To improve Core Web Vitals, measure your real-user scores in Search Console, fix the pages rated "Poor" first, and work through the three metrics in order: loading (LCP), visual stability (CLS) and responsiveness (INP). Do it for the right reason. Good scores will not lift a page on their own, but they protect the leads you already earn from search, maps and ads. A slow, jumpy site loses visitors before your offer is read. We treat speed as part of a site that is built to convert, and that is how this guide approaches it.
TL;DR
- Core Web Vitals are three real-user measurements: LCP (loading), INP (responsiveness) and CLS (visual stability). A page passes when all three are "good" for at least 75% of visits.
- Start in Search Console, fix "Poor" URLs first, then the groups with the most pages or the most revenue behind them.
- The biggest wins are usually the hero image, heavy scripts and elements that load without reserved space.
- Passing does not replace relevant content, but it decides between pages that are otherwise comparable, and it helps visitors turn into leads.
- Field data takes about 28 days to reflect a fix, so judge progress in weeks, not days.
Table of Contents
- What Core Web Vitals measure, in plain terms
- Check where you stand before you change anything
- Prioritize: what to fix first
- How to improve LCP (loading)
- How to improve CLS (visual stability)
- How to improve INP (responsiveness)
- How long until Search Console shows the improvement
- Do Core Web Vitals help rankings?
- The better reason: speed and conversions
- A 30-day plan for a small business site
- Where we can take the work off your hands
- Our point of view
- FAQ
- Sources
What Core Web Vitals measure, in plain terms
Core Web Vitals are Google's way of scoring how a page feels to a real visitor. Google describes them as real-world measures of loading, interactivity and visual stability that it recommends for success with Search. There are three:
- Largest Contentful Paint (LCP): how long the biggest visible element, usually a hero photo or headline block, takes to appear.
- Interaction to Next Paint (INP): how quickly the page responds after a tap, click or keypress. INP replaced First Input Delay on March 12, 2024, so older guides that mention FID are out of date.
- Cumulative Layout Shift (CLS): how much the page jumps around while it loads. It is the reason you tap "Call now" and hit an ad instead.
The thresholds are published by Google in its web.dev definition of the thresholds:
| Metric | Good | Needs improvement | Poor |
|---|---|---|---|
| LCP | 2.5 seconds or less | 2.5 to 4 seconds | Over 4 seconds |
| INP | 200 milliseconds or less | 200 to 500 milliseconds | Over 500 milliseconds |
| CLS | 0.1 or less | 0.1 to 0.25 | Over 0.25 |
A page passes only when it meets all three targets at the 75th percentile of page loads, reported separately for mobile and desktop. In practice, that means three out of four visits must have a good experience. If one visitor in four is on an older phone with a weak signal, that visit counts. This matters for local businesses, because many of your customers search on a phone, often while out and about. Our mobile search optimization guide covers why that audience deserves the most attention.
Check where you stand before you change anything
Guessing wastes budget. Two free Google tools give you a real picture, and they answer different questions.
PageSpeed Insights: one page, last 28 days
PageSpeed Insights shows page-level and origin-level performance for the past 28 days, plus a list of suggestions. Enter your home page, your top service page and your contact page. Read the field data section at the top first. That is what real visitors experienced. The lab score below it is a simulated test, useful for finding causes but not the number Google assesses.
A common surprise: a page can show a performance score in the high 90s and still fail the assessment. The score is a lab estimate. The assessment uses real visits, including slow phones and poor connections.
Search Console: the whole site at once
The Core Web Vitals report in Search Console uses real-user field data for LCP, INP and CLS and groups similar URLs together. A group takes the status of its worst metric. So one weak metric holds a whole group of pages back.
For a typical local business site, you will see a few groups: service pages, location pages, blog posts and perhaps a store or booking template. Fixing a template fixes every page built from it, which is why this report is the best starting point.
If a page does not appear, it may not have enough traffic for Google to report on it. That is normal for small sites. Test your key pages in PageSpeed Insights and rely on the origin-level view for the overall picture.
Prioritize: what to fix first
Google's own advice for non-technical owners is simple. Fix all "Poor" issues first, then prioritize by the number of affected URLs or by importance. The same page suggests keeping mobile pages under 500 KB with fewer than 50 resources.
A practical order for a small business:
- Pages that earn money. Your top service pages, your booking or quote page, and any landing page your ads send traffic to.
- Templates over single pages. One fix to a shared template improves dozens of URLs.
- The worst metric in each group. Improving a metric that already passes changes nothing.
- The checkout or form step. Research on retail data found customers are most sensitive to delays at the checkout stage. For a service business, the equivalent is the quote form and the thank-you step.
The web.dev team also makes a point that owners should hear: optimizations cannot be judged on technical merit alone, and the most realistic, highest-impact wins come first, not theoretical perfection. A fix that your site can actually ship this week beats an ideal one that sits in a backlog.
