What Does a Website Speed Audit Actually Tell You?

You ran a speed check on your website. Maybe it was PageSpeed Insights. Maybe it was GTmetrix. Maybe someone sent you a screenshot of your Lighthouse score and it was sitting at 34, glowing red, and you did not know what to do about it.

Now you are looking at a wall of numbers, warnings, and terms like "Largest Contentful Paint" and "Cumulative Layout Shift" and wondering: what does any of this actually tell me? And more importantly, what am I supposed to do with it?

That is the right question. Because an audit is only useful if it tells you something you can act on. Scores without context are just noise. A number on its own does not fix your website. Knowing what that number means, what is causing it, and what to do first, that is what actually moves the needle.

Let me walk you through what a website speed audit measures, what a good one looks like, what a bad one looks like, and how to use the results to make real decisions about your site.

What a Website Speed Audit Actually Measures

When someone runs a speed audit on your website, they are checking a specific set of things. Here is what those things are, in plain English.

Lighthouse Scores

Google Lighthouse is the engine behind most speed audits. It loads your page in a simulated browser and scores you across four categories, each from 0 to 100:

• Performance: How fast your page loads and becomes usable
• Accessibility: Whether people with disabilities can use your site
• Best Practices: Whether your site follows basic web development standards
• SEO: Whether search engines can find, understand, and index your pages

Each category gets a colour. Red (0 to 49) means real problems. Orange (50 to 89) means it needs work. Green (90 to 100) means you are in decent shape. I wrote a full breakdown of what these scores mean and what to aim for in what is a good Lighthouse score.

The scores are a snapshot. They tell you where you stand right now. They do not tell you what to fix or in what order. That comes from the details underneath them.

Core Web Vitals

These are the three specific metrics Google uses to measure real-world page experience. They are the heart of any speed audit because they directly affect your search rankings and how visitors experience your site.

Largest Contentful Paint (LCP) measures how long it takes for the largest element on your page to appear. Usually that is a hero image, a headline, or a video. Google says LCP should be under 2.5 seconds. If yours is at 4 or 5 seconds, visitors are staring at a blank or partially loaded screen, and a lot of them are leaving. I have seen sites where fixing LCP alone increased time-on-page by 30% or more. If you want the full breakdown, read how to fix LCP.

Cumulative Layout Shift (CLS) measures how much your page layout jumps around while it loads. You know when you click a link and the page shifts at the last second and you end up clicking the wrong thing? That is layout shift. Google wants CLS under 0.1. High CLS is annoying for users and a known ranking factor. It is also one of the easiest things to fix once you know what is causing it. See how to fix CLS for the practical steps.

Interaction to Next Paint (INP) measures how fast your page responds when a user clicks or taps something. It replaced the old First Input Delay metric in 2024 and it is harder to fix because it measures your site's ongoing responsiveness, not just the initial load. If your INP is over 200 milliseconds, your site feels sluggish. Over 500 milliseconds, it feels broken. The fixes are usually about reducing JavaScript bloat and deferring non-essential scripts. Read how to fix INP for what that actually involves.

Real User Data (CrUX)

Lighthouse runs in a simulated environment. It uses a simulated mid-range phone on a throttled 4G connection. That is useful for diagnosing problems because the test is consistent and repeatable. But it is not the same as what your actual visitors experience.

That is where CrUX data comes in. CrUX (Chrome User Experience Report) is data Google collects from real Chrome users who visit your site. It shows the actual LCP, CLS, and INP that real people experienced over the previous 28 days. If your Lighthouse score is 90 but your CrUX data is failing, it means your site works fine on fast devices but struggles on slower ones. Both numbers matter. I explain the difference in detail in Core Web Vitals explained.

Accessibility, SEO, and Best Practices

A proper speed audit does not just look at performance. It also checks whether your site is usable by people with disabilities (alt text, colour contrast, keyboard navigation, proper headings), whether search engines can understand your pages (meta tags, crawlable links, mobile usability), and whether your site follows basic technical standards (HTTPS, no console errors, no deprecated code).

These areas overlap with speed. A site that loads 14 JavaScript files is slow AND has poor best practices scores. A site with no alt text on images has accessibility issues AND is harder for search engines to understand. A good audit looks at the whole picture, not just one number.

What a Good Audit Report Looks Like

A good audit report does three things: it identifies specific issues, it prioritises them by impact, and it tells you how to fix them. Here is what that looks like in practice.

Specific issues, identified by name. Not "your site is slow." But "your homepage hero image is 4.2MB and unoptimised, causing your LCP to sit at 3.8 seconds on mobile." That is a specific, verifiable finding. You can look at that image, check its file size, and confirm the problem exists.

