Speed is one of the few website problems with a clear, measurable fix. It affects conversions directly, it is a ranking factor, and on mobile connections the difference between a fast and slow site is the difference between a visit and a bounce.
The good news is that the causes are predictable. For most small business sites, one problem accounts for most of the delay.
- 1
Measure before you touch anything
Run pagespeed.web.dev on your three most important pages and record the mobile Core Web Vitals. Use Search Console’s Core Web Vitals report too — that is real visitor data rather than a lab simulation, and it is what Google actually uses.
- 2
Fix the images first
Images are the cause in the large majority of slow small business sites. Export at the size actually displayed rather than uploading a five-megapixel original, use WebP, and compress. A hero image should be well under 200KB. This single step often halves load time.
- 3
Lazy-load everything below the fold
Images and embeds outside the initial view should load only as the visitor scrolls. Most platforms support this with an attribute or a setting. Embedded videos and maps are particularly heavy and particularly worth deferring.
- 4
Audit what third-party scripts cost you
Chat widgets, tracking pixels, review embeds, font libraries, cookie banners. Each is a request to another server that blocks your page. List every one and justify it. Removing two unused tracking scripts is often worth more than any code optimisation.
- 5
Sort out fonts
Self-host rather than fetching from a third party, limit yourself to two families and only the weights actually used, and set font-display to swap so text is readable while the font loads. Each additional weight is another file every visitor downloads.
- 6
Check your hosting is not the bottleneck
Look at time to first byte in the report. Anything consistently over about 600ms points at the server rather than the page. Cheap shared hosting is a common cause and no amount of front-end optimisation will compensate for it.
- 7
Measure again and record the result
Re-run the same tests on the same pages and compare against your baseline. Real-visitor data in Search Console takes up to 28 days to reflect changes, so the lab test is your immediate feedback and the field data is the verdict.
What to aim for
Largest Contentful Paint under 2.5 seconds, Interaction to Next Paint under 200ms, Cumulative Layout Shift under 0.1 — measured on mobile. Hitting "good" on all three matters far more than any headline score out of 100.
On layout shift
Content jumping as the page loads is usually images without width and height attributes, or ads and banners injected above existing content. It is intensely annoying, easy to fix, and frequently ignored.
Frequently asked questions
It is a genuine but modest ranking factor. The larger effect is on conversion — visitors abandon slow pages, particularly on mobile data.
Would rather we did it?
These guides are here so you can do it yourself. If you would rather hand it over, that is what we do.
Start a conversation