Core Web Vitals: How Does Page Speed Affect SEO and Conversion Rates?

Page speed isn’t just a technical issue. If a user has to wait a long time for the main content, buttons are slow to respond, or interface elements shift while loading, it’s not just the user experience that suffers. The risk of losing a visit, a lead, or a sale also increases.

Core Web Vitals help evaluate three elements of this experience: how quickly the main content loads, the page’s responsiveness to interactions, and visual stability. Google uses these metrics in its ranking systems, but at the same time emphasizes that a good Core Web Vitals score does not guarantee a high ranking. The relevance and quality of content remain crucial.

Therefore, when designing and developing fast websites, Core Web Vitals should be treated as one of the elements of a site’s quality, alongside content, UX, architecture, and conversion path effectiveness.

Key Takeaways

• Core Web Vitals currently include LCP, INP, and CLS. Good values are, respectively: LCP up to 2.5 seconds, INP up to 200 ms, and CLS up to 0.1.
• Assessments should be based primarily on real-user data, while lab results are mainly used to diagnose the causes of problems.
• Core Web Vitals can support SEO and user experience, but they do not replace good content, proper intent, or correct site architecture.
• The greatest business value comes from addressing issues on pages that generate traffic, leads, or sales.

What are Core Web Vitals, and what exactly do they measure?

LCP – Largest Contentful Paint

Largest Contentful Paint measures the time it takes to render the largest visible content element on the screen, such as a hero image, a large banner, or a block of text. A good LCP score is up to 2.5 seconds. Scores between 2.5 and 4 seconds need improvement, and scores above 4 seconds are considered poor.

INP – page responsiveness

Interaction to Next Paint evaluates a page’s responsiveness to user interactions, such as clicking a button, expanding a menu, or selecting a form field. A good INP score is 200 ms or less. Values in the 200–500 ms range require improvement, and values above 500 ms indicate poor responsiveness.

CLS – Visual Stability

Cumulative Layout Shift measures unexpected shifts in page elements while using the site. A good CLS score is 0.1 or lower; a score between 0.1 and 0.25 requires improvement; and a score above 0.25 is poor. A common issue occurs when a user tries to click a button, but at the last moment, the element shifts due to a newly loaded image, ad, or banner.

In the Core Web Vitals assessment, the 75th percentile of user experiences is what matters, separately for mobile and desktop devices. This means that it’s not about a single “good test,” but rather whether the vast majority of visits fall within the recommended thresholds.

How Do Core Web Vitals Affect SEO?

Google confirms that Core Web Vitals are used by ranking systems as one component of a broader assessment of page experience. However, there is no single “page experience signal” that alone determines a page’s ranking. A technically fast page that poorly addresses the user’s intent will not automatically outrank a result that is substantively better.

In practice, this means that optimizing Core Web Vitals makes the most sense when the SEO foundation is already sound: the page answers the right query, is indexable, has valuable content, and has a logical structure. Only then can issues with loading speed, responsiveness, or stability become an additional limitation on the quality of the user experience.

How does page speed affect conversion?

The impact on conversion is more direct than the impact on rankings. Users don’t see technical metrics. Instead, they see whether the content loads quickly, whether the form responds immediately, and whether the interface remains stable.

A poor LCP can delay the display of a key message or CTA. A high INP can cause a user to click a button and receive no response for a moment. A high CLS can shift a form or button just before a click. Each of these issues creates friction in the user journey.

This does not mean that improving CWV will always result in an identical increase in sales. The impact depends on the type of website, traffic source, devices, and stage of the funnel. However, there are well-documented business examples. In an A/B test of an optimized landing page, Rakuten 24 recorded a 33.13% increase in its conversion rate, and redBus reported approximately a 7% increase in sales after improving its INP by 72%. These are results from specific implementations, not universal benchmarks, but they clearly demonstrate that performance can have a measurable business impact.

How should you interpret the results, and where can you check Core Web Vitals?

The most important distinction concerns real-user data versus lab data. The Chrome UX Report (CrUX) collects data from real visits and presents it in tools such as PageSpeed Insights and Search Console. In CrUX, values are calculated over a rolling 28-day window, so the real-world result does not change immediately after implementing a fix.

PageSpeed Insights combines data from CrUX with the Lighthouse lab test. Real-world data answers the question, “How does the page perform for real users?” while lab data helps answer, “What might be causing the problem, and what should be checked?” If the two results differ, this does not indicate an error in the tool. The lab test is conducted under controlled conditions, whereas users access the site using various devices, networks, and configurations.

