To answer this question you have to know 2 things—what is considered “fast” and what is the average page speed of your store.
1. The threshold of patience
Back in the days people used to say “if your visitors can’t understand what your store is about in 60 seconds, they’ll leave”. Later this number got to 30s, then to 10s and then to 5s. Now it is even worse: “if your site doesn’t show something on the screen in about 3s, most of your visitors will leave”. No wonder why bouncerates of 75-80+ are very common in the e-commerce industry: most online stores are notoriously big, heavy, complex and as a result—slow to load.
“Three seconds”. That’s your yardstick.
Now, let’s see how you measure against it.
2. A speed test is not a speed test is not a speed test
Most of the online tools for speed testing load your site over a high-bandwidth connection on a very fast server. That’s why a slow site really has to put in some work to show up as slow as its customers are perceiving it. Few of the popular tools are trying (or have the option) to throttle the connection to approximate 3G speeds and lag. Still, these are just approximations.
Nothing beats actual load time measurements from the devices of your visitors. And that’s what Google Analytics’s Site Speed Reportcan provide for you.
When you go to Behavior > Site Speed you’ll see how much time it took your actual visitors to load your pages from start to finish. You can see the average page load time and check if a particular page is causing a slow down.
If some of your pages are showing 0 seconds, this doesn’t mean they are lightning fast. It just means Google Analytics hadn’t gathered any data for that page in that period.
Sidebar: the default sampling rate for speed timings is 1%—if you have 1000 views for the current period, you’ll get just 10 data points recorded.
To fix the lack of data in the report start increasing the date range till you hit a major redesign/theme change/app installation or removal milestone. With the increase in range of time you should get more data in the report.
Why is this better than PageSpeed Insights, GTMetrix, Pingdom Speed Test or WebPageTest.org?
Because this report contains data from your actual visitors being themselves in their context. Maybe your target audience uses high-powered iMacs with retina screens on a gigabit broadband, or maybe they are people on the go with sketchy signal coverage fluctuating between 2G/3G/4G speeds. You won’t get all of that detail in the report, but you’ll see how it impacts their page load timings.
And that’s when you measure against the “3s yardstick”. Is your average page load time good enough? Which pages need attention to clear the “three second mark”?
P.s. No, this article doesn’t mean the above mentioned tools are of no good use for you. Yes, their speed reports are better than nothing, but still not the “real thing”. Yet, where these tools excel is in automatic diagnostic of the issues that slow down your Shopify store. Knowing what slows your site down is the first step to a faster store, decrease in bouncerate and higher conversions.