We put Lighthouse scores on our case studies, so this might sound strange coming from us. A 100 is not the point. A 100 has never been the point.
Lighthouse scores have become a sales prop. Agencies screenshot a green circle and call it proof. Sometimes it is. Often it is a desktop score, on a stripped test page, from the one run out of five that came back highest. You deserve to know the difference.
What Lighthouse actually measures
Lighthouse is a lab test. A program loads your page on a simulated device with a throttled connection and grades the result from 0 to 100. Same page, same tool, different runs can produce different scores. That is normal. It is a controlled experiment, not a surveillance camera.
Two things are true at once:
- A low score almost always points at real problems. A 40 on mobile means real users are having a bad time.
- A perfect score does not by itself mean the site makes money. It means the lab run was clean.
The score is a composite of several metrics with different weights. Three of them are worth your attention, because Google measures them on real visitors too.
The three numbers that map to money
These are the Core Web Vitals. Google collects them from actual Chrome users on your actual site and uses them as one signal among many in ranking. Not the whole game, and anyone who says otherwise is selling something. But real.
| Metric | What it measures | What your customer feels |
|---|---|---|
| LCP | Time until the main content is visible | "Is this thing loading or not?" |
| INP | How fast the page reacts to a tap or click | "Did my tap even register?" |
| CLS | How much the layout jumps around | "The button moved and I hit the wrong thing" |
LCP protects the first impression and your ad spend. INP protects the moment someone tries to use your quote form. CLS protects them from rage-tapping the wrong link. These map to behavior. The composite score is just a summary of a lab run.
Where score-chasing goes wrong
Lab is not the field. The score comes from one simulated load. Field data comes from thousands of real visitors on real phones and real networks. A site can score high in the lab and still feel slow to actual users. When lab and field disagree, the field is telling the truth. Google publishes field data through the Chrome UX Report, and PageSpeed Insights shows it right at the top of the results once a site has enough traffic. Free to check. Hard to fake.
The tricks. There are ways to inflate a score without making anything better for a human. Deferring scripts so the test finishes before the weight arrives. Lazy-loading content the test never scrolls to. Stripping features that made the business money. The circle turns green. Nothing improved.
The screenshot game. Desktop scores run far higher than mobile. If someone shows you a 100 without saying which device, which page, and which data, they are showing you the most flattering number they could generate.
The honest version
Here is how to read performance claims from anyone, including us.
- Ask for the mobile score, not desktop.
- Ask whether it is lab data or field data from real users.
- Ask for the URL and run the test yourself. It is free.
- Ask what LCP, INP, and CLS look like, in plain terms.
We publish our scores because they are checkable, not because a round number is the goal. The goal is a site that loads before your customer's patience runs out, reacts when they tap, and holds still while they read. Do that honestly and the green circle tends to show up on its own. Chase the circle instead, and you can end up with a trophy for a site that still loses customers.
If a pitch leads with a 100 and cannot answer which device and whose data, keep your hand on your wallet.