One often repeated piece of advice I see for improving web performance is to minimize the number of plugins you use.
Plugins, conventional wisdom says, increase the load on your server and database and slow down your site. Hell, it’s even in the WordPress Codex.
It’s also 100% wrong.
Plugins are not the problem #
Plugins are not bad for performance. Badly written plugins are.
The Harvard Business School’s Digital Initiatives website runs more than 20 plugins. So does this site. PAWS New England has almost 30 plugins.
They all start rendering content in around 1 second (slightly more on initial load and under that on subsequent page views).
Properly written plugins should have a minimal or nonexistent impact on your site’s performance.
What makes a plugin good (or bad)? #
- Run a new database query and recreate “The Loop” instead of modifying the original query (to, for example, display every single post on one page or only display events that haven’t happened yet for a custom post type like “Events”).
- Modify the query before it happens to avoid multiple database calls.
How do you know if a plugin is good or not? #
Honestly, it’s really hard to figure this out. I’ve found the process so frustrating that I end up writing a lot of my own instead of using third-party stuff.
Get the Spare Parts Newsletter
Every week, I send out a short email packed with web development resources and interesting stuff from around the web.