Search Console is a good starting point for larger websites because it groups similar URLs and shows which areas of the site have issues. PageSpeed Insights and developer tools work better later on, during the diagnosis of a specific subpage.

How can you improve poor Core Web Vitals scores?

Optimization should begin by identifying the specific metric and the element that is causing it to suffer. “Speeding up the site” without a proper diagnosis often results in a series of changes that have little impact on the actual user experience.

When LCP is the problem

• Check what the LCP element is. It’s often the hero image or a large block of content visible immediately upon landing on the page.
• Ensure that the key image is detectable in the HTML and loaded with high priority. Do not use lazy loading for an image that is the LCP element.
• Minimize render-blocking resources and check the server response time. With a high TTFB, even a well-optimized image will start loading too late.

When INP is the problem

• Look for long-running JavaScript tasks that block the browser’s main thread.
• Minimize unnecessary code executed on startup and external scripts that compete for resources.
• Test real-world interactions: forms, filters, menus, search, and CTAs. A good initial view isn’t enough if the page starts to stutter during use.

When CLS is the problem

• Set dimensions or aspect ratios for images, videos, and embedded elements so the browser can reserve space for them before loading.
• Pay attention to cookie banners, ads, widgets, and dynamically reloaded modules.
• Avoid animating CSS properties that change the page layout if you can achieve the same effect without moving other elements.

What should you fix first?

The priority should not be the URLs with the worst technical performance, but rather those where the problem has the greatest business impact. A campaign landing page, service page, product page, or lead-generation form usually deserves attention before a subpage with marginal traffic.

It’s good practice to combine CWV data with analytics, user behavior, and a UX audit. If the drop in conversions occurs exclusively on mobile devices, and at the same time the problematic URLs have poor INP or LCP scores, this suggests a hypothesis that can be verified. A red status in the report alone does not provide such a diagnosis.

Example: A Slow Landing Page and Loss of Conversions

Let’s assume that a campaign’s landing page has an attractive offer and generates valuable traffic, but its main hero image is large and loads only after JavaScript executes. The LCP on mobile is poor, and the user cannot see the most important message and CTA for several seconds.

The diagnosis shouldn’t stop at simply compressing all the images on the site. First, you need to determine whether the hero image is actually an LCP element, whether its URL is immediately available in the HTML, whether lazy loading is enabled, and whether it’s being blocked by earlier resources. After fixing the loading process, we retest the page and then monitor real-user data and the landing page’s conversion rate.

Common Mistakes in Core Web Vitals Optimization

• Treating a PageSpeed score of 100/100 as an end in itself.
• Making decisions based solely on a single lab test.
• Optimizing all subpages simultaneously without establishing business priorities.
• Ignoring analytics scripts, ad scripts, chat scripts, and other external components that affect INP.
• Implementing changes without re-measuring and monitoring post-implementation real-world data.

When Should You Measure Core Web Vitals?

It’s a good idea to monitor Core Web Vitals on a regular basis, but especially after a migration, redesign, change in front-end technology, implementation of new marketing scripts, addition of widgets, or major changes to the hero section. Monitoring is also important because a website can gradually lose performance as its content and features expand.

Expert Insight

The biggest mistake is treating Core Web Vitals as a separate technical project that ends once you achieve green metrics. Good optimization starts with the question: where is performance preventing the user from completing an important action? Only then do you choose the metric and the fix.

FAQ

Are Core Web Vitals a ranking factor?

Google uses Core Web Vitals in its ranking systems, but they are not a standalone guarantee of ranking. Relevance and content quality remain paramount, and user experience is just one of many factors taken into account.

Does a 100/100 score on PageSpeed Insights mean the site is well-optimized?

No. The Lighthouse score is the result of a lab test and helps with diagnostics, but it does not replace real-user data. A website can have a high score while still causing issues under specific conditions or during certain interactions.

Which is more important: lab data or real-user data?

Field data, such as CrUX, is more important for evaluating the real-world experience. Lab data, on the other hand, is very useful for diagnosing causes and testing changes before implementation.

How long does it take to see a change in results after optimization?

Laboratory results can be checked immediately after implementation. CrUX data, however, is based on a rolling window of the previous 28 days, so improvements in field data appear gradually as new visits come in.

Summary

Core Web Vitals allow you to translate part of the user experience into specific metrics: LCP describes how quickly the main content is displayed, INP describes responsiveness, and CLS describes layout stability. Improving these metrics can support SEO and, above all, reduce technical friction along the user journey.

The best optimization isn’t about chasing the highest score in a tool. It’s about identifying a problem that affects real users, determining its business impact, implementing the right fix, and remeasuring the effect.

You can find more resources on SEO, UX, and website performance on our blog.