Most React developers I know are confused about React Server Components. I was too. Do they actually improve performance — or is it just hype? So Nadia Makarevich did what most of us rarely do: She measured it. Every rendering strategy. Real Chrome DevTools profiles. Real numbers. Here’s what the data shows 👇 🐢 CSR — ~4s before users see anything. JS downloads, boots React, then fetches data. ⚡ SSR — first paint drops to ~1.78s. 🤯 RSC alone — surprisingly little difference in mixed apps. 🌊 RSC + Suspense + Streaming — this is the real unlock. The server streams HTML as data resolves. The performance profile changes completely. The takeaway: • RSC without Suspense = wasted effort • Suspense without server-first data = half the benefit • All three together = the performance profile you actually wanted If you're building with Next.js App Router, your Suspense boundaries matter more than you think. Talk + article in the comments 👇 #React #ReactServerComponents #WebPerformance #NextJS #Frontend #ReactAdvanced
This matches what I’ve been seeing as well. RSC alone often doesn’t change much in mixed apps, but once you combine it with Suspense and streaming the UX difference becomes much more noticeable. The tricky part seems to be designing the right Suspense boundaries and data flow so you actually benefit from server-first rendering. Curious how teams are structuring those boundaries in larger Next.js apps.
📺 Watch the talk (22 min):https://www.youtube.com/watch?v=_RQAMXlWSd0 📖 Full deep-dive:https://www.developerway.com/posts/react-server-components-performance