The MAZ SolutionsThe MAZ

How to Read a PageSpeed Insights Report (and What to Fix First)

How to Read a PageSpeed Insights Report (and What to Fix First)
pagespeed insightshow to read pagespeed insightspagespeed insights report explainedcore web vitalsimprove pagespeed score
Back to blog

You ran your website through Google PageSpeed Insights, got a number in red or orange, and a long list of "opportunities" written in developer language. Now what?


This guide walks through the report from top to bottom, explains which parts actually matter for a small business site, and gives you a sensible order for fixing things. No jargon you don't need.


First, the two halves of the report


A PageSpeed Insights report has two very different sections, and mixing them up is the most common source of confusion.


"Discover what your real users are experiencing" is *field data*. It comes from the Chrome User Experience Report: real Chrome visitors to your site over the last 28 days. If your site doesn't get enough traffic, this section says there isn't enough data, which is normal for many small sites.


"Diagnose performance issues" is *lab data*. This is a single Lighthouse test Google runs right now, on a simulated mid-range phone over a slowed-down mobile connection. The big score from 0 to 100 lives here.


Why it matters: the 0–100 score is a lab test. It moves a few points between runs, even when nothing on your site changed. The field data is what real visitors experienced, and it's what Google's Core Web Vitals assessment is based on.


Also check the tab at the top. PageSpeed Insights shows Mobile by default. Mobile scores are almost always lower than desktop because the test simulates a slower phone and network. Most local business traffic is on phones, so mobile is the one to work on.


The Core Web Vitals, in plain English


Google's Core Web Vitals are three measurements. Each one is judged at the 75th percentile of visits, so most visitors have to get a good experience, not just the lucky ones.


  • LCP (Largest Contentful Paint): how long until the biggest thing on screen, usually the hero image or main heading, appears. Good is 2.5 seconds or less.
  • INP (Interaction to Next Paint): how quickly the page responds when someone taps a button or opens a menu. Good is 200 milliseconds or less.
  • CLS (Cumulative Layout Shift): how much the page jumps around while loading, like a button moving just as you tap it. Good is 0.1 or less.

If all three are green in the field data, your site passes the Core Web Vitals assessment, whatever the lab score says.

What the lab metrics mean


The lab section shows five metrics that make up the performance score:


  • First Contentful Paint (FCP): when anything first appears.
  • Largest Contentful Paint (LCP): same idea as above, measured in the lab.
  • Total Blocking Time (TBT): how long the page is too busy running scripts to respond. INP can't be measured in a lab test without a real person tapping, so TBT is the lab stand-in.
  • Cumulative Layout Shift (CLS): layout jumping, as above.
  • Speed Index: how quickly the visible page fills in overall.

In current Lighthouse versions, TBT, LCP and CLS carry most of the weight in the score. So a slow hero image or heavy scripts will drag the number down far more than a slightly slow first paint.


The colors: 90–100 is green, 50–89 is orange, 0–49 is red.


Reading the "opportunities" and "diagnostics"


Below the metrics, Lighthouse lists suggestions. They're useful, but they are not a to-do list you must finish. A few tips:


  • Use the "Show audits relevant to" filter. Click LCP, TBT or CLS to see only the suggestions that affect that metric. Start with whichever metric is worst.
  • Expand each item. The detail usually names the exact file, image or script responsible. "Reduce unused JavaScript" means little until you see that most of it is a chat widget or a slider plugin.

What to fix first on a typical small business site


Every site is different, but on the WordPress, Shopify and Webflow sites we look at, the problems tend to show up in this order:


1. The hero image. A huge photo at the top of the homepage is the most common LCP problem. Resize it to the size it's actually shown, serve a modern format like WebP or AVIF, and make sure it is *not* lazy-loaded (the first image should load right away; lazy loading is for images further down).

2. Third-party scripts. Chat widgets, review carousels, booking embeds, heat-map tools and multiple tracking tags all compete for the phone's processor. That shows up as high TBT. Remove what you don't use, and delay the rest until after the page loads or until someone interacts.

3. Page builders and plugin bloat. Heavy themes and builders load a lot of CSS and JavaScript on every page, including features the page doesn't use. Disabling unused modules or loading plugin assets only where needed often helps more than any caching plugin.

4. Fonts. Several font families and weights from an outside service add requests before text can appear.

5. Layout shift. Give images and embeds a width and height (or a fixed aspect ratio), and reserve space for banners and cookie notices so content doesn't get pushed down.

6. Hosting and caching. If "Reduce initial server response time" shows up, the server itself is slow. Page caching and a better host fix this; front-end tweaks won't.


Re-test after each change. Run the test two or three times and look at the typical result, not the best or worst one.


Mistakes to avoid


  • Chasing 100. A score of 100 is nice for a screenshot, but customers don't see the score. Passing Core Web Vitals and a fast-feeling mobile page matter more than the last few points.
  • Testing only the homepage. Your service pages and landing pages for ads are often slower, and they're where people decide to contact you.
  • Expecting rankings to jump. Speed is part of Google's page experience signals, and a faster site helps visitors. It is not a guarantee of higher rankings.

A quick way to get a second opinion


If the report still reads like another language, you can run our free speed check. It uses the same Google test and translates the result into your mobile score, LCP, CLS and TBT, plus the top three fixes in plain English. No signup.


And if you'd rather have someone do the fixing, our Speed Fix is a fixed-price job with a before-and-after report, so you know what you're paying before work starts.


FAQ


Why does my PageSpeed score change every time I test?


The lab test runs on shared Google servers and simulates a slow phone and network, so small differences in timing change the result. A few points of movement between runs is normal. Compare typical results, not single runs.


Is a low mobile score bad if my site passes Core Web Vitals?


Not necessarily. Passing Core Web Vitals in the field data means real visitors are getting a good experience. The lab score is still useful for spotting problems, but it's a diagnostic, not a verdict.




Got a report you're not sure how to read? Send me your URL and I'll tell you honestly which one or two fixes would make the biggest difference.