Performance
Weighted aggregate of LCP, FCP, TBT, CLS, and Speed Index. Below 90, the page feels slow even with good content. Google uses these metrics as ranking factors.
Test your site’s real speed
Enter a URL. You get Lighthouse scores and Core Web Vitals from Google PageSpeed Insights, the same data search uses for ranking signals.
Get a free estimateWhat each score means
Lighthouse splits the audit into four categories. Each measures something different. A site can score SEO 100 and Performance 40 at the same time.
LCP (Largest Contentful Paint)
Time until the largest on-screen element appears: usually a hero image, video, or large text block. Google’s “good” threshold is under 2.5 seconds. Most often caused by an uncompressed image or slow server.
CLS (Cumulative Layout Shift)
Sum of unexpected layout shifts. If a cookie banner pushes a button down right as you tap, CLS increases. Target: under 0.1.
INP (Interaction to Next Paint)
Time from click or tap until the page responds visually. Replaced FID in March 2024. Heavy JavaScript or too many third-party scripts (chat, analytics, pixels) degrade it.
SEO
Checks if the page is indexable: unique title, meta description, crawlable links, hreflang if multilingual, valid structured data. Does not replace a content audit, but catches technical blockers.
Best Practices
HTTPS, up-to-date library versions, images with correct aspect ratios, no critical console errors. Signals the site is maintained, not abandoned.
Why speed matters
Google does not rank pages on keywords alone. It also compares how fast the user gets an answer to their search intent. A slow site loses on two fronts: ranking and conversion.
Lab data vs. real users
PageSpeed Insights shows two sources: the Lighthouse simulation (controlled network and device) and CrUX (Chrome UX Report), aggregated from real users. Good lab scores with poor CrUX mean the problem is your actual audience: slow mobile networks, older devices, or a server far from Romania.
Google indexes the mobile version
With Mobile-First Indexing, Googlebot sees your site as it renders on a phone. Always test the mobile strategy. A site that loads in 1 second on desktop but 6 on 4G is judged on the slow version.
LCP is usually one resource
In most cases, slow LCP comes from one large JPEG or web fonts blocking render. A full redesign is often unnecessary: WebP/AVIF, correct dimensions in HTML, and preload on the LCP resource fix many cases.
Unused JavaScript costs time
Total Blocking Time measures how long the main thread is blocked by JS. WordPress plugins, tag managers with 15 scripts, and chat widgets add up. The user sees the page but cannot interact. TBT above 300ms feels like lag.
Speed also affects Google Ads
Landing Page Experience feeds into Quality Score. A slow page means higher CPC for the same keyword. You pay extra per click versus a competitor with a fast site, even with an identical ad.
What you can fix yourself
Compressed images, lazy loading below the fold, missing meta tags, wrong robots.txt: quick CMS fixes. CLS from ads or widgets, TBT from WordPress architecture, or LCP above 4 seconds on mobile: usually needs a refactor or migration to a modern stack.
How to read the score
90-100
Good
Metrics pass Google’s “good” threshold. The page responds before users feel delay. Does not guarantee position 1, but removes a negative factor.
50-89
Needs work
Mix of OK and weak metrics. Small ranking losses that compound over time. Users sometimes feel lag on scroll or click.
0-49
Poor
Below the “poor” threshold. Google flags the experience as deficient. Bounce rate rises, and Ads cost more per visit.
Frequently asked questions
Yes, both use the Lighthouse engine. The difference: PageSpeed Insights runs on Google’s servers with standardized settings. DevTools on your laptop may differ if you have extensions or a faster network.
Lighthouse simulates different devices and networks. Mobile uses slower CPU and 4G. Google indexes the mobile version, so the mobile strategy is what matters for SEO.
No. We send the URL to Google’s API and display the response. Nothing stays on KODERS servers.
Practical target: 90+ Performance on mobile. Below 70, the impact on experience and ranking is clear. The Lighthouse SEO score checks technical indexability, not your keyword positions.
Opportunities like “serve images in next-gen formats” or “reduce unused JavaScript” are actionable in a CMS or via plugins. If TBT stays high after surface fixes, the problem is architectural. That is where we step in.
Architecture issues do not fix themselves with plugins.
If TBT or LCP stay poor after surface optimizations, the way the site is built needs to change.