How to Move Customers Through Your Site Faster: Page Speed Reports

A potential customer hears about one of your new releases on the way home. They pull out their phone, search for it and tap the link.

Then they wait.

The screen stays mostly blank. When something finally appears, it’s not the part they came for. They start to scroll, the page jumps, their thumb lands on the wrong thing, and they get taken to a new page.

That’s poor user experience at best. At worst, a page that moves just as someone taps is a dark pattern, because it gets people to press things they never meant to. Customers rarely tell you when this happens. They go back to the search results and find a site that isn’t going to waste their time. Google knows how frustrating this is, so it measures it as layout shift, and sites with a lot of it can be marked down in its evaluations.

Your team, or your agency, may have shown you a speed report with a lot of numbers and colours. This post covers how to find where customers are being held up: what the report measures, what good looks like, and what to ask.

Three numbers describe your customer’s wait

Google’s speed tools boil the experience down to three things that a customer can feel and that can be easily measured.

The example figures are mobile results on a real product page, before and after a round of speed work. The overall score went from 45 to 84 out of 100.

The middle row is a big improvement, 8.6 seconds down to 3.8, but it’s still short of the mark. When someone reports progress, ask where it sits against the good mark as well as how far it moved.

Google Judges the Phone Version of Your Site First

You may know that most of your customers shop on desktop. Google ranks your site using its mobile version regardless, so mobile is the number that counts for search. Desktop is often already good, and in this example it only moved from 93 to 99.

Google takes a month to notice a fix

A speed test loads one page, once, in a controlled environment. It shows the effect of a change straight away, but a single run can wobble by a few points.

Google’s own verdict, called Core Web Vitals, is based on 28 days of visits by real customers. It trails any change by about a month. A page can test well today and still show a failing verdict for weeks, and that’s normal. That matters if you have a peak sales period coming. How a page loads and behaves is one of the factors Google uses to score your site for search. Any search benefit builds over at least a month of sustained improvement, so do the work well before your peak sales period, not during it.

Five questions that find where customers are held up

1. Which page and which device did you test? A homepage on desktop tells you little about a product page on a phone.
2. What were the three numbers before and after? Not just the overall score, which can hide the one customers feel.
3. Where does each sit against Google’s good mark? A measure can improve a lot and still fall short of the mark. Ask for how far it moved and where it landed.
4. When will Google’s verdict reflect it? The honest answer is about 28 days after the change.
5. What business number will we watch alongside it? A better page is unlikely to bring many more customers by itself. Its main job is to stop people dropping out on the way through the site, like greasing the funnel, and that’s where conversion rate gains can come from. Watch conversion rate as well as traffic.

What a better score buys you, and when to stop

A page that loads quickly and has a stable layout is a better experience for the customer, and it’s one of the signals Google uses in ranking. It’s one of many, so it doesn’t guarantee more traffic or more sales on its own.

What it does is stop your site working against you. Improving it is usually a series of small, specific fixes, not a rebuild, which makes it one of the more affordable ways to improve how a site performs.

Returns also shrink as the score rises. Taking a page from poor to good is usually the most valuable work. Once it’s past Google’s good mark, each further gain costs more effort for less reward, so chasing a perfect score is rarely worth it.

If you’d like to know where customers are being held up on your own site, we’d be glad to take a look.