Hook
Your website might be losing customers right now, and you would never know it. A page that takes 4 seconds to load loses roughly 25% of visitors before they even see your offer. But when someone tells you to "check your Core Web Vitals" or "run a Lighthouse audit," it sounds like a foreign language.
I have been building websites for over 16 years, and the performance tooling world has become increasingly complex. The good news: you do not need to understand the technical internals to measure website performance effectively. You need to know which numbers matter, which free tools to use, and what the results mean for your business.
This guide covers exactly that. No code. No jargon without explanation. A clear process you can follow in under 30 minutes.
TL;DR Summary
- Website performance means how fast your pages load and how responsive they feel to visitors.
- Three metrics matter most: LCP (how fast your main content appears), INP (how quickly the page reacts to clicks), and CLS (whether content jumps around while loading).
- Free tools like Google PageSpeed Insights and GTmetrix give you these numbers in seconds.
- Test on mobile, not just desktop. Most of your traffic is probably on a phone.
- Slow pages directly cost you revenue: longer load times mean fewer conversions, lower search rankings, and higher bounce rates.
Need a hand with your website or web app?
Free 30-min strategy call. I'll review your situation and give you a clear next step.
Table of Contents
- Why measuring performance matters for your business
- The 5 metrics that actually matter (in plain English)
- Free tools to measure your website performance
- Step-by-step: run your first performance test
- How to read your results without a developer
- What "good" looks like (benchmarks by industry)
- When to call in a professional
- FAQ
Why measuring performance matters for your business
I have had dozens of conversations with business owners who redesigned their site, saw traffic drop, and blamed their SEO agency. In many cases, the real culprit was performance. The new design looked great but loaded 3 seconds slower.
Here is what the data says:
- A 1-second delay in page load time reduces conversions by approximately 7%, according to research by Akamai.
- Google uses page speed as a ranking factor. Slower sites rank lower in search results.
- 53% of mobile users abandon a site that takes more than 3 seconds to load, based on Google's own mobile speed study.
Performance is not a technical nice-to-have. It is a revenue lever. When your site loads slowly, you are paying for ads that send people to a page they leave before it finishes rendering.
The first step to fixing a performance problem is knowing you have one. That requires measurement, and measurement is something anyone can do.
The 5 metrics that actually matter (in plain English)
Performance tools spit out dozens of numbers. Most of them are noise. Here are the five you should care about, translated into business terms.
1. Largest Contentful Paint (LCP)
What it measures: How long it takes for the biggest visible element on your page to appear. Usually a hero image, headline, or large text block.
Why it matters: LCP is the moment your visitor feels the page has "loaded." Too long, and they leave.
Good target: Under 2.5 seconds. Between 2.5 and 4 seconds needs work. Over 4 seconds is a problem.
2. Interaction to Next Paint (INP)
What it measures: How quickly the page responds when someone clicks a button, taps a link, or types in a form. INP replaced the older First Input Delay (FID) metric in March 2024.
Why it matters: A page can look loaded but feel broken if buttons do not respond. INP captures that sluggishness.
Good target: Under 200 milliseconds. Between 200 and 500 milliseconds needs improvement. Over 500 milliseconds feels broken.
3. Cumulative Layout Shift (CLS)
What it measures: Whether elements jump around while loading. You have experienced this: you tap a button, an ad loads above it, the button shifts, and you tap the wrong thing.
Why it matters: Layout shifts frustrate users and hurt your Google rankings.
Good target: Below 0.1. Between 0.1 and 0.25 needs work. Above 0.25 is a poor experience.
4. Time to First Byte (TTFB)
What it measures: How long your server takes to start sending data after someone requests a page. Think of it as the response time of your hosting.
Why it matters: Slow TTFB delays everything else. It is often the first clue that your hosting or server configuration needs attention.
Good target: Under 800 milliseconds. Ideally under 200 milliseconds.
5. Total page weight
What it measures: The total size of everything the browser downloads to display your page: images, fonts, scripts, and stylesheets. Measured in megabytes (MB).
Why it matters: Larger pages take longer to load, especially on mobile networks. A 5 MB homepage loads slowly on a 3G connection regardless of your server quality.
Good target: Under 2 MB for most pages. Under 1 MB is ideal.
