phones, on real connections.

This walkthrough shows you how to run an audit, how to read the main scores and metrics, and how to turn the recommendations into practical next steps.

Google PageSpeed Insights is a free tool that analyses a web page and reports on performance, accessibility, best practices and search optimisation. The performance part is what most people care about. It is powered by Lighthouse, which is the same auditing engine built into the Chrome browser.

It is important to know what the tool is doing. It is a helpful diagnostic. It is not a perfect representation of every visitor's experience. Think of it like an MOT for a car. It can highlight issues and risks, but you still need to apply common sense and prioritise what matters.

Also, chasing a perfect 100 score is rarely the best use of time for a small business. The goal is a site that feels fast and smooth to your customers, especially on mobile, and a site that is easy for Google to crawl and understand.

For many local businesses, your website has one job. It needs to get a visitor to take action. That might be calling, filling in a form, booking an appointment, or getting directions.

When pages are slow, three things tend to happen. First, people leave. This is even more true for mobile users, which is most local traffic these days. Someone searching "electrician near me" while standing in their kitchen is not going to wait for a heavy homepage slideshow to load.

Second, fewer people convert. Even if they stay, slow pages create friction. Buttons feel unresponsive. Layouts jump around. Forms take longer to submit. All of that reduces trust.

Third, you lose some advantage in Google. Page experience is not the only ranking factor, but performance and usability signals do matter, especially when Google is choosing between similar local options.

You only need three things to get useful results.

You need the exact web page address you want to test, not just the domain. Your homepage may perform very differently to a service page or a booking page.

You need to decide whether you care most about mobile or desktop. In most cases, start with mobile.

You need a way to take notes. The most useful audits are the ones you repeat after changes, so you can prove what improved.

Method one, PageSpeed Insights website: Open your browser and go to pagespeed.web.dev. Copy and paste the full address of the page you want to test, then run the analysis. When the report loads, you will see two tabs near the top for mobile and desktop. Start with mobile. Many sites look fine on desktop and struggle on mobile, so the mobile report is usually more revealing. This method is great because it can show you two types of data, which we will cover next.

Method two, Lighthouse in Chrome: Open your page in Google Chrome. Right click anywhere on the page and choose Inspect to open Developer Tools. Then find the Lighthouse tab. If you do not see it, click the double arrow menu to reveal more tabs. Choose the categories you want to test. For speed, choose Performance. Then run the audit. This method is handy because it is quick, and you can test changes on a staging site or a page that is not publicly accessible. It is also helpful if you want to test while blocking scripts, testing different device settings, or running repeat audits.

Near the top of PageSpeed Insights, you may see a section called "Discover what your real users are experiencing". This is field data, collected from real Chrome users who have visited your site. It is also called CrUX data.

Below that, you will see "Diagnose performance issues". This is lab data. It is a controlled test run at the moment you run the report.

Both are useful. Field data tells you what is happening in the real world, across different phones, networks and times of day. It is the best indicator of genuine user experience. Lab data helps you troubleshoot. It gives specific diagnostics and a repeatable way to measure improvements as you make changes.

If you have a smaller site with low traffic, you may not have field data available yet. That is normal. In that case, rely on lab data and your own analytics.

The performance score is a number out of 100. It is calculated from several metrics, weighted by how strongly they relate to perceived speed and usability.

Treat it as a guide, not a judgement. A page can score 85 and still feel slow if it is jumpy or if the main content loads late. A page can score 65 and still be perfectly acceptable for a simple contact page, especially if the key actions work smoothly. The more important point is whether you are improving over time, and whether your key pages meet Google's Core Web Vitals.

Largest Contentful Paint, often shortened to LCP, measures how quickly the main content becomes visible. On many small business sites, this is the large hero image, headline, or banner at the top. If LCP is poor, people feel like the site is slow, even if some parts load in the background. Common causes include large images, slow servers, heavy themes, and too much code loading before the main content.

Interaction to Next Paint, shortened to INP, measures responsiveness. It looks at how quickly the page reacts when someone taps a button, opens a menu, or types into a form. If INP is poor, users feel like the site is laggy or broken, which is especially damaging for enquiry forms and booking steps. Common causes include too much JavaScript, too many plugins, and third party scripts fighting for attention.

