Benchmarking Website Performance: Difference between revisions
David Smith (talk | contribs) No edit summary |
David Smith (talk | contribs) No edit summary |
||
| Line 8: | Line 8: | ||
The following KPIs can be tested for: | The following KPIs can be tested for: | ||
===== '''Average Response Time''' ===== | ===== '''Average Response Time and Average / Best Loading Time''' ===== | ||
We can benchmark the Time to First Byte (TTFB) and the Time to Last Byte (TTLB) as these two metrics will give us a great indication as to how long the average user must wait for the website to a) start loading and b) to finish loading. If the average response time is slow, most of your website's visitors will leave before the page has even finished loading. | We can benchmark the Time to First Byte (TTFB) and the Time to Last Byte (TTLB) as these two metrics will give us a great indication as to how long the average user must wait for the website to a) start loading and b) to finish loading. If the average response time is slow, most of your website's visitors will leave before the page has even finished loading. | ||
Using Pingdom, we can test your website's average and best loading times to see what kind of experience your visitors are receiving when they visit. For example: | |||
http://tools.pingdom.com/fpt/#!/cPphSi/http://www.parasol-island.com | |||
===== '''Page Grade''' ===== | ===== '''Page Grade''' ===== | ||
| Line 21: | Line 24: | ||
We can use a piece of software called ApacheBench (https://en.wikipedia.org/wiki/ApacheBench) to test how efficiently your website's server can handle a user's request. An example test would be: | We can use a piece of software called ApacheBench (https://en.wikipedia.org/wiki/ApacheBench) to test how efficiently your website's server can handle a user's request. An example test would be: | ||
ab -n 100 -c 10 "http://www.parasol-island.com/" | ab -n 100 -c 10 "http://www.parasol-island.com/" | ||
Latest revision as of 15:07, 21 March 2016
What is a benchmark?
[edit]A benchmark is a "standard or point of reference against which things can be compared".
Why benchmark test a website's performance?
[edit]To see how a website is performing and to see where performance can be enhanced.
Which Key Performance Indicators (KPIs) can we test for?
[edit]The following KPIs can be tested for:
Average Response Time and Average / Best Loading Time
[edit]We can benchmark the Time to First Byte (TTFB) and the Time to Last Byte (TTLB) as these two metrics will give us a great indication as to how long the average user must wait for the website to a) start loading and b) to finish loading. If the average response time is slow, most of your website's visitors will leave before the page has even finished loading.
Using Pingdom, we can test your website's average and best loading times to see what kind of experience your visitors are receiving when they visit. For example:
http://tools.pingdom.com/fpt/#!/cPphSi/http://www.parasol-island.com
Page Grade
[edit]There are two services that we can use to quickly "grade" your website's performance and these are linked below:
Google PageSpeed Insights (https://developers.google.com/speed/pagespeed/insights/) GTMetrix (https://gtmetrix.com)
These two services will give you a grade and if the grade isn't good, we will be given explanations on what needs to be improved.
Request Handling Speed
[edit]We can use a piece of software called ApacheBench (https://en.wikipedia.org/wiki/ApacheBench) to test how efficiently your website's server can handle a user's request. An example test would be:
ab -n 100 -c 10 "http://www.parasol-island.com/"