Article WordPress

Redis Object Cache for WordPress and WooCommerce: what it speeds up and what it does not

Redis Object Cache can reduce database round-trips, but it does not replace page cache, OPcache or query optimization. This guide separates the cache layers and shows what to measure.

Hamid Aarab Senior WordPress and WooCommerce Backend Developer
Published
Reading time
2 min read

Hamid Aarab November 22, 2026 2 min read

Terminal-style Redis Object Cache cover for WordPress and WooCommerce
Contents (7)

When a WordPress site is slow, “install Redis” is often suggested too early. Redis can help, but only when you understand the layer it serves. Object cache keeps application objects and query results available across requests; it does not cache complete HTML pages or compiled PHP bytecode.

The default WordPress object cache is not persistent

WordPress has an in-memory object cache for the lifetime of a request. Persistence between requests requires an object-cache.php drop-in backed by a service such as Redis or Memcached.

What can improve

Sites with repeated queries, large option sets, taxonomy-heavy workloads or busy WooCommerce traffic can reduce database round-trips and stabilize TTFB with a persistent object cache. The real benefit depends on hit rate and invalidation patterns, not on whether the Redis process is merely running.

What it does not replace

  • Images and CSS/JS → browser/CDN caching.
  • Fully rendered HTML → page cache.
  • Compiled PHP bytecode → OPcache.
  • Bad SQL or missing indexes → fix the query/database.
  • Slow external APIs → use integration-specific caching and timeouts.

WP_CACHE is not Redis

Defining WP_CACHE by itself does not enable Redis. Persistent object caching uses a dedicated drop-in. Also remember that transients may live in an external object cache when one is active, or in the database when it is not.

How to verify the layer

wp cache type
wp cache supports flush_group
wp cache get some_key some_group

Use cache metrics too: hit/miss ratio, memory usage and latency. A low hit rate is often a signal to inspect invalidation and cache groups before simply allocating more memory.

My WooCommerce rule

Treat Redis as one performance layer, not a turbo button. Measure the bottleneck first, then cache the thing that is actually expensive.

Sources

$ ./request_review

Facing the same problem in your own store?

Senior WordPress and WooCommerce backend engineering for complex production problems.

Request a Technical Assessment

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *