Public benchmarks
Reproducible benchmark protocols for modern WordPress and headless architectures. We publish the methodology before the measurements so anyone can re-run the test on their own infrastructure and verify the numbers.
Benchmark blog posts that lead with a single number rarely survive scrutiny. We publish the test corpus, the hardware, the network conditions, the deployment topology and the measurement script before the run. The number arrives later and replaces the "preliminary" disclaimer in the page. The methodology stays under a stable URL so a future reader can compare their own run to ours.
Each protocol below has its own page with the methodology embedded, plus a separate methodology index that lists every measurement decision in one place.
Same WooCommerce origin, same content, same Cloudflare Workers deployment. Astro 5 and Next.js 15 compared on TTI, LCP, INP, JavaScript bundle size and serverless cost per 100k pageviews.
What is measured: Time to interactive, LCP, INP, JS bundle size, cost per 100k pageviews
Identical Next.js application served from Cloudflare Workers and Vercel Edge. Time to first byte measured from synthetic clients in Warsaw, Frankfurt, Stockholm and Lisbon over a multi-day window.
What is measured: Time to first byte (median, p95) across EU PoPs
The same content, the same traffic profile, two architectures: a managed WordPress monolith and a headless WordPress + Cloudflare Workers stack. Total cost of ownership compared over a twelve-month operating window.
What is measured: Hosting cost per month, plus support and incident overhead, over twelve months
Send the architecture, the constraints and the question you want answered. We reply within one working day. Pricing is individual.
Contact us