How to improve LCP (loading)
On most small business sites, LCP is the first metric to fix because the cause is usually obvious: one large element is slow to arrive. Look at what PageSpeed Insights names as your LCP element. It is almost always a hero image, a slider or a large heading block.
Work through these in order:
- Shrink the hero image. Resize it to the largest size it is actually displayed at, compress it, and use a modern format such as WebP or AVIF. A 3 MB photo straight from a camera is the single most common cause of a poor LCP.
- Do not lazy-load the hero. Lazy loading is for images below the fold. The main image should load right away, and you can tell the browser it is a priority.
- Cut what blocks the first paint. Large style files and scripts that load before the page content delay everything. Remove unused ones and delay the rest.
- Speed up the server response. If the first byte takes a long time, nothing else can start. Caching, a content delivery network (a network of servers that serves files from a location near the visitor) and better hosting all help here.
- Remove the slider. Rotating banners load several large images to show one. A single strong image with a clear headline usually loads faster and converts as well.
There is also a lesser-known gain. The back/forward cache lets the browser restore a page instantly when someone presses Back. web.dev reports that it improves LCP and eliminates layout shifts, yet many sites are still ineligible for it. If your pages do not load sensitive information, check whether your site qualifies.
How to improve CLS (visual stability)
CLS is about surprises. Something loads late, pushes the content down, and the visitor taps the wrong thing. web.dev's research found that shifts of 0.15 or more were consistently seen as disruptive, which is why the good threshold sits at 0.1.
The usual causes and fixes:
- Images and videos without dimensions. Set width and height, or reserve the space with an aspect ratio, so the browser knows how much room to leave.
- Ads, embeds and chat widgets. Give each one a fixed-size container. A review widget or booking embed that appears late should not push your headline around.
- Web fonts that swap. When a fallback font is replaced by your brand font, the text can reflow. Preload the key font and choose a fallback with similar proportions.
- Banners injected above content. Cookie notices and promo bars should overlay the page or sit in reserved space, not shove everything down.
CLS has one quirk worth knowing. CLS is still measured only in Chromium-based browsers, while LCP and INP can now be measured across all major browser engines since Safari 26.2 shipped in December 2025. Your Search Console numbers reflect Chrome-family users, so test there.
How to improve INP (responsiveness)
INP is the newest of the three and the one owners understand least. It measures the delay between a visitor's action and the next visual update, across the whole visit, not just the first tap. A menu that takes half a second to open on a phone is an INP problem.
The cause is almost always too much JavaScript running at once. Every script competes for the same processing time, and when the browser is busy, it cannot respond. Typical culprits on small business sites:
- Several analytics and tracking tags installed over the years, some no longer used.
- Chat widgets, pop-up tools and heat-map scripts loading on every page.
- Page builders and plugins that load their full code on pages that use a fraction of it.
- Heavy sliders, animations and scroll effects.
The fix is to audit what runs. List every script, ask what it contributes to leads or reporting, and remove or delay anything that does not earn its place. Wharton researcher Santiago Gallino recommends weighing the benefit of added tracking and features against the slowdown they cause. That is a good rule for any marketing stack.
To diagnose INP, web.dev suggests looking at CrUX data in PageSpeed Insights or a real-user monitoring tool and starting from the slowest interactions. If your site is built on a page builder, many INP gains come from removing what the page does not need rather than adding a speed plugin.
How long until Search Console shows the improvement
Expect to wait. The field data behind both tools is a rolling 28-day window, so a fix made today moves the number gradually. PageSpeed Insights reports the past 28 days, and a day after your fix you are only a small fraction of the way into a fresh window.
In Search Console, once you have fixed an issue, click Start Tracking. This begins a 28-day validation session, and the issue counts as fixed only if no URL shows it throughout that window. Use the lab results in PageSpeed Insights in the meantime to confirm you are moving in the right direction.
Plan on roughly a month between shipping a fix and seeing a green status. That timeline is worth telling anyone who is waiting for a report.
Do Core Web Vitals help rankings?
They help, but less than most guides imply. Google presents Core Web Vitals as part of the broader page experience its ranking systems aim to reward, and it does not promise a ranking gain from passing. Relevance and content quality remain the stronger factors. Passing mainly matters when Google already rates two pages as similarly relevant.
For a local business, that tells you where to spend. If your service pages are thin, your Google Business Profile is incomplete or your pages are not being indexed, fix those first. We cover the common reasons in why Google may be ignoring your website and in our piece on Google indexing problems. Our SEO best practices guide shows where page experience fits among the other priorities.
When you are already competing well in the map pack and in organic results, good vitals are the polish that keeps you from losing to a similar competitor with a faster site.
The better reason: speed and conversions
The stronger case for fixing Core Web Vitals is what happens after the click. Visitors who wait leave, and the leads you paid for in SEO or ads disappear.
The best-known study is Deloitte's Milliseconds Make Millions, covering 37 European and American brands and over 30 million sessions. It found that a 0.1-second improvement across four speed metrics was associated with 10% higher travel booking rates, 9.2% higher retail average order value and a 21.6% increase in lead-generation form progression. The full Deloitte report also reports an 8.3% mobile bounce-rate reduction on lead-generation information pages.

