Page Speed Is a Business Metric, Not a Score
Performance work has an image problem. It's presented as a score out of a hundred, which makes it look like a video game, which makes it easy for an owner to deprioritize in favour of something that sounds like it makes money.
So let's translate.
What the score is actually measuring
A Lighthouse score is a weighted average of lab measurements taken on a simulated device on a simulated connection, on your machine, at one moment. It's a useful diagnostic and a terrible target.
Two things make it misleading as a goal. It's synthetic — your real visitors are on different devices and networks. And it's non-linear — going from 40 to 70 is a genuinely different experience; going from 90 to 100 is often invisible to every human being.
What to measure instead
Field data. What actual visitors experienced, from real-user monitoring or the Chrome UX Report. This is the number that reflects your business and the one search engines use.
Time to first meaningful content on a mid-range phone on cellular. Not the score — the seconds. Everyone in the room understands seconds.
The bounce curve by load time. If your analytics can segment by speed, this is the most persuasive chart in the building. It reliably shows that visitors who waited longer left more.
The translation that lands
Take real numbers off the site. Suppose 8,000 organic sessions a month, a 2.4% enquiry rate, and an average job worth $2,400.
Now: your mobile pages take 5.8 seconds to become useful, and mobile is 61% of your traffic and converting at less than half the desktop rate.
The question stops being "should we spend on performance" and becomes "what is the mobile gap costing per month." That's a question a business owner can answer, and the answer is usually much larger than the cost of the work.
Where the time actually goes
For most business sites, in this order: oversized images, third-party scripts, render-blocking CSS and fonts, and unnecessary platform overhead. That's the whole list, and the first two are usually more than half the problem.
Which is worth saying out loud, because it means the work is finite. Speed optimization is not an endless money pit — it's a handful of specific fixes with a clear end point.
What "good enough" looks like
For a small business website: something meaningful on screen within about two seconds on a mid-range phone on 4G, interaction that responds immediately, and no layout that jumps around while loading.
Past that, keep the discipline rather than chasing the number. Set a page weight budget, check field data monthly, and don't let a new marketing tag quietly add 400 KB in November.
Nobody has ever chosen a supplier because their Lighthouse score was 100. Plenty have left because a page took six seconds.