Somebody tells you your website is slow. You open it on your office wifi and it loads fine. You tell them it is fine. They tell you to try it on a phone, on a bad signal, in a car park.
Both of you are right, which is the annoying part. Speed is not one number, and the version that matters is the one your customer sees.
What does "fast" actually mean for a website?
It means three different things, and Google has named them. These are called Core Web Vitals, which is a set of measurements of how a page feels to use. The three are explained on web.dev:
- Largest Contentful Paint (LCP): how long until the biggest thing on the screen, usually a main image or headline, has appeared. The recommended target is 2.5 seconds or less.
- Interaction to Next Paint (INP): how quickly the page reacts when someone taps or clicks. This one is about responsiveness, not loading. The recommended target is 200 milliseconds or less.
- Cumulative Layout Shift (CLS): how much the page jumps around while loading. The recommended target is 0.1 or less. You know this one: you go to tap a button and an advert shoves it away.
Google's guidance says to measure at the 75th percentile, which means that three out of four visits should hit the target. A page passes when all three metrics meet the targets at that level.
Does speed affect how I show up in Google?
Somewhat, and this is where I would stay calm. Google's description of page experience says its core ranking systems look to reward content that provides a good page experience. It also says site owners should not focus on only one or two aspects. Core Web Vitals are on its self-assessment list, alongside secure serving, mobile display and avoiding intrusive pop-ups.
My view: speed is a tiebreaker and a courtesy, not a magic lever. A slow site with a clear offer will still beat a fast site that says nothing. But a site that makes a customer wait is a site that loses the customer before the clear offer is read. That is an opinion, and the evidence is in your own enquiry numbers, not in mine.
How do I check my own site in five minutes?
Use a free tool and use a phone-sized test.
- Open PageSpeed Insights and paste your address.
- Choose the mobile tab. Not desktop. Most owners test on desktop and feel better than they should.
- Note the three measures above, and the list of suggestions underneath.
- Test your three most important pages, not only the home page. Usually a service page, a product or quote page and the contact page.
- Then test it for real. Take your phone, turn off wifi and load the page on mobile data. Time it with your thumb. A tool is a good guide. A real phone is a witness.
Write the numbers down with the date. In a month you will want to know whether anything you changed made a difference.
What should I fix first?
Fix the thing that is blocking the page, not the thing that is easiest. When I audited a new B2B site, the overall grade was B+, with A+ for on-page work and 87 for mobile PageSpeed. The thing that was slowing mobile loading was a render-blocking file, which is a file the browser must finish loading before it can show anything. That was a specific, findable cause. Fixing it was worth more than polishing a dozen small things. It is a small example from one site, so please do not read it as a promise for yours.
A sensible order, in my experience:
- Heavy images first. Oversized pictures are the most common cause of a slow main image.
- Files that block the page. The test tool will name them.
- Extra tools and add-ons. Every plugin adds code that each visitor downloads. I wrote about this in too many SEO plugins.
- Layout jumps. Give images and adverts a fixed space so the page stops shuffling.
When is it not worth fixing?
When the cost is bigger than the gain. On that same B2B site I deliberately left a larger rebuild alone, because it would have removed a single redirect hop and cost far more than it was worth. Chasing a perfect score is a hobby, not a business plan. If your numbers are in the recommended range and enquiries are healthy, spend your time elsewhere.
Also remember that speed cannot help a page Google has not stored. If a page is missing from search entirely, read why Google will not index your pages first.
What should I do this week?
Run the test on three pages, on mobile, and fix the single biggest item the tool names. Then test again. If you would like a hand reading the results, book a call. You can do the above without me.
Frequently asked questions
What is a good Core Web Vitals score?
Google's targets are an LCP of 2.5 seconds or less, an INP of 200 milliseconds or less and a CLS of 0.1 or less, all measured at the 75th percentile of visits. Check the current numbers on web.dev because thresholds can be updated.
Why is my site fast on my computer and slow on my phone?
In my experience phones often have slower connections and less processing power than an office computer, so the same page can feel slower. That is why I test on mobile first.
Do I need to hit a perfect score of 100?
No. A perfect score is not required and chasing it can cost more than it returns. Aim for the recommended targets and for pages that feel quick to a real customer.
Will a faster website guarantee more enquiries?
No, and nobody honest can promise that. A faster page removes one reason for people to leave. It does not create demand or fix an unclear offer.
