Checking speed and Core Web Vitals in SEO by Zaweba

SEO by Zaweba measures speed in two ways. Out of the box it records local LCP (Largest Contentful Paint), CLS (Cumulative Layout Shift) and page weight from its own renderer. If you switch on a free Google PageSpeed Insights key, it adds Google's lab and real-user Core Web Vitals as well, using one request per audit.

Speed sits in the UX/Speed category of the scoring, alongside the other eight categories. It's rarely the whole story of why a page ranks, but a slow, jumpy page is a poor experience, and that's worth fixing whatever it does for search.

What the local figures tell you

When Render pages with JavaScript is on, the app reads the page in Microsoft Edge WebView2 and takes measurements from that renderer. LCP is roughly how long it takes for the biggest thing on the screen to appear, usually a hero image or a heading. CLS measures how much the layout jumps about while the page loads, which is what makes you tap the wrong button on a phone. Page weight is how much the page downloads in total.

These figures are measured on your PC and connection, so treat them as a guide rather than the last word. They're most useful for comparison, because the same measurements are taken for the competitors in the audit.

Adding PageSpeed Insights

Google's PageSpeed Insights gives lab results and, where Google has enough visits to the page, real-user data. To use it in SEO by Zaweba you need a free key, which you add in the optional integrations. Keys are stored encrypted for your Windows account, and the integrations are off by default. With the key on, the audit makes one request per audit and adds the results to the speed checks.

Real-user data is the stronger evidence, because it reflects actual visitors. Not every page has it; quieter pages may have lab results only.

Using the numbers well

Compare rather than chase. The Comparison tab puts you, each competitor and the top-3 average side by side. If your LCP is much slower than the pages above you, that's worth attention. If you're already in line with them, speed is unlikely to be what's holding you back, and the action plan will usually point elsewhere.

Look at page weight alongside LCP. A heavy page is often slow because of large images, and the image checks report file size and natural versus displayed size, which shows where pictures are bigger than they need to be.

Common mistakes

Testing once and drawing conclusions is the main one. Local timings vary from run to run with your connection. Run the audit again on another day, and History will show whether a change is real.

Another is fixing speed at the cost of content. Stripping out useful images or text to shave weight can make a page worse for readers. Fix what's genuinely wasteful first: oversized images, things that shift the layout, and anything that loads but isn't used.

Where speed sits in the plan

Every check carries a severity, an effort and a plain reason, and priority is worked out from severity, weight, confidence and effort. A severe speed problem that's easy to fix will rise to the top of the next steps. A small one that would take days will sit lower, which is usually the right call.