Your developer sends a PageSpeed report full of red circles and three short codes: LCP, INP and CLS. You want to know one thing. Is your website slow for customers, and does it cost you on Google?
Those three codes are Google's Core Web Vitals. They are three scores that measure how a real visitor feels your page: how fast it shows up, how fast it reacts, and whether it jumps around. This guide explains each one in plain words, the numbers that count as a pass, and how to check your own site for free.
Quick answer
What are Core Web Vitals?
Core Web Vitals are three measurements Google uses to judge the experience of a web page. LCP checks loading, INP checks how quickly the page reacts when someone taps or clicks, and CLS checks whether things move around unexpectedly. Each one has a "good" line set by Google's web.dev team.
Think of them as a health check done from your customer's phone, not from your office computer. The numbers come from real Chrome users who visit your site.
LCP: how fast the main content shows
LCP stands for Largest Contentful Paint. It measures the time until the biggest image, text block or video on the screen appears. On most business sites this is the big banner photo or the main heading at the top.
If LCP is slow, a customer on mobile data sees a blank or half-built page. Many will press back before your shop name even loads.
INP: how fast the page reacts to a tap
INP stands for Interaction to Next Paint. It looks at every click, tap and key press during a visit and reports one of the slowest. Scrolling and hovering do not count.
A bad INP feels like this: a customer taps "Add to cart" or opens your menu, and nothing happens for a moment. INP replaced an older score called FID (First Input Delay) on 12 March 2024. Any guide that still talks about FID is out of date.
CLS: does the page jump while it loads?
CLS stands for Cumulative Layout Shift. It is a score with no unit that measures how much the page moves around without the visitor asking for it. Google's own example is a person tapping the wrong button because the layout shifted at the last moment.
You have probably seen this. You go to tap "Call now", an image loads above it, and you tap something else.
What scores count as good?
Google sorts each score into three bands: good, needs improvement and poor. The table below uses the official lines from web.dev and the PageSpeed Insights documentation, as of October 2026.
| Score | Good | Needs improvement | Poor |
| LCP (loading) | 2.5 seconds or less | 2.5 to 4 seconds | over 4 seconds |
| INP (reacting) | 200 ms or less | 200 to 500 ms | over 500 ms |
| CLS (jumping) | 0.1 or less | 0.1 to 0.25 | over 0.25 |
Here "ms" means milliseconds. 200 milliseconds is one fifth of a second.
How do you pass the Core Web Vitals assessment?
You pass when all three scores are in the good band for at least 75% of real visits, measured over the last 28 days. Google checks mobile and desktop separately. If there is not enough INP data, PageSpeed can still pass a page when LCP and CLS are both good.
The 75% rule matters for Indian sites. If a quarter of your visitors are on slow phones or weak networks, their experience decides your result, not your office Wi-Fi.
Do these scores affect your Google ranking?
Yes, but less than most agencies claim. Google's Search Central documentation says Core Web Vitals are used by its ranking systems. The same page also says Google "always seeks to show the most relevant content, even if the page experience is sub-par".
So a slow page that answers the search well can still beat a fast page that does not. Google also says good scores in its reports do not mean your pages will rank at the top. Nobody can promise a first position on Google, so be careful with anyone selling a "speed fix for rankings".
The better reason to care is your customers. A visitor who waits, or taps the wrong button, is a visitor who may call your competitor instead.
How do you test your website's page speed for free?
The quickest website speed test is Google's own PageSpeed Insights. It is free, needs no login, and shows both real-user data and a lab test. Search Console also has a free report that groups all your pages.
- Open PageSpeed Insights in Chrome and paste the address of your home page.
- Wait for the test, then make sure the "Mobile" tab is selected.
- Read the top section first. It shows real-user data and says "passed" or "failed" for the assessment.
- Scroll to the lab section. It shows a performance score from 0 to 100, where 90 or above is good and below 50 is poor.
- Repeat for your most important page, such as the page your Google Ads send people to.
- In Search Console, open the Core Web Vitals report to see which groups of pages are good, need improvement or poor.
Why the lab score and the real-user score differ
The lab test loads your page once, on one simulated device and network. The real-user data comes from the Chrome User Experience Report, which collects anonymous timings from actual Chrome visitors. Your lab score can be 95 while real users still fail, because real phones and networks are slower and more varied.
When the two disagree, trust the real-user section. That is the data Google's Search Console report is built on.
What if PageSpeed says there is no data?
Small or new websites often see "no data" in the real-user section. Chrome needs enough visits to a page before it shows numbers. In that case, use the lab result as your guide and check again after your traffic grows.
What usually causes poor scores on a business website?
Most poor results on small business sites have a few common causes. The big three are heavy images, too many add-on scripts, and no space kept for things that load late. You do not need to fix these yourself, but knowing them helps you ask the right questions.
| Problem you see | Score it hurts | Common cause |
| Blank screen, then a big photo pops in | LCP | Heavy banner image, slow server, or the main image set to load late |
| Buttons do nothing for a moment | INP | Too much JavaScript (the code that makes a page interactive), often from chat, tracking or popup tools |
| Text jumps down as you read | CLS | Images and ad slots with no space reserved, or fonts loading late |
If your current site fails on most of these, a rebuild can be cheaper than patching it page by page. You can see how we plan that on The Beyond Horizon's website development service for Indian businesses.
How to improve Core Web Vitals: what a developer changes
Each score has its own set of fixes. For loading, a developer shrinks images and makes the main image load first. For reacting, they cut heavy scripts, and for jumping, they reserve space for every image and banner.
Loading (LCP)
Google's web.dev guidance says the main image should be in the first HTML the browser receives. It should never be set to load late, and it should use a modern format such as WebP or AVIF. A faster server or a CDN (a network of servers that keeps copies of your site closer to visitors) also helps.
Reacting (INP)
The usual fix is less work for the phone. Every chat bubble, popup, tracker and slider adds code that runs when someone taps. Ask which add-ons are still in use, and drop the rest.
Jumping (CLS)
Every image and video needs its width and height written into the page, so the browser keeps a space for it. Cookie banners and offer strips should appear in reserved space or at the bottom, not push the page down.
A buyer's tip before you pay for a fix
Ask the developer for a PageSpeed screenshot of your top pages on the Mobile tab, before and after the work. Remember that the real-user section covers the previous 28 days. A real fix can take up to four weeks to show fully there, so do not judge it the next morning.
Closing note
You do not need to understand code to keep these three scores healthy. Test your main pages on mobile once a month, and ask for proof whenever someone says they fixed speed.
If your report shows red and you want someone to explain it page by page, call or WhatsApp The Beyond Horizon on +91 75973 92744.