Skip to content

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

A PageSpeed Insights report can look like a wall of red and orange. Here is what each part means, which findings deserve your time, and what to fix first on a typical local business site.

Our Abbotsford marketing team6 min read
KEY TAKEAWAYS
  1. The big coloured score matters less than the three Core Web Vitals and whether real visitors passed them.
  2. Test the mobile tab first, because many of your customers arrive on a phone and its results are the stricter ones.
  3. Fix the heaviest image and the scripts nobody needs before chasing the small savings further down the list.
  4. Run the test on several pages, and again after each change, since one reading can swing from run to run.
01

What the report is actually telling you

PageSpeed Insights is a free Google tool. Paste in a web address and it studies the page in two ways: a lab test on a simulated mid-range phone with a slow connection, and, if enough people have visited, measurements gathered from real Chrome users. These are two different kinds of evidence, and the report keeps them apart.

The top of the page shows the real-visitor result, marked as passed or not passed, followed by the lab score out of one hundred. Green starts at ninety, orange covers the middle and red is anything below fifty. A small local site often has no real-visitor data at all, so the lab score is all you get. Treat it as a guide, not a grade.

02

The three numbers behind the score

Google measures page experience through three signals called Core Web Vitals. You do not need the jargon to use them, because each one answers a question a visitor would ask and has a clear line between good and poor. Read these before anything else, since they describe how a visit feels better than the single score does.

The lab section adds more figures, including First Contentful Paint, Total Blocking Time and Speed Index. Treat them as clues rather than targets. If the main picture is slow, ask which file it is and why it is heavy. If things jump around, look for images with no set size, banners that appear late and fonts that swap after the page has loaded.

  • Largest Contentful Paint: how long until the main picture or headline appears, where 2.5 seconds or less is good
  • Interaction to Next Paint: how quickly the page reacts to a tap or click, where 200 milliseconds or less is good
  • Cumulative Layout Shift: whether content jumps while loading, where 0.1 or lower is good
03

Reading the opportunities without drowning in them

Below the numbers sits a long list of findings with estimated time savings. Many of them save a few hundredths of a second and are not worth a developer’s afternoon. Look for the one or two findings with the biggest savings, and for any that name a specific file, because a named file is something you can shrink, move or remove.

Ignore advice that does not fit your setup. A tip about newer image formats helps only if your site can serve them. A warning about third-party code means a script from another company, such as a chat bubble, a review feed or a tracking tag. Ask of each one whether it earns its place, since you can often drop one without anyone noticing.

04

What to fix first on a typical local business site

On small business sites the same few problems appear again and again. Photos taken on a modern phone are several megabytes each and get uploaded without resizing. A homepage slider loads four pictures to show one. Plugins and widgets pile up over the years. And budget hosting can answer slowly before any page work begins.

Start with the biggest win, then retest. If the top image is your slowest element, fixing it can move the result more than everything else combined. Change one thing at a time where you can, so you know what helped and what did not, and keep a note of each change with the date.

  • Resize and compress photos before upload, starting with the top image on each page
  • Replace sliders and autoplay video with a single still image
  • Remove plugins, widgets and tags that nobody uses
  • Set a width and height on images so the page stops jumping
  • Ask your host about caching, and whether a faster plan is overdue
  • Cut the fonts down to the ones you really use
05

How often to test, and when to stop

Scores shift from run to run, so test each key page three times and note the middle result. Check your homepage, your busiest service page and your contact page, the places where people decide. Compare like with like, mobile with mobile, and write down the date and what you changed.

There comes a point where more speed work stops paying. Once your main pages pass the Core Web Vitals and open in a couple of seconds on a phone, the next gain is more likely to come from clearer copy or an easier enquiry form. Speed helps people stay, but relevance and useful content still matter more, so do not strip a good page to chase a perfect score.

WHAT'S NEXT · EMERGING TECH

Make speed a habit, not a project

A site that passes today can slip next month after a plugin update, a new banner or a batch of heavier photos. Set a reminder to run the report on your three key pages every quarter, and again after any big change. If the same findings keep coming back, the cause is probably a habit, such as uploading photos at full size, rather than the site itself.

  1. 01Test three key pages every quarter
  2. 02Resize photos before they are uploaded
  3. 03Review plugins and scripts twice a year
Q&A

Questions & answers

NEXT ARTICLEWebsite Not Displaying Properly on Mobile? How to Find the Cause and Fix It →

Want to put this
into practice?

+1 (604) 800-3883(604) 800-3883 Start a project