4 minutes readKey takeaways Most WordPress slowness comes from plugins and unoptimised images, not the core. Caching plus image optimisation covers the majority of the available gain. Audit plugins periodically. Each one adds queries and assets to every page. We managed to get a 99/100 Performance Score with Google Lighthouse, check the details below how we did… Read more »
4 minutes readKey takeaways CLS measures unexpected movement, which is why a page can feel broken while loading fast. The usual causes are images without dimensions, injected banners and late-loading fonts. Reserve space for anything that arrives late and most of the score fixes itself. Cumulative Layout Shift or CLS, Web Vitals and Core Web Vitals, these… Read more »
3 minutes readKey takeaways Cypress runs in the browser alongside your app, which makes debugging far more direct. It suits front-end integration tests better than cross-browser coverage. Its retry behaviour hides flakiness. Investigate a test that only passes on retry. Cypress is one of the best web testing frameworks created so far. With more and more new… Read more »
2 minutes readKey takeaways Selenium IDE is the fastest way to capture a journey, and the worst way to maintain one. Use it to draft a test, then export and clean up the locators. Recorded selectors break on the next layout change, so treat them as a starting point. Selenium IDE is an open source record and… Read more »
2 minutes readKey takeaways TestCafe runs without WebDriver, which removes most of the setup friction. Its automatic waiting cuts the flakiness that manual sleeps introduce. End to end tests are expensive, so keep them to the journeys that matter. TestCafe is a Node.js tool to automate end-to-end web testing. What are End to End tests and why… Read more »
< 1 minute readKey takeaways Running Lighthouse from an extension gives you an audit without leaving the page you are testing. One-off audits vary. Schedule them if you want a trend rather than a snapshot. Treat the score as a prompt for investigation, not a target to optimise for its own sake. LoadFocus uses Google Lighthouse as a… Read more »
2 minutes readKey takeaways FMP tried to capture when the main content appeared, rather than the first pixel. It proved hard to measure consistently and has largely been replaced by LCP. Use LCP for anything current; FMP appears mainly in older reports. Time to First Meaningful Paint (TTFMT) is a modern performance metric introduced by few engineers… Read more »
