Website speed for Indian mobile visitors: a Core Web Vitals guide
Most visitors to an Indian business website arrive on a phone, and many of those phones are budget or mid-range Android models on a mobile connection that drops from 4G to something slower between one street and the next. A site that feels instant on an office laptop over fibre can take several seconds to show anything useful in that setting, and visitors who wait rarely wait long.
Core Web Vitals are Google's way of measuring that experience from real visits. They cover three things: how fast the main content appears, how quickly the page responds when someone taps, and whether the layout jumps while loading. This guide explains each metric and its thresholds, the difference between field and lab data, the culprits we see most often on Indian sites, how to fix them and how to keep them fixed.
What LCP, INP and CLS measure
Each metric has a good threshold, and Google assesses it at the 75th percentile of page loads, split between mobile and desktop. A page passes on mobile only if at least three quarters of mobile visits meet the threshold, so a fast median can hide a slow tail: the slowest quarter of your visitors decides the result.
| Metric | What it measures | Good | Poor |
|---|---|---|---|
| Largest Contentful Paint (LCP) | Time until the largest image or text block in the viewport is rendered | 2.5 seconds or less | More than 4 seconds |
| Interaction to Next Paint (INP) | Delay between a tap, click or key press and the next visual update, across the visit | 200 milliseconds or less | More than 500 milliseconds |
| Cumulative Layout Shift (CLS) | How much visible content moves unexpectedly while the page is in use | 0.1 or less | More than 0.25 |
INP replaced First Input Delay as a Core Web Vital in March 2024. The change matters for sites heavy with scripts, because INP looks at responsiveness throughout the visit rather than only the first interaction. A page that loads quickly but freezes when someone opens a menu or a form now fails.
Core Web Vitals feed into Google's page experience signals, but Google is clear that relevant content still matters more. Treat them as a measure of whether visitors can use your site, not as a ranking trick.
Field data and lab data
There are two kinds of measurement, and confusing them causes most of the arguments between marketing teams and developers.
- Field data comes from real visits. The Chrome UX Report (CrUX) collects it from Chrome users who have opted in, over a rolling 28-day window. PageSpeed Insights shows it at the top of the report, and Search Console's Core Web Vitals report groups your URLs by it. This is what Google assesses.
- Lab data comes from a single simulated load, usually Lighthouse, which PageSpeed Insights runs with an emulated mid-range phone on a throttled connection. It is repeatable and good for diagnosing problems, but it is one load, not your audience.
- A standard Lighthouse page-load test cannot measure INP, because no one taps the page. Total Blocking Time in Lighthouse is the closest lab signal for responsiveness.
Smaller sites often have no page-level field data, because CrUX needs enough traffic. PageSpeed Insights then falls back to origin-level data, or shows none. In that case, collect your own field data with the open-source web-vitals JavaScript library and send it to your analytics, so decisions rest on your visitors rather than on one lab run.
Why mid-range Android phones and patchy networks matter
Most smartphones in India run Android, and a large share of them are budget and mid-range models. Their processors take much longer than a flagship phone or laptop to parse and run JavaScript, so a script bundle that costs little on a developer's machine can block the main thread for a long time on a visitor's phone. That shows up as slow INP and, often, slow LCP.
Networks add a second layer. Coverage varies between buildings, basements, lifts and moving cabs, and latency matters as much as bandwidth: every extra connection to a new domain for a font, a tag or a widget costs round trips before anything downloads. Many visitors also watch their data use, so a page that pulls in several megabytes of video and images costs them money as well as time.
The practical rule is to test where your visitors are. Keep a real mid-range Android phone in the office, open the site on mobile data away from Wi-Fi, and compare it with your field data. A website designing company in Gurgaon that tests only on desktop browsers will miss most of what your visitors experience.
Fixing slow LCP: hero images, servers and fonts
On most business sites the LCP element is the hero image or the main heading. LCP has four parts: server response time, the delay before the browser starts fetching the LCP resource, the time to download it, and the delay before it is rendered. Find which part is large before changing anything.
Hero images
Serve the hero in AVIF or WebP, at the size the screen actually needs, using responsive images. Never lazy-load the LCP image, and give it fetchpriority set to high so the browser fetches it early. If the hero is set as a CSS background or injected by JavaScript, the browser discovers it late; a plain image element in the HTML is usually faster.
Server response
A slow server delays everything. Host close to your visitors or behind a CDN with points of presence in India, cache pages that do not change per visitor, and watch for slow database queries on WordPress sites with many plugins.
Web fonts and fallback metrics
If the LCP element is text, a late web font can hold it back. Self-host fonts, preload only the one or two files needed above the fold, subset them to the characters you use, and set font-display so text appears in a fallback font straight away. Indian-language fonts carry many more glyphs than Latin fonts, so subsetting matters even more for Hindi or other Indian-script pages. To stop the swap from moving text, adjust the fallback font's metrics with size-adjust and the ascent and descent overrides, so both fonts take up the same space.
Common culprits and their fixes
This table pulls the most frequent problems together. Start with the rows that match the metric failing in your field data.
| Culprit | Metric hit | Fix |
|---|---|---|
| Oversized hero image | LCP | AVIF or WebP, responsive sizes, fetchpriority high, no lazy loading |
| Hero set as CSS background or by script | LCP | Plain image element in the HTML |
| Slow hosting or no caching | LCP | Page caching, CDN with Indian points of presence, faster queries |
| Web fonts from several domains | LCP, CLS | Self-host, subset, preload one or two files, tune fallback metrics |
| Bloated tag manager container | INP | Remove unused tags, delay the rest until after load |
| Chat or WhatsApp widget | INP, LCP | Lightweight button that loads the widget on tap |
| Hero slider or carousel | LCP, CLS | Single static hero, or reserved space and no injected slides |
| Images and embeds without dimensions | CLS | Width and height or aspect ratio on every element |
| Cookie banner or sticky bar pushing content | CLS | Overlay or reserved space |
| Large JavaScript bundles | INP, LCP | Split code, load per page, remove unused libraries |
How to monitor Core Web Vitals
Speed decays. A new campaign adds a pixel, a page builder update adds a script, a marketing team uploads a large banner. Monitoring catches the slide before visitors and rankings feel it.
- Check Search Console's Core Web Vitals report monthly, and open any URL group that moves to needs improvement or poor.
- Run PageSpeed Insights on your key templates after every release: home, service, location, product or project, and landing pages.
- Collect field data with the web-vitals library and report it by template and device in your analytics.
- Set a page weight and script budget per template, and fail a release that breaks it.
- Review the tag manager every quarter with the marketing team.
Building speed into a website project
Speed is far cheaper to design in than to retrofit. The decisions that decide Core Web Vitals, such as the hero format, the font choice, the number of third-party tools and whether a slider is needed at all, are made in design, not development. Agree targets in the brief and test them on staging, on a mid-range phone, before launch.
If your current site fails on mobile, a focused website redesign can often fix templates without starting again. If you are planning a new site, a website design company in Gurgaon should show you field data from sites it has launched and explain how it will meet LCP, INP and CLS targets for your visitors. We build to these targets from our South City 1 office as a web design agency in Gurugram, testing on the phones and networks your customers use.