Google PageSpeed Insights gives your site a score between 0 and 100. It's a useful diagnostic tool, but the score itself is not what you're optimising for. The underlying metrics that produce the score are what matter for user experience and search performance. Here's how to read PageSpeed Insights correctly, understand what the score tells you and what it doesn't, and address the recommendations that produce the most improvement for WordPress sites.
The PageSpeed score is a weighted composite of several performance metrics measured in a simulated lab environment using a Lighthouse audit. The weighting changes periodically as Google updates Lighthouse. It's not a direct ranking signal, Google uses field data from the Chrome UX Report (real user measurements) for rankings, not the lab simulation score.
A site with a lab score of 68 but consistently good Core Web Vitals field data (real users loading it quickly) can outperform a site with a lab score of 95 in search rankings. The lab score and field data don't always align because the lab simulates a controlled environment (specific device, network conditions, empty cache) that may not reflect your actual user base.
That said, improving the lab score almost always improves the real-user field data too, because the root causes are the same: slow server response, heavy images, render-blocking scripts. Use the score as a diagnostic proxy, not as a target in itself. Focus on the specific recommendations and the individual metric values rather than the composite number.
When you submit a URL to PageSpeed Insights (pagespeed.web.dev), it does two things: it fetches field data from the Chrome UX Report for that URL (if sufficient traffic data exists) and it runs a Lighthouse lab audit from a Google server. The field data section shows real Core Web Vitals (LCP, INP, CLS) for the past 28 days. The lab section shows the simulated results with the score and individual metric breakdowns.
The "Diagnostics" and "Opportunities" sections below the score are the most actionable parts of the report. Each recommendation includes an estimated time savings in seconds. Prioritise the ones with the highest estimated savings, these are the issues most responsible for your current score and real-world load time.
Reduce initial server response time (TTFB). This appears whenever your server takes more than 600ms to respond. It's the most impactful single item in the report because TTFB delays everything else. Fix: enable server-side page caching. LiteSpeed Cache on a LiteSpeed host reduces cached page TTFB to under 50ms. This one change often accounts for 1–2 seconds of improvement in LCP and substantially moves the score.
Eliminate render-blocking resources. CSS and JavaScript loaded synchronously in <head> that prevents the page from rendering. Fix: add defer to non-critical scripts. Inline critical CSS. Load full stylesheets asynchronously: <link rel="preload" as="style" onload="this.onload=null;this.rel='stylesheet'">.
Properly size images. Images rendered at smaller dimensions than their actual file dimensions. The report will list each oversized image with the bytes wasted. Fix: upload images at the size they'll be displayed. Use srcset for responsive images so mobile visitors download smaller variants.
Serve images in next-gen formats. JPEG and PNG where WebP would be 25–35% smaller. Fix: convert images to WebP, or configure ShortPixel, Imagify, or Smush to serve WebP automatically from your media library.
Remove unused JavaScript / CSS. Plugins that load assets on every page regardless of whether those assets do anything on that page. Fix: audit plugins and conditionally dequeue scripts on pages where they're not needed.
PageSpeed Insights tests both mobile and desktop. The mobile test simulates a mid-range Android device (Moto G4 class hardware) on a 4G connection with 150ms round-trip latency. Most WordPress sites score significantly lower on mobile than desktop. A site scoring 90 on desktop might score 45 on mobile.
Focus on mobile first. It's tested with more restrictive conditions, the score improvement is typically more substantial, and mobile traffic now accounts for the majority of web traffic for most sites. If your mobile score is 40, identifying and fixing the top 2–3 recommendations (usually TTFB, image sizing, and render-blocking scripts) will typically get you to 70+ before you make any other changes.
The desktop score is useful for identifying server-side issues in isolation, because the faster network and faster simulated device mean front-end bottlenecks (heavy JavaScript, large images) have less impact on desktop scores. If your desktop score is good but mobile is poor, the bottleneck is client-side performance under constrained conditions, heavy JavaScript and large images. If both scores are poor, the server-side issues (TTFB, caching) are likely the primary cause.
The score is composed of these Lighthouse metrics with approximate weightings: First Contentful Paint (10%), Speed Index (10%), Largest Contentful Paint (25%), Total Blocking Time (30%), and Cumulative Layout Shift (25%). Note that Total Blocking Time (TBT) is the lab-equivalent of Interaction to Next Paint, it measures how much the main thread is blocked by long JavaScript tasks.
If your score is low primarily because of LCP, the fix is server-side (TTFB, image optimization). If it's low because of TBT, the fix is JavaScript (fewer plugins, defer, remove heavy libraries). If it's low because of CLS, the fix is image dimensions and font loading. The individual metric breakdowns tell you which category to focus on.
A well-configured WordPress site on LiteSpeed hosting with optimised images, minimal render-blocking resources, and a lean plugin list should achieve 85+ on mobile and 95+ on desktop. These targets are achievable without custom development or expensive optimisation services.
Scores below 50 on mobile almost always indicate a combination of high TTFB and heavy unoptimised images. Address the hosting layer first (enable page caching) and optimize images second. On most WordPress sites, these two changes alone move a 45 mobile score to 70–80 without touching JavaScript, CSS, or any other optimization.
PageSpeed Insights is the starting point, but other tools provide different diagnostic angles. GTmetrix gives a Lighthouse-based score alongside detailed waterfall charts showing asset load order and timing, useful for identifying which specific resources are blocking render. WebPageTest (webpagetest.org) offers the most detailed filmstrip view, connection views, and the ability to test from multiple geographic locations and real mobile devices. Google Search Console's Core Web Vitals report shows aggregate field data across all your pages grouped by URL pattern, the most useful view for identifying which page templates have systemic performance issues.
LiteSpeed with LSCache resolves the TTFB issue before you've optimised anything else. That alone moves the score more than any plugin combination.
View Hosting Plans