If you want a deeper understanding of the first three metrics, which Google collectively calls Core Web Vitals, I wrote a separate guide that goes into more detail.
Free tools to measure your website performance
You do not need to buy software or hire someone to get your performance numbers. These tools are free, require no account setup, and give you results in under a minute.
Google PageSpeed Insights (PSI)
URL: pagespeed.web.dev
Best for: Getting your Core Web Vitals scores with Google's own data.
PSI is the most important tool on this list. It shows "field data" from real Chrome users over the past 28 days, and "lab data" from a simulated test. Field data is what Google uses for rankings.
Type in your URL, hit "Analyze," and you get LCP, INP, CLS, and a 0-100 performance score. Green means good. Orange means needs improvement. Red means poor.
GTmetrix
URL: gtmetrix.com
Best for: Visual breakdowns and waterfall charts that show exactly what is loading and when.
GTmetrix runs a test from a real browser and gives you a timeline of every file your page loads. The waterfall chart is revealing even for non-developers. You can see which images are massive, which scripts take forever, and where the bottlenecks are.
The free tier tests from Vancouver, Canada. That is fine for a baseline.
WebPageTest
URL: webpagetest.org
Best for: Testing from multiple locations and seeing filmstrip views of how your page loads frame by frame.
WebPageTest is more advanced, but the filmstrip view is worth it. You see screenshots of what your page looks like at each second of loading. It makes performance tangible.
Google Search Console
URL: search.google.com/search-console
Best for: Seeing how Google evaluates your site's performance over time.
If you have Search Console set up (and you should), the "Core Web Vitals" report shows which pages pass or fail Google's thresholds. This is the closest thing to Google telling you directly which pages have a speed problem.
Chrome DevTools (built into your browser)
Best for: Quick spot checks on individual pages.
Right-click anywhere on your page, select "Inspect," then go to the "Lighthouse" tab. Click "Analyze page load." You get the same Lighthouse score as PageSpeed Insights, but tested from your own machine. Results will vary based on your computer speed and network.
Step-by-step: run your first performance test
This process takes about 20 minutes and gives you a clear picture of where your site stands.
Step 1: Pick your 5 most important pages
Do not test your entire site at once. Start with the pages that matter most to your business:
- Your homepage
- Your highest-traffic landing page (check Google Analytics to find this)
- A product or service page
- Your contact or booking page
- A blog post that gets decent traffic
Step 2: Test each page on PageSpeed Insights
Go to pagespeed.web.dev and enter each URL. For each page, note:
- The overall performance score (0-100)
- LCP, INP, and CLS values
- Whether results are field data or lab data only
Save these results. A screenshot or spreadsheet works fine.
Step 3: Check mobile results
PageSpeed Insights defaults to mobile testing, which is what you want. Mobile results are almost always worse than desktop, and mobile is how Google evaluates your site for rankings.
Step 4: Run GTmetrix for your worst page
Take the lowest-scoring page and run it through GTmetrix. In the waterfall chart, look for:
- Large images (anything over 200 KB is worth questioning)
- Long bars (files that take a long time to download)
- Red items (failed requests or missing resources)
Step 5: Document your baseline
Write down the scores. This is your baseline. You cannot improve what you do not measure, and you need a "before" snapshot so you can tell whether future changes actually help.
I keep a spreadsheet with columns for: Page URL, Test Date, Performance Score, LCP, INP, CLS, and Notes. The point is having a record.
How to read your results without a developer
You ran the tests. You have a bunch of numbers. Here is how to interpret them.
The traffic light system
Both PageSpeed Insights and GTmetrix use green, orange, and red indicators. Here is what they actually mean:
| Color | Meaning | What to do |
|---|---|---|
| Green | Passing Google's thresholds | No urgent action needed |
| Orange | Needs improvement | Plan to address within the next quarter |
| Red | Poor experience | Fix this soon, it is actively hurting your business |
Performance score ranges
The 0-100 score from Lighthouse and PageSpeed Insights breaks down like this:
| Score | Rating | Business impact |
|---|---|---|
| 90-100 | Good | Your site is fast. Protect it during future redesigns. |
| 50-89 | Needs improvement | There are meaningful gains available. Worth investigating. |
| 0-49 | Poor | Your site is likely losing visitors and rankings. Act on this. |
Common issues you will see in reports
"Reduce unused JavaScript" means your site loads code that never runs. Common with heavy page builders and too many plugins.