Academic work points the same way. A Wharton interview about a peer-reviewed study of retailer data reports that a 10% slowdown in site speed can cut online sales by 4.2% and conversion rates by 2%, and that shoppers are most sensitive to delays late in the purchase process.
How to read these numbers
Treat them as direction, not a promise. The Deloitte data was collected over 30 days at the end of 2019, mostly from large brands, and it measures association, not a controlled test. Your own gain will depend on your traffic, your offer and your starting point.
Here is a simple way to apply it. Suppose your quote form page gets 400 mobile visits a month and 20 of them submit the form. Your own numbers will differ, but the logic holds: if faster loading moves even a few more visitors from "reading" to "submitting", the extra leads come from traffic you already have, with no added ad spend. Compare that to the monthly cost of buying those clicks, and speed work often looks like one of the cheapest ways to get more leads.
That is why we judge every site change on return, not on a score. A page that passes Core Web Vitals but has a weak offer, buried phone number or confusing form still leaves leads on the table. Our guides on conversion rate optimization and turning your website into a conversion machine show what to do once the page loads quickly.
A 30-day plan for a small business site
Week 1: measure. Record your Search Console status and run PageSpeed Insights on your five most important pages. Write down the LCP element and the failing metric for each.
Week 2: fix the biggest cause. Compress and resize the hero image, remove sliders, and set dimensions on images and embeds. These three changes usually address LCP and CLS together.
Week 3: audit scripts. List every tag, plugin and widget. Remove what you no longer use and delay chat and tracking tools that do not need to run immediately. This is the INP work.
Week 4: validate and watch. Click Start Tracking in Search Console for the fixed issues. Keep a record of leads from the pages you changed, so you can see whether speed moved the business number, not just the score.
If the issue sits in the site's design or template, a patch will not hold. A site built from the start for fast loading and clear conversion paths avoids the problem.
Where we can take the work off your hands
Core Web Vitals are one piece of a site that has to be found and also has to turn visitors into customers. Our High Converting Website Design For Small Business service is built around conversion, and it is part of a wider set of services that includes local SEO, Google Business Profile optimization and reputation management. For businesses that want visibility and leads handled together, Local Domination is listed at $450 per month, and Be Found Everywhere at $1,595 per month. If you would rather pay for outcomes, Pay Per Result SEO is listed at $650 to $2,500 per month. You can see everything we offer on our services page, or start from the Digital Marketing All home page to talk about your site.
Our point of view
Most owners come to us with a PageSpeed score and a worry. The score is not the goal. The goal is a site that Google can find, that a customer can use in the first few seconds and that gives them an easy way to call or book.
We look at everything from a return-on-investment view. A speed fix that protects the leads from your map-pack ranking is worth doing. A week spent chasing the last few points of a lab score is not. That is why we start with the pages that earn money and the template behind them.
We also think visibility has widened. People now find businesses through search results, maps, autosuggest and AI recommendations. A fast, stable, clearly written site supports all of those at once, and it is the foundation for the AI-driven and local strategies we build for small and medium-sized businesses.
If your site is slow because it was built without conversion in mind, the better answer is a site designed properly from the start, with local SEO and reputation work running alongside it.
Diane O'Brien
FAQ
What is a good Core Web Vitals score?
There is no single score. Each metric has its own target: LCP of 2.5 seconds or less, INP of 200 milliseconds or less and CLS of 0.1 or less. Google assesses them at the 75th percentile of page loads, and you must meet all three to pass.
How do I pass the Core Web Vitals assessment?
Check Search Console to see which metric is failing, then fix the cause: oversized images for LCP, missing dimensions and late embeds for CLS, and excess scripts for INP. After the fix, use Start Tracking to begin the 28-day validation. A group passes when its worst metric is good.
Will improving Core Web Vitals raise my Google rankings?
Not on its own. Google recommends good vitals for search success, but relevance and content quality matter more. Think of passing as a tiebreaker between similar pages and as protection for your conversion rate.
Does page speed really affect leads and sales?
Research says yes, with caveats. Deloitte's study of more than 30 million sessions linked a 0.1-second improvement to a 21.6% increase in lead-form progression. Treat that as a direction, since your result depends on your traffic and offer.
Sources
- Understanding Core Web Vitals and Google search results
- How the Core Web Vitals thresholds were defined
- Core Web Vitals report, Search Console Help
- Web Vitals overview, web.dev
- INP becomes a Core Web Vital, web.dev
- Getting started with measuring Web Vitals
- Optimizing Core Web Vitals, web.dev
- How to improve Core Web Vitals, Vercel
- Milliseconds make millions, web.dev case study
- Milliseconds Make Millions, Deloitte report
- Why Online Retailers Have a Need for Speed, Wharton Magazine
- Need for Speed: The Impact of In-Process Delays on Customer Behavior, Operations Research
