How to test your website speed in two minutes

A slow website loses visitors before they have read a word, and most owners have no idea whether theirs is slow. The phone in your pocket loads your site from cache, over your own wifi, in a second — which tells you nothing about the person finding you on mobile data in a car park.

Testing it properly takes about two minutes and needs nothing installed.

Step 1: Test the page people actually land on

Open PageSpeed Insights, paste a URL, and run it. Use the page visitors arrive on — often a service page rather than the homepage — and read the mobile score first. Mobile is where the problems are, because the network is slower and the processor is smaller.

Run it twice. A single measurement is a sample, not a verdict.

Step 2: Read three numbers, ignore the rest

Google's Core Web Vitals are three measurements, and Google publishes the threshold for each:

• Largest Contentful Paint (LCP) — how long until the biggest thing on screen appears. Good is within 2.5 seconds.
• Interaction to Next Paint (INP) — how long the page takes to respond when someone taps. Good is 200 milliseconds or less. INP became a stable Core Web Vital in 2024, replacing First Input Delay.
• Cumulative Layout Shift (CLS) — how much the page jumps around while loading. Good is 0.1 or less.

These are assessed at the 75th percentile of page loads, split by mobile and desktop. In plain terms: three out of four visits should be good, not the average visit.

The overall score out of 100 is a summary. The three numbers are the diagnosis.

Step 3: Match the number to the cause

Each of the three fails for a small set of reasons, and you can usually guess yours.

LCP is slow — almost always an oversized image. A 4 MB photograph straight off a camera is the single most common cause of a slow small-business site.

INP is slow — usually scripts: chat widgets, tracking, a booking embed, three of them at once.

CLS is bad — something loads late and pushes the content down. Images without dimensions, a cookie banner, a web font swapping in.

Step 4: Fix the image problem first

For most small sites, resizing images fixes most of the score.

• Export photographs at the size they are displayed, not the size the camera produced
• Use a modern format — WebP or AVIF — rather than a large JPEG
• Give every image its width and height so nothing shifts as it loads
• Remove any image that is decoration rather than information

Step 5: Re-test, and write the number down

Run the same page again and note the three values with the date. Speed drifts — a widget added in March shows up in April — and a number from six months ago is the only way to notice.

When the problem is not the images

Resizing pictures fixes most small-business sites. If yours is already lean and two of the three numbers are still red, the cause is one of these.

A slow Interaction to Next Paint is almost always third-party scripts. A chat widget, a booking embed, a tag manager and a review carousel each add work the browser must finish before it can respond to a tap. Remove them one at a time and re-measure; most sites are carrying at least one nobody has used in a year. What remains can often load only on the page that needs it rather than on every page.

A bad Cumulative Layout Shift is something arriving late. The usual four are images without width and height, an advert or embed with no reserved space, a cookie banner that pushes content down instead of overlaying it, and a web font that swaps in and re-flows the text. Each has the same fix: reserve the space before the thing arrives.

A slow Largest Contentful Paint with small images is usually the hosting or a render-blocking stylesheet. This is the point where the answer stops being something you can do from the content side, and the honest next step is to ask whoever runs the site.

How fast is fast enough?

Meeting all three thresholds is a good target and a reasonable stopping point. Chasing a perfect 100 is rarely worth it for a small business site: the last points cost disproportionate effort and no visitor perceives them.

What is worth doing is checking twice a year, and always after you add something that loads on every page.

Once the site is quick, the question becomes what visitors do next — see how to get more enquiries from your website and where your visitors come from.

Sources

• Web Vitals · web.dev
• Largest Contentful Paint (LCP) · web.dev
• Page experience in Google Search results · Google Search Central
• PageSpeed Insights · Google

Frequently asked

Does website speed affect my Google ranking?

Page experience, including Core Web Vitals, is one of many signals Google documents for search. It will not lift a page above more relevant results, but between two comparable pages the faster one has an advantage, and speed affects whether visitors stay regardless of ranking.

Do I need a perfect score of 100?

No. Meeting the three thresholds is a sensible stopping point for a small business site. The final points cost disproportionate effort and no visitor perceives the difference.