"Serve images in next-gen formats" means your images are in JPEG or PNG when they could be WebP or AVIF, which look the same but are much smaller.
"Eliminate render-blocking resources" means certain CSS and JavaScript files force the browser to wait before displaying anything.
"Largest Contentful Paint element" tells you which element takes the longest to appear. Usually a large hero image that has not been optimized.
You do not need to fix these yourself. But understanding what they mean lets you have an informed conversation with whoever does.
What "good" looks like (benchmarks by industry)
"Fast" and "slow" are relative. An e-commerce site processing payments has different performance expectations than a five-page brochure site. Here are realistic targets:
| Site type | LCP target | Performance score target |
|---|---|---|
| Brochure/portfolio site | Under 1.5s | 90+ |
| Blog or content site | Under 2.0s | 85+ |
| E-commerce store | Under 2.5s | 75+ |
| Web application (SaaS) | Under 2.5s | 70+ |
| Site with heavy media/video | Under 3.0s | 65+ |
These are realistic targets based on what I have seen across 250+ projects, not Google's official thresholds. If you meet them, you are ahead of most competitors.
One pattern I have seen: businesses that monitor performance regularly maintain it. The ones that check once let it degrade as new features and content get added. Set a reminder to re-test quarterly.
When to call in a professional
You can measure performance yourself. Fixing it is where things get technical. Here are the situations where bringing in a developer makes sense:
Your scores are in the red across multiple pages. A single slow page might be an oversized image. Consistent poor performance usually points to deeper issues with hosting, code, or third-party scripts.
You have tried the obvious fixes. Compressed images, removed plugins, switched hosts, and scores still have not improved? The remaining issues are likely in the code itself.
You are about to redesign or rebuild. Build performance in from the start rather than bolt it on later. I wrote about this in my guide on website speed optimization.
Performance is costing you real money. If you can tie slow pages to lost conversions or dropping rankings, the ROI on professional optimization usually pays for itself within months.
If you are in that situation, I work with businesses to audit and fix performance issues as part of my website development services. We start with measurement, identify root causes, and implement fixes with before-and-after benchmarks.
Want to talk through your situation? Get in touch and I will take a look at your numbers.
FAQ
How often should I test my website performance?
Test after every significant change to your site, including content updates, plugin installations, redesigns, and hosting migrations. At minimum, run a full test quarterly to catch gradual performance degradation before it impacts your search rankings or conversion rates.
Is a PageSpeed Insights score of 70 good enough?
A score of 70 is acceptable for complex sites like e-commerce stores or web applications, but most business websites should aim for 85 or higher. The score matters less than the individual Core Web Vitals metrics. A site scoring 70 with passing LCP, INP, and CLS is often fine.
Does website speed really affect my Google rankings?
Yes. Google has explicitly confirmed that page experience signals, including Core Web Vitals, are ranking factors. In competitive search results where content quality is similar between sites, faster pages tend to rank higher than slower ones.
Can I measure website performance on my phone?
Yes. PageSpeed Insights and GTmetrix both work in mobile browsers. You can also use Chrome on Android with the "Lighthouse" feature in DevTools. Testing from your phone shows you the real mobile experience, though the score may differ from desktop lab tests.
What is the difference between field data and lab data?
Field data comes from real users who visited your site using Chrome over the past 28 days. Lab data comes from a single simulated test run. Field data reflects actual visitor experience and is what Google uses for ranking decisions. Lab data is useful for debugging specific issues but does not represent real-world conditions.
What to do next
You now have the knowledge and tools to measure website performance without relying on anyone else. Start with your five most important pages, run them through PageSpeed Insights, and document your baseline scores.
If your numbers look good, set a quarterly reminder to re-test and protect those scores during future changes.
If your numbers are orange or red, you have two paths. Quick wins like image compression and plugin cleanup are manageable on your own. Deeper performance issues in code or architecture are where a developer earns their fee.
Either way, you are making decisions based on data instead of guesswork. That puts you ahead of most business owners I talk to.
