Google PageSpeed: What the Numbers Actually Tell You
The Score Is a Signal, Not a Verdict
Google PageSpeed Insights gives you a number between 0 and 100. It feels definitive. It is not. The score is a weighted summary of several underlying metrics, and those metrics carry different weights depending on the page type, the device tested, and the data Google has collected from real users visiting your site.
A score of 65 on one page can mean something very different from a 65 on another. One might have a slow server response time dragging the number down. Another might have an unoptimised third-party script that fires late but never actually blocks the user from doing anything useful.
Lab Data vs Field Data: Two Different Stories
PageSpeed Insights pulls from two sources. Lab data comes from a simulated test run on a throttled connection with a mid-range device. Field data comes from the Chrome User Experience Report, which aggregates real visits from real people on real connections.
These two sources regularly disagree. A page can score 45 in the lab and still pass all three Core Web Vitals thresholds in the field, because real users on good connections do not experience the same bottlenecks a throttled simulation creates. When they conflict, understanding which score actually matters saves you from chasing problems that do not exist.
Field data is harder to game. It also takes time to update, usually 28 days of rolling data. So if you made a change last week, the field data will not yet reflect it.
The Three Metrics Worth Watching Closely
Core Web Vitals are the metrics Google uses to assess real-world experience. They are also the ones tied to search ranking. Focus here first.
- Largest Contentful Paint (LCP) measures how long the biggest visible element takes to load. Under 2.5 seconds is good. Over 4 seconds is a problem.
- Interaction to Next Paint (INP) measures responsiveness when a user clicks, taps, or types. Under 200ms is good.
- Cumulative Layout Shift (CLS) measures how much the page jumps around while loading. Under 0.1 is good. A score above 0.25 means buttons and text are visibly shifting as the page settles.
These three metrics carry the most weight in the scoring formula and the most consequence for search performance. Fix these before anything else on the diagnostics list.
What the Diagnostics Tab Is Actually For
Below the score, PageSpeed lists diagnostics and opportunities. Many site owners treat every item as urgent. Most are not. Some items, like eliminating render-blocking resources or deferring offscreen images, can meaningfully improve LCP. Others, like removing unused CSS on a WordPress site running a well-built theme, may save a few kilobytes but move the score by two points at most.
Read each diagnostic alongside its estimated saving. If an item shows a saving of under 0.1 seconds, it is not the root cause. Work down from the largest savings first, and cross-reference against which Core Web Vital it affects.
Why WordPress Sites Often Score Lower Than They Should
WordPress loads a lot by default. Plugins add scripts. Themes add stylesheets. Builders add both. The lab test captures all of it, including scripts that may only fire after a user interaction in the real world.
The practical fix is not to strip WordPress back to nothing. It is to audit what is loading on each page, defer what is not needed above the fold, and serve images in modern formats. A well-configured caching layer and a fast host solve a large portion of the remaining gap. The architecture matters more than the score target.
A Score to Aim For, Not Chase
Targeting a score of 100 is a distraction. Google does not expect it and does not reward it over a score of, say, 78 if your Core Web Vitals are passing. The goal is a site that loads fast for real users on real devices, passes the three Core Web Vitals thresholds, and does not create friction between a visitor and whatever they came to do.
Measure that first. Then let the diagnostics guide where to spend the technical effort. If you want a second opinion on where your site actually stands, get in touch and book a free audit.