GuardCue

Am I losing people because my site is slow?

A slow page does not lose you customers who complain. It loses the ones who leave before it finishes, and they never tell you they were there.

Free. No sign-up. We check the one page you give us, and only what any visitor can already see.

What this check can and cannot see

Speed measured from a single run can swing, so where we can we lean on how your site behaves for real visitors rather than one lab test, and we say when a number is a single measurement.

What we will never doChange anything in your app, or keep your customers’ data. A finding holds a count, the column names, and one sample blanked past its first two characters.
How it happens

Nobody chooses this. It is what is left out.

Speed on your own machine is not the number that matters. You are on a fast connection, the site is already cached, and you know where you are going. Your visitor is on a phone, on mobile data, deciding in about two seconds whether to wait.

The usual causes are dull and easy to fix. A photo uploaded at full camera size and shown two inches wide, so the phone downloads several megabytes to display a thumbnail. Code that runs before anything appears. A font that holds the text back until it arrives.

The honest part is the hardest. One speed test tells you almost nothing: on TruckTuck we measured the same unchanged site scoring between 53 and 98. So where we can we use how your site behaved for real visitors over time, and where we cannot, we say the number came from a single run rather than pretending it is the truth.

What it looks like

Two results, printed as you would get them.

Neither of these is your app. They are the words our own checker prints, on test apps we built with the faults we see in the wild.

on a phone, on mobile data
On a mid-range phone, the main part of the page takes about 6.4 seconds to appear.

About half of visitors leave a page that takes longer than three seconds, and they do not come back to try again.
ExampleCosts you customers

Your page takes over six seconds to show anything useful.

What it means for youMost of the people who click your link never see your app. They are not annoyed customers you could win back; they are strangers who decided you were broken and went elsewhere, and nothing in your numbers records them.
what we didloaded your page as a mid-range phone on mobile data
what we found6.4 seconds before the main content appeared
on the home page
3 images on your site are bigger than a phone should have to download, the largest being hero-photo.png at 4.2 MB.

That one file is often larger than the rest of the page put together, and every visitor waits for it.
ExampleWorth a spare hour

One photo is larger than the whole of the rest of the page.

What it means for youThis is the single most common cause and the easiest to fix: the picture is shown at a few hundred pixels wide and stored at several thousand. Shrinking it costs an hour and usually takes seconds off the number above.
what we didmeasured every image against the size it is displayed at
what we found3 too large, the worst 4.2 MB
If the answer is yes

It says you are slow. Now what?

Start with the images, because they are usually most of it and the fix needs no code: export at the size they are shown and save as a modern format. That alone often moves a page from six seconds to under three.

After that it is what runs before the page appears, which does need the tool that built your app. The free check names the exact files and their sizes; paying gets you the words to paste and a weekly re-check, because speed is the thing that creeps back one change at a time.

What a plan adds →

Before you paste your address

The questions people ask.

My site feels fast to me. Is the test wrong?
Probably neither. You are on a laptop with the site already cached, and we measure a mid-range phone on mobile data arriving for the first time. Both numbers are real; only one of them is what your next customer experiences.
Why do I get a different score every time I test?
Because a single run genuinely swings that much. We measured the same unchanged site at 53 one minute and 98 the next. That is why we prefer how your site behaved for real visitors over time, and why we tell you when a number came from one run rather than dressing it up as fact.
Does this actually cost me money, or is it just a score?
It costs you twice. People leave before the page loads, and Google ranks slow pages lower, so you get fewer visitors and lose more of the ones you get. It is the rare problem where the technical number and the business number are the same number.
Am I losing people because my site is slow? · GuardCue