Prioritised by impact. A good report does not list 47 issues in no particular order. It tells you which 3 to 5 issues are causing the most damage and which ones can wait. Usually the highest-impact issues are the ones affecting Core Web Vitals on mobile, because those directly influence both user experience and search rankings. I wrote about the most common ones in common website mistakes businesses make.

Clear fix recommendations. For each issue, the report should tell you what to do about it. Not in vague terms like "optimise your images." But in specific terms: "compress the hero image to under 200KB, convert it to WebP format, and add width and height attributes to prevent layout shift."

Here is an example of what a good audit finding looks like:

Issue: Render-blocking JavaScript in the document head (3 files, 340KB total) Impact: High. These files block the browser from rendering your page content until they finish loading. This is the primary cause of your slow LCP. Fix: Defer or async-load two of the three scripts (analytics and chat widget). Inline the critical CSS for the above-the-fold content. Expected improvement: LCP from 3.8s to under 2.5s.

That is something you can act on. You can hand it to a developer and they know exactly what to do. Or if you have the skills, you can do it yourself.

What a Bad Audit Looks Like

Bad audits are unfortunately common. Here are the warning signs.

Just scores, no explanation. A report that says "Performance: 45, Accessibility: 62, SEO: 78" and nothing else. That tells you nothing. You already knew your site was not great. You needed to know why and what to do about it.

Generic advice with no specifics. "Install a caching plugin" is not a fix recommendation. Which caching plugin? What settings? What is causing the slow load that caching would address? If the advice could apply to literally any WordPress site on the internet, it is not tailored to your site.

No prioritisation. A list of 40 issues presented as if they all matter equally. They do not. Fixing a 4MB hero image will transform your mobile load time. Fixing a missing favicon will not. If the report does not distinguish between the two, it is not helping you make decisions.

Sales pitch disguised as an audit. The report finds alarming problems, uses scary language, and then pivots to "but we can fix all of this for $2,000." If the audit does not show you specific, verifiable findings before asking for money, it is a sales funnel, not an audit. I wrote about how to spot this in how much does a website audit cost.

No mention of real user data. If the audit only shows lab data (Lighthouse) and ignores CrUX field data, it is missing half the picture. Your Lighthouse score might be fine while real users on slow phones are having a terrible experience. A good audit looks at both.

Free Tools vs a Professional Audit

You can run a free speed check right now. Google PageSpeed Insights takes 30 seconds and costs nothing. GTmetrix does the same thing with a slightly different interface. Both use Lighthouse under the hood.

These tools are genuinely useful. I am not going to pretend they are worthless. If you have never run one, go do it right now. The problem is not that free tools are bad. The problem is expecting them to do something they were not designed to do.

Here is the difference.

Free tools give you raw output. Lighthouse runs its tests and dumps every finding into a list. "Eliminate render-blocking resources." "Reduce JavaScript execution time." "Serve images in next-gen formats." "Minify CSS." These are all technically correct, but if you do not know what render-blocking resources are or which ones on your site are actually causing the problem, the advice is not actionable.

A professional audit gives you interpretation. It takes that raw Lighthouse output, adds manual review, cross-references it with real user data, and translates it into plain English. Instead of "eliminate render-blocking resources," you get "your theme is loading three JavaScript files in the head that block rendering. Here is what each one does, which ones you can defer, and what the expected impact is."

Free tools treat all issues equally. Lighthouse lists everything with the same weight. A critical performance problem and a minor best-practice nitpick sit side by side in the same list. A professional audit tells you which 3 things to fix first and which 20 things can wait.

Free tools do not look at your specific setup. They do not know what CMS you are on, what theme you are using, what plugins you have installed, or what your hosting environment looks like. A professional auditor does, and that context changes the recommendations. The fix for a slow Shopify store is different from the fix for a slow WordPress site, which is different again from a slow Wix site. I wrote about those differences in how to speed up WordPress and why is my Shopify store slow.

The short version: free tools tell you what is wrong. A professional audit tells you what to do about it and in what order.

Why Scores Alone Are Not Enough

This is the part that trips up most business owners. You run PageSpeed Insights. You get a Performance score of 42. You think "okay, I need to get that to 90." But the score is not the goal. The score is a symptom. The goal is a fast, usable website that does not lose customers.

Chasing a score without understanding what is driving it leads to wasted effort. I have seen business owners spend hours minifying CSS to gain 2 points on their Performance score while ignoring a 4MB hero image that was adding 3 seconds to their load time. The minification did almost nothing. The image fix would have transformed the site.

A score of 42 could mean your site has one massive problem (a huge unoptimised image) or twenty small ones (bloated theme, too many plugins, no caching, render-blocking scripts, uncompressed assets). The fix for each scenario is completely different. Without knowing which one applies to your site, you are guessing.

This is why the audit report matters more than the audit score. The score tells you there is a problem. The report tells you what the problem is, how bad it is, and what to do about it.

