Online Tool Store
· 5 min read
Best 3 Core Web Vitals References Compared
Heshan Fernando
Co-founder & COO
Someone forwards a performance report and it is full of three-letter acronyms with numbers attached. LCP 3.1s. INP 240ms. CLS 0.18. You need to know, quickly, which of those are bad, how bad, and what each one is actually measuring before you can say anything useful about them.
The information is not hard to find. It is hard to find briefly. The authoritative sources are excellent and long, written to teach you the metric properly, and what you needed was one line per acronym and the threshold that separates good from not. Meanwhile the metrics themselves have changed — INP replaced FID — so half the material you land on via search is describing a version of Core Web Vitals that no longer applies.
How to judge a Core Web Vitals reference
Is it current? INP became a Core Web Vital and FID was retired. Any page still centring FID is out of date, and the dates are not always obvious.
Does it give thresholds, not just definitions? Knowing what CLS measures does not tell you whether 0.18 is a problem.
Does it say what percentile the threshold applies at? The numbers are assessed at the 75th percentile of page loads, which changes how you read a single measurement entirely.
Is it a reference or a measurement tool? Looking up a definition and measuring your own site are different tasks that get conflated constantly.
The comparison
| Tool | Best for | Free tier | Watch out |
|---|---|---|---|
| web.dev Web Vitals | The authoritative definitions and thresholds | Free, no account | Thorough enough that a quick lookup means reading past a lot of context |
| PageSpeed Insights | Measuring a real URL against the thresholds | Free, no account | It measures rather than explains — the definitions are links away |
| MDN Web Performance | Understanding the underlying browser APIs and timings | Free, no account | Broader web performance scope, so Core Web Vitals is one topic among many |
Facts checked August 2026; plans can change.
web.dev Web Vitals
This is the source of truth. It defines Web Vitals as Google’s initiative to give unified guidance on quality signals, and it states the three Core Web Vitals and their targets plainly: LCP within 2.5 seconds of when the page starts loading, INP of 200 milliseconds or less, and CLS of 0.1 or less. It also makes the point most summaries drop — that these are assessed at the 75th percentile of page loads, segmented across mobile and desktop.
Its thoroughness is the trade-off. Each metric has its own deep article behind it, and if you arrived wanting one number you will do some scrolling to reach it.
PageSpeed Insights
PageSpeed Insights answers a different question: how is your page doing. Paste a URL and it reports field data where available alongside lab measurements, colour-coded against the same thresholds. For turning an abstract metric into an actionable number for a specific page, nothing else is as direct, and it is free with no account.
It is a measurement tool, not a glossary. The definitions are one click away rather than on screen, and its opportunities section tells you what to fix without necessarily explaining what the metric means.
MDN Web Performance
MDN covers the layer underneath. If you want to know how LCP is actually observed in a browser, which Performance APIs report it, or what a layout shift is at the rendering level, MDN explains the mechanism rather than the score. That depth is what you need when you move from measuring a problem to fixing it.
Core Web Vitals is one part of a much larger performance section there, so it is a reference for understanding rather than a quick lookup for a threshold.
Core Web Vitals Glossary
Ours is deliberately the shallow one. It gives a plain-language definition for each Core Web Vitals metric and related signal, with the threshold Google considers good, in a format built for a ten-second lookup during a meeting. It runs entirely in your browser.
The limitation is that it is a glossary and nothing more. It does not measure your site, does not tell you why your LCP is slow, and does not replace reading the full metric documentation when you are about to make an engineering decision. It gets you oriented; the tools above take it from there.
Which one to pick
- If you are making a decision that depends on the precise definition, read web.dev.
- If you need to know how a specific page scores, use PageSpeed Insights.
- If you are implementing measurement or debugging a layout shift, MDN explains the machinery.
- If you just need to remember what INP is and whether 240ms is bad, use ours.
How to do it with Core Web Vitals Glossary
- Open the Core Web Vitals Glossary.
- Find the metric from the report you are reading.
- Compare your number against the stated “good” threshold.
- Measure the actual page afterwards if the number looks wrong. More web tools are in the tools directory.
You might also need
- Page Speed Checklist — for the practical steps once you know which metric is failing.
- Responsive Breakpoint Tester — since layout shifts often show up on one viewport and not others.
Frequently asked questions
Is there a free Core Web Vitals reference that doesn’t need an account?
Yes — web.dev, PageSpeed Insights, and MDN are all free without registration, and our glossary has no accounts because the site has no signup. Nothing in this category should cost you anything.
What are the current Core Web Vitals thresholds?
Per web.dev’s Web Vitals article: LCP should occur within 2.5 seconds of the page starting to load, INP should be 200 milliseconds or less, and CLS should be 0.1 or less — each assessed at the 75th percentile of page loads.
What happened to FID?
Interaction to Next Paint replaced First Input Delay as the responsiveness Core Web Vital. INP measures the latency of interactions across the whole page visit rather than just the first one, which reflects real responsiveness more honestly.
Final thought
Keep a glossary for reading reports and a measurement tool for fixing pages, and do not confuse the two. Most Core Web Vitals confusion is someone looking up a definition when they needed a measurement, or the reverse.