Cumulative Layout Shift, or CLS, measures visual stability. If elements move around while the page loads, that is a layout shift. You have seen it when you try to click something and the page shifts, and you end up clicking the wrong link. It is frustrating, and it reduces trust. Common causes include images with no defined size, cookie banners appearing and pushing content, and fonts that load late and change the spacing.

First Contentful Paint is when the first bit of content appears. Speed Index is an estimate of how quickly the page looks complete. These are helpful, but for most small business owners, LCP, INP and CLS are the ones to focus on first.

Once you scroll down, you will see diagnostics and opportunities. Some will feel technical, but most map to a few common issues.

For local business sites, images are often the main culprit. Large photos from a phone, uncompressed banners, and sliders with multiple images can quickly add megabytes. If PageSpeed suggests serving images in next gen formats, it is talking about formats like WebP or AVIF. If it suggests properly sizing images, it means you might be uploading a huge 3000 pixel wide photo and displaying it at 600 pixels. If it suggests deferring offscreen images, it is pointing to lazy loading. Action you can take today: check your largest images, especially on the homepage and top service pages, and re export them at the size you actually display. If your site runs on WordPress, ensure image compression is enabled and that modern formats are being generated.

Reduce unused JavaScript and CSS: your site might be loading code for features you are not using. When PageSpeed points to unused JavaScript or render blocking resources, it is telling you that code is being loaded too early, too often, or unnecessarily. Action: audit your plugins and remove what you do not truly need. Every plugin can add scripts, styles and database load.

Improve server response time: if PageSpeed flags initial server response time or time to first byte, the issue may be your server. Cheap shared hosting can struggle at busy times. Action: check your hosting plan, your server location, and whether you have proper caching in place. UK based hosting, a modern server setup, and good caching can make a noticeable difference.

Enable text compression: this means sending files like HTML, CSS and JavaScript in a compressed format so they download faster. If PageSpeed flags this, it can often be fixed at the server level. Many modern hosts already enable it, but some do not. This is usually a job for whoever manages your hosting.

Reduce the impact of third party scripts: chat widgets, tracking pixels, cookie consent tools, social media embeds, booking systems, and some map embeds can be heavy. Action: be ruthless. Do you need that chat widget on every page, or only on the contact page. Do you need three analytics tools, or can you simplify. Also, be mindful of cookie banners. Some are well built and lightweight. Some are not.

Start with one page that matters, not the whole site. For many local businesses, that is either the homepage or the highest traffic service page.

Run a mobile audit and write down the LCP, INP and CLS values, plus the performance score. Then make one change at a time, and re test. If you change ten things at once, you will not know what worked.

Most sites get the biggest improvements from these three areas. First, fix the biggest image on the page, usually the banner image at the top. Second, remove or delay anything that is not essential for the first view, such as sliders, heavy animations, and non essential scripts. Third, make sure caching and compression are properly set up.

If you do that, you often see a noticeable improvement without rebuilding the whole site.

One is that "mobile score is low so my site is broken". Mobile tests are strict, and they simulate slower devices. A lower mobile score is common. The aim is steady improvement and meeting Core Web Vitals on key pages.

Another is that "my competitor scores higher so I need the same". Not necessarily. Some businesses have simpler sites with fewer features. If your site needs an online booking system, you may accept a small performance cost, but you should still optimise around it.

A third is that "a 100 score means more sales". Speed helps, but clarity, messaging, trust and a good call to action matter just as much. A fast site with confusing copy will not outperform a slightly slower site that explains your services clearly and makes it easy to enquire.

If PageSpeed repeatedly flags complex items like reducing JavaScript execution time, eliminating render blocking resources, or issues tied to your theme and build setup, it may be time for professional support.

For many small businesses, the best value is not hours spent trying to squeeze out five extra points. It is investing in a cleaner setup, lighter templates, better image handling, and a sensible plugin stack.

As a web design company in Leighton Buzzard, we often find that a few targeted improvements can make a site feel dramatically faster, without changing the look and feel your customers already recognise.

After you make improvements, re run the same test on the same page, on mobile, ideally at a similar time of day. Performance can vary, so run it a couple of times and look for consistent improvement.

Also check your real world results. Are more people reaching the contact page. Are calls increasing. Are fewer people bouncing straight away. Google PageSpeed is a tool, but your business results are the real scorecard.

If you want, share the link to one page you would like to improve and tell us what your site is built with. WordPress, Wix, Squarespace and Shopify all have different common causes. We can point out which PageSpeed recommendations are likely to be worth your time first.