by Daniel Sim
In this series we are looking at three methods of measuring your website performance: 1. Sampling your visitors 2. Analysis and recommendation tools 3. Time measurement. All are distinctly different ways of measuring the performance of your website, with advantages and disadvantages to each.
Of the three methods, time measurement is the simplest. It is starting a stopwatch when a page is requested and stopping it when it is ready for the user to use. That's the time in seconds the page takes to load.
This can be easily done in any desktop web browser.
There Are Different Definitions Of 'Loaded'
Due to the way modern browsers and websites work, there are different definitions of what 'loaded' means. Choose the wrong one and you may falsely believe your shop feels fast or spend a lot of time and money chasing an unattainable goal with no business benefit.
We look at in detail which definition of loaded to use in our article about How A Web Page Is Loaded. Our conclusion is that 'first meaningful paint' time is the best definition to use of loaded. It achieves all of the goals that speed optimization has.
Can Be Affected By Uncontrollable Factors
If your computer is running slowly, you have a bunch of websites open, others are streaming on your connection etc. this can affect the time you are measuring.
However this is less of a problem here than when sampling users speed. Usually the workflow is: measure the time, make a speed change, measure the time again. With a short gap between both measurements, the test conditions should be largely the same.
Manual And Small Data
Measuring your speed in this way is time-consuming: it's not something that's worthwhile to do every day.
It's also a small amount of data since we're testing the speed at one point in time infrequently. It won't catch occasional slowdowns of your shop or slowness related to geography/device/browser. For this time of monitoring use visitor speed sampling.
No Insights Into Why
With just a single number, the loading time in seconds, there's no analysis of what that number is made up of.
To get these insights use speed analysis and recommendations tools.
Is The Only Speed Metric That Directly Improves Conversion Rate
There have been many studies showing tangible, large improvements in conversion rate when a website improves its speed. Speed here is loading time in seconds, not PageSpeed score or visitor sampling or something else.
For this reason your loading time in seconds is a valuable metric if you are interested in speeding up your shop to improve your conversion rate.
When Not To Use
As we looked at, there are better alternatives for monitoring your website speed over longer time periods and getting ideas for speeding up your shop.
When It's Useful
Because it's an instant measurement of speed it is very useful to use for speed optimization work. It cannot be gamed and is not a vanity metric that will have no benefit on your business.
Your first meaningful paint time in seconds is a valuable metric to use as the basis for speed optimisation work. The other methods we've looked at in this series also have their role to play. Use each for its best purpose and you'll have a great holistic view of your speed.