Bundlers: ship-size vs build-speed tradeoff Build speed does not equal shipped speed. Measure both. — webpack: best tree-shaking + code-splitting maturity. Cold build of a 500-module app: ~25-40s. — e…
preload vs preconnect: 180ms vs 40ms Both are resource hints. They are not interchangeable. — preconnect: opens the socket (DNS + TCP + TLS handshake) early. Saves ~40-120ms per third-party origin. — …
CrUX vs Lighthouse: 28-day lag vs single sample Field data (CrUX = Chrome User Experience Report, real visitors) and lab data (Lighthouse, one synthetic run) measure different things. — Lighthouse: 1 …
TTFB 1.9s -> 160ms with one missing index A search results page filtered 2.4M rows on an unindexed status + created_at combo — full table scan every request. — EXPLAIN showed a sequential scan, 1,520m…
INP 540ms -> 130ms by virtualizing a table A reporting tool rendered all 4,000 rows to the DOM. Sorting re-laid-out 4,000 nodes on every header click. — Layout + paint per sort: 380ms. DOM nodes: 52,0…
If you follow us for Site speed / performance optimization, these belong in your list too: — @RootAccessDaily — Trench-level VPS tips for webmasters who SSH in and fix it… — @HostingHeresy — We debunk…
