Averages multiple successive Lighthouse tests to arrive at a more accurate PageSpeed score.
$ npm install --global @bkwld/lightkeeper
$ lightkeeper https://yourdomain.com
The results will be something like this:
Mobile Results
┌─────┬───────┬───────┬───────┬───────┬───────┬───────┬───────┐
│ │ Score │ FCP │ SI │ LCP │ TTI │ TBT │ CLS │
├─────┼───────┼───────┼───────┼───────┼───────┼───────┼───────┤
│ #1 │ 57 │ 3.8s │ 5s │ 6.2s │ 5.6s │ 328ms │ 0 │
├─────┼───────┼───────┼───────┼───────┼───────┼───────┼───────┤
│ #2 │ 66 │ 2.5s │ 3.5s │ 5.5s │ 5.2s │ 406ms │ 0.005 │
├─────┼───────┼───────┼───────┼───────┼───────┼───────┼───────┤
│ #3 │ 70 │ 2.5s │ 3.5s │ 5.4s │ 5s │ 307ms │ 0 │
├─────┼───────┼───────┼───────┼───────┼───────┼───────┼───────┤
│ AVG │ 64.3 │ 2.9s │ 4s │ 5.7s │ 5.3s │ 347ms │ 0.002 │
├─────┼───────┼───────┼───────┼───────┼───────┼───────┼───────┤
│ SD │ 5.4 │ 646ms │ 719ms │ 354ms │ 263ms │ 42ms │ 0.002 │
└─────┴───────┴───────┴───────┴───────┴───────┴───────┴───────┘
The summary rows (which can be exclusive returned with the -s option) contain the following rows:
AVG- The statistical meanSD- The standard deviation
From lightkeeper --help:
ARGUMENTS
<url> The URL to test
OPTIONS
-b, --block <urls> Comma seperated URLs to block, wildcards allowed
-d, --desktop Test only desktop
-m, --mobile Test only mobile
-s, --summary Only show summary rows
-t, --times <count> The number of tests to run
default: 10
For example:
lightkeeper https://www.bukwild.com- Runs 10 desktop and 10 mobile testslightkeeper https://www.bukwild.com -m -t=30 -b=googletagmanager- Runs 30 mobile tests while blocking Google Tag Managerlightkeeper https://www.bukwild.com -ms -t=100 && say "All done"- Runs 100 mobile tests and only show the summary at the end. And says "All done" aloud on a Mac.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent