Somewhere in a report or an agency pitch you have probably met the phrase Core Web Vitals, flanked by acronyms and traffic-light charts. Behind the jargon sits something refreshingly simple: three measurements of whether your website feels good to use. Google collects them from real Chrome visitors, uses them as a ranking input, and publishes the thresholds. Here is the plain-English version, 2026 edition.
The three vitals, translated
LCP, Largest Contentful Paint: how quickly the main thing shows up. The clock starts when a visitor requests your page and stops when the biggest element, usually the hero image or headline, is on screen. Good is 2.5 seconds or less. A slow LCP is the classic "staring at a blank page" experience, and it is where most small business sites fail.
INP, Interaction to Next Paint: how quickly the page responds when touched. INP replaced the older FID metric and measures the delay between a visitor tapping or clicking anything and the page visibly responding. Good is 200 milliseconds or less. A poor INP is the button that seems to do nothing, so the visitor taps it three more times.
CLS, Cumulative Layout Shift: whether the page holds still. CLS scores how much content jumps around while loading. Good is 0.1 or less. It is the metric behind the universally hated experience of reaching for a button just as an ad or banner shoves it down the page.
Do they actually affect rankings?
Yes, as one signal among many, and with a sense of proportion: brilliant vitals will not rescue thin content, and genuinely useful content can rank despite mediocre scores. The larger commercial effect is indirect. Slow, janky pages lose visitors before the content is ever read, and those abandonment signals feed back into everything, a chain we traced in the real cost of a slow website. Treat vitals as a floor: passing does not win you the race, but failing means running with a limp.
How to check your own site, free
PageSpeed Insights gives you both lab tests and, where your site has enough traffic, real-visitor data, which is the data that counts. Search Console's Core Web Vitals report shows which groups of pages pass and fail across the whole site. Our AI Site Audit rolls a performance check in alongside SEO and mobile signals if you want one report instead of three tabs. Whichever you use, judge the mobile numbers: that is where most visitors are and where scores are hardest to earn.
What actually fixes failing scores
- For LCP: compress and properly size images, serve them in modern formats, preload the hero image, use decent hosting or a CDN, and cut render-blocking scripts. On WordPress, theme and plugin bloat is usually the root cause; the architecture arguments in static versus WordPress are largely LCP arguments.
- For INP: less JavaScript, later. Third-party widgets, chat bubbles, tag managers stuffed with old pixels: each one competes for the browser's attention exactly when the visitor is trying to interact. Audit what is loading and remove what no longer earns its place.
- For CLS: give images and embeds fixed dimensions so the browser reserves their space, preload fonts to stop text reflowing, and never inject banners above content after load.
The business owner's takeaway
You do not need to memorise the acronyms. You need three habits: check the real-visitor numbers a few times a year, treat mobile as the benchmark, and when scores fail, fix causes rather than chasing points, because the causes are what your customers are actually feeling. Performance is built into every site we ship, not bolted on afterwards, for one simple reason: it is far cheaper to build fast than to make slow things fast later. If your scores are red and you want them green, we do that.
