Core Web Vitals Explained for Business Owners

Core Web Vitals are three measurements Google invented in 2020 to try to put a number on how fast and smooth your website feels to a real human on a real phone. They were made a direct ranking signal in 2021 and the specific metrics have been refined twice since. In 2026 the three vitals are Largest Contentful Paint (LCP), Interaction to Next Paint (INP) and Cumulative Layout Shift (CLS). Pass all three and Google considers your "Page Experience" healthy. Fail one and you have a ranking liability.

For a local service business owner, the practical importance is this: a slow, janky mobile site now simultaneously damages (1) your Google rankings, (2) your AI engine citations, (3) your Google Business Profile ranking and (4) your conversion rate. One underlying problem creates four separate business consequences.

LCP measures how long it takes for the biggest visible element on your page to appear. On most service business websites the biggest element is the hero image or hero video at the top of the home page. LCP is measured in seconds. Google's thresholds in 2026 are: good is 2.5 seconds or faster, needs improvement is 2.5–4 seconds, poor is above 4 seconds.

The most common LCP killers on local service sites are oversized hero images (a 3 MB JPEG where a 150 KB WebP would do the same job), hero images loaded with lazy-loading accidentally applied to the above-the-fold image, custom fonts that block paint, and render-blocking JavaScript. The fixes are standard: compress images, serve WebP or AVIF, pre-connect to critical origins, defer non-critical scripts, use a CDN.

Realistic improvement expectation: on a WordPress site with a page builder, moving LCP from 6 seconds to 2.3 seconds is typically a 2–4 day project. On a well-built React or Next.js site, LCP under 2 seconds should be the starting point, not a goal.

INP replaced the older First Input Delay (FID) metric in March 2024. It measures how quickly your page responds to real user interactions — taps, clicks, scrolls — not just the first interaction but the slowest one. Google's thresholds in 2026 are: good is 200 milliseconds or faster, needs improvement is 200–500 ms, poor is above 500 ms.

INP is the metric most local service sites now fail. The reason is accumulated JavaScript bloat. Every third-party script you add — Google Analytics, Meta Pixel, Hotjar, a chat widget, a popup builder, a review widget, three A/B testing scripts — adds to the main thread workload. By the time a user taps a button, the browser is busy running analytics code and the tap feels sluggish.

The fix is editorial: remove scripts you do not actively use. Then shift what remains to defer or async loading. Then move analytics calls to lightweight server-side implementations where possible. On a bloated WordPress site, INP under 200 ms is often a 1–2 week engagement rather than a quick fix.

CLS measures how much visible content "jumps" during loading. The classic example: you are about to tap a button, the page shifts because a late-loading ad or image pushed content down, and your tap lands on the wrong element. Google thresholds: good is 0.1 or less, needs improvement is 0.1–0.25, poor is above 0.25.

CLS failures almost always come from missing dimensions. Images without width and height attributes, iframes without reserved space, web fonts that swap in with different metrics than the fallback, cookie banners that appear 2 seconds in and push the page down. Each of these creates a layout shift.

CLS is usually the easiest Core Web Vital to fix — set explicit width and height on all images and iframes, reserve space for ads and widgets, and use the font-display: swap strategy with closely-matched fallback fonts.

The failure mode we see repeatedly in audits is accumulated weight over time. The site is fast when the agency hands it over. Over 18 months, every vendor adds a script tag — the chat widget, the two analytics tools, the review widget, the appointment booking widget, the Facebook retargeting pixel, the Microsoft Clarity session recorder. Each is a reasonable addition in isolation. Together they turn a 1.8 second site into a 6 second site, and the ranking slide starts.

Our standard recommendation is a quarterly Core Web Vitals audit for any client site, specifically to catch regressions early. One bad vendor install can tank the vitals in a week; catching it in the first month is cheap, catching it after six months of ranking damage is expensive.

The AI engines (ChatGPT, Claude, Gemini, Perplexity) do not publicly document their ranking criteria the way Google does, but the leaked prompts and the observed behavior make clear that page experience is a confidence signal. A fast, clean, well-structured page is more likely to be cited than a slow, bloated one of equivalent content quality. The engines appear to treat page experience as a proxy for whether the content is worth summarizing in a generated answer.

Practically, improving Core Web Vitals usually improves AI visibility within 30–60 days of the engines re-crawling the pages. This is why we recommend addressing Core Web Vitals before investing heavily in GEO content work — the content investment compounds much better on a technically-healthy site.

Three tools cover the real measurement needs.

CLS fixes are usually editable in a page builder or theme settings and can be DIY. Image compression and WebP conversion are DIY with free tools. Removing unused scripts is usually DIY (the pain is the inventory, not the deletion).

INP fixes almost always require a developer unless your site is on a simple static builder. LCP fixes above the easy level (compress images, defer scripts) typically require developer time to configure CDN, pre-connect hints and HTTP/2 or HTTP/3. For sites on complex platforms (Shopify Plus, headless architectures, custom React), budget for a specialist engagement.