Advanced performance is mostly about planning
Geneva was a workshop, not a talk, so we went deeper than usual and wrote code together. No video was recorded, but the full slide deck is below.
The surprise for most people: the advanced part isn’t a clever trick. It is planning what to cache, for how long, and for whom.
Measure first
Before you change anything, run a test and write the numbers down. Otherwise you can’t tell whether your change helped. Pingdom, PageSpeed Insights and Lighthouse are enough to start.
Transients, the underrated cache
WordPress has a built-in way to store the result of an expensive operation for a set time: the Transients API. Check for a transient, and only if it is missing run the heavy query and save the result for, say, twelve hours. On a site with a persistent object cache, those transients live in memory and are very fast.
Plan your caching
Page cache, browser cache and object cache each solve a different problem. Start from the use case: how much traffic you expect, how often the content changes and what visitors expect to see. Then choose.
Scale up or scale out
A bigger server buys time. A load balancer with several smaller servers buys resilience. Neither replaces fixing slow queries, which is why profiling with Query Monitor comes before buying hardware.
Website speed matters more than ever. That was true in 2022, and it is even more true now.
