I value your privacy
This website uses cookies to analyze traffic and improve user experience. You can accept all cookies, reject them, or customize your settings.
Find more information in our Privacy Policy
4 service/s
Facebook advertising tool used to track conversions and optimize advertising campaigns.
Purpose: Marketing and advertising
Google Analytics service that tracks and reports website traffic.
Purpose: Analytics
Microsoft Clarity service that helps understand how visitors interact with the website.
Purpose: Analytics
Cloudflare Analytics service that tracks website traffic and helps understand user behavior.
Purpose: Analytics
Technical insights, tutorials, and best practices for web development.
Everyone tunes the page cache. Meanwhile, with an undersized OPcache buffer, a WooCommerce store with six popular plugins recompiles ~1,200 PHP files on every single page view - doubling worker RAM, saturating CPU, and cutting throughput by 29%. I measured all of it, request by request. No dashboard shows you any of this.
I read 150 EmDash hot takes in a week. Almost none came with data. So I built two identical sites, ran 4,732 cold-isolated curl measurements over ~3.5 days, and asked an SEO crawl-stats expert to grade the result blind. Here is what actually happened.
Performance tests tell you how fast your store is. They don't tell you if your checkout actually works. Playwright does. Here's how to write your first test in under 10 minutes.
The default WooCommerce search is slow by design. I'm building a custom plugin using RediSearch and SHORTINIT to see how fast autocomplete can get. Early results from a cheap VPS inside.
Your checkout page is the only page that can't be cached. Every millisecond of latency runs at bare server speed — and it's costing you conversions. Here's what's actually slow and how to fix it.
A 1:1 database clone sends real emails, charges real cards, and corrupts your analytics. Here's how to build staging environments that are functionally identical but externally dead.
Code optimization on a terrible host is a dead-end loop. I share real benchmarks - same code, two servers - and why the order of operations fundamentally changes the outcome.
A case study on building the Shift64 website using Astro, exploring the technology choices, architecture decisions, and lessons learned.
Automate WordPress theme versioning end-to-end — from release pipeline setup to daily commits and PRs. One Claude Code skill sets up Semantic Release, three commands handle the rest.