The speed cost most teams have not measured
Page speed has been a confirmed Google ranking factor since 2010 and a mobile ranking factor since 2018. Core Web Vitals, Google's user experience metrics covering loading performance, interactivity, and visual stability, have been a ranking signal since 2021. Most marketing professionals are aware of this at a conceptual level. Fewer have actually measured whether their website's speed is materially affecting their organic rankings or their conversion rates.
The research on this is consistent and significant. A one-second delay in page load time reduces conversions by an average of 7%. Pages that load in five seconds have 70% longer average sessions than pages that load in nineteen seconds. For mobile users, where the baseline experience is often slower than desktop, the impact is even more pronounced. A website that feels slow is not just a mildly worse experience; it is actively losing business that it would otherwise capture.
What Core Web Vitals actually measure
Google's Core Web Vitals framework measures three specific dimensions of page experience. Largest Contentful Paint measures loading performance: specifically, how quickly the largest visible content element on the page loads. A good LCP score is under 2.5 seconds. Interaction to Next Paint measures responsiveness: how quickly the page responds to user interactions like clicks and keyboard inputs. First Input Delay was the previous metric here; INP, which became the official metric in 2024, is a more comprehensive measure of overall interactivity. Cumulative Layout Shift measures visual stability: the degree to which page elements move around as the page loads, which creates a jarring experience that can cause users to click on the wrong element.
These three metrics are measurable, improvable, and connected to both search performance and user experience. A site that scores well on all three is both more likely to rank competitively and more likely to convert the visitors it attracts.
Every second of load time is a percentage of the audience you are not converting. Most websites are losing meaningful revenue to page speed problems that are entirely fixable.
How to measure your current performance
Google's PageSpeed Insights tool, available free at pagespeed.web.dev, provides a Core Web Vitals assessment for any URL with a single query. It reports field data from real users where available, and lab data for pages without sufficient real-user data. The report shows the score for each Core Web Vitals metric, a list of specific issues contributing to poor performance, and guidance on what to fix.
Running this assessment on your three to five most important pages, specifically the homepage, main product or service pages, and key landing pages, gives a realistic picture of where speed problems are concentrated and how severe they are. Most websites have two or three specific issues that account for the majority of the speed problem, and those issues are often more straightforward to fix than the overall score suggests.
The most common speed problems and their fixes
Unoptimised images are the most common cause of slow LCP scores. Images that are served at full resolution and without compression when a smaller, compressed version would render identically in the browser. The fix is straightforward: compress images before upload, serve them in modern formats like WebP where browser support allows, and implement lazy loading for below-fold images so they do not block the initial page render.
Render-blocking resources, typically JavaScript files and CSS stylesheets that are loaded in a way that prevents the page from rendering until they are complete, are the second most common cause of poor loading performance. Moving non-critical scripts to load asynchronously or deferred, and ensuring that critical CSS is inlined or loaded in a way that does not block render, typically produces significant LCP improvements.
Third-party scripts, analytics tags, chat widgets, advertising pixels, and tracking scripts, can add significant load time when they are added without considering their performance impact. Auditing third-party script load and implementing a tag management approach that loads scripts without blocking page render is worth doing for most websites that have accumulated a significant number of third-party integrations.
The mobile priority
Google uses mobile-first indexing, meaning that the mobile version of a site is the primary version used for ranking assessment. A website that performs well on desktop but poorly on mobile is being evaluated primarily on its mobile performance. For most business websites, mobile traffic represents 50 to 65 percent of total visits. Treating mobile speed as secondary to desktop is both a search performance mistake and a conversion mistake.
Testing specifically on mobile, using PageSpeed Insights or Google Search Console's Core Web Vitals report filtered to mobile URLs, often reveals a performance gap between mobile and desktop that represents a significant untapped conversion opportunity.
Speed as a continuous maintenance responsibility
Website speed is not a project with a completion point; it is a continuous maintenance responsibility. New plugins, new third-party integrations, new images, and new page templates all have the potential to introduce speed regressions. Building a monthly speed check into the website maintenance routine, and treating significant score drops as issues requiring immediate investigation, maintains the performance investment over time rather than allowing it to degrade between periodic audits.