How to Use Audit Results to Decide What to Fix First

Once you have a proper audit in front of you, here is how to use it.

1. Look at mobile, not desktop. Over 60% of web traffic in 2026 comes from mobile devices. Google uses the mobile version of your site for indexing and ranking. If your mobile performance is poor, that is where to focus. Your desktop score being 90 is irrelevant if your mobile score is 35.

2. Start with Core Web Vitals. If your LCP, CLS, or INP are failing, fix those first. They have the biggest impact on both user experience and search rankings. LCP is usually the highest priority because it directly affects how fast your main content appears.

3. Fix the highest-impact issues, not the easiest ones. It is tempting to start with quick wins like minifying CSS because they feel productive. But if your hero image is 4MB, compressing it will do more for your speed score than every other fix combined. Impact first, effort second.

4. Work in batches and re-test. Fix the top 2 or 3 issues, then run the audit again. See what improved and what did not. This tells you which fixes actually mattered and which ones were noise. It also stops you from making changes blindly.

5. Get help for the stuff you cannot do. Some fixes are genuinely simple (image compression, alt text, meta descriptions). Others need a developer (render-blocking scripts, theme code, server configuration, INP optimisation). Be honest about which is which. I wrote a full framework for this in should I hire someone to fix my website or do it myself.

What WL Tech Can Do With Your Audit

At WL Tech, the process is designed to give you exactly what you need at each stage, without pressure.

Free audit. You start with a free audit. It runs real Lighthouse tests, checks Core Web Vitals, scans for accessibility and SEO issues, and gives you a scored report immediately. No sales call, no credit card. You see exactly where your site stands.

$150 Developer Report. If the free audit shows problems and you want the full picture, the Developer Report gives you a detailed, plain-English breakdown of every issue on your site with fix instructions. You can take it to any developer, including us, or use it to tackle fixes yourself.

$250 Quick Fix. If you want someone to handle the top three issues from your audit, the Quick Fix does exactly that. This is the sweet spot for most small businesses: measurable improvement without a big spend.

$500+ Full Fix. If you want everything fixed end to end, the Full Fix covers the lot. Core Web Vitals, theme code, server config, database, accessibility. Pricing depends on scope and we always tell you the exact cost upfront after the audit.

You can read more about how these compare in our guide to website fix costs.

The Bottom Line

A website speed audit tells you three things: where your site stands right now, what is causing the problems, and what to fix first. The scores are a starting point, not the finish line. What matters is what you do with the information.

A good audit gives you a clear, prioritised action plan in plain English. A bad audit gives you a number and leaves you to figure out the rest. Free tools are a solid starting point, but they do not replace interpretation, context, and prioritisation.

If you have never audited your site, start free. Run it through our free audit and see what comes back. If the results show problems you want to understand better, the Developer Report at $150 gives you the full breakdown. If you want fixes, the Quick Fix at $250 or Full Fix from $500 gets it done.

The audit is not the product. The audit is the map. The fixes are the journey. But you need the map before you can start walking.

Get your free audit →

---

FAQ

What does a website speed audit actually measure? A website speed audit measures Lighthouse scores for performance, accessibility, SEO, and best practices, along with Core Web Vitals metrics including Largest Contentful Paint (LCP), Cumulative Layout Shift (CLS), and Interaction to Next Paint (INP). It also checks real user data from Chrome users and flags specific technical issues affecting your site.

What is the difference between a free audit tool and a professional audit? Free tools like PageSpeed Insights and GTmetrix give you raw scores and a list of flagged issues with no prioritisation or context. A professional audit includes manual review, plain-English explanations of what each issue means, a prioritised action plan, and specific fix recommendations tailored to your site.

Are Lighthouse scores enough to know what to fix on my website? No. Lighthouse scores tell you how your site performed in a simulated test, but they do not tell you which issues matter most, what to fix first, or how to fix them. You need a prioritised report that explains each issue in context and gives you a clear order of operations.

What should a good website audit report include? A good audit report should include specific issues identified by name, each issue prioritised by its impact on real users and search performance, plain-English explanations of what is wrong, and clear fix recommendations you or a developer can act on. It should also cover Core Web Vitals, accessibility, SEO, and best practices, not just a single performance score.

Related reading

• What Is a Good Lighthouse Score?
• Core Web Vitals Explained: What Every Business Owner Needs to Know
• How Much Does a Website Audit Cost in 2026?
• Should I Hire Someone to Fix My Website or Do It Myself?
• How Slow Websites Lose Customers

---

About the author

Christopher Welsh is a systems engineer and founder of WL Tech. He specialises in website performance audits, technical SEO, and AI visibility optimisation for small businesses worldwide. No retainers, no jargon, just clear analysis and practical fixes.

Get a free website audit →