Time to first byte

Time to first byte, TTFB, is the interval from the start of the request to the arrival of the first byte of the response. Good is under 0.8 seconds and poor is over 1.8 seconds, as web.dev publishes them. It is not a Core Web Vital, but every other timing inherits it.

By , founder of Porteur · Updated 15 September 2026 · Markdown

What the number contains

  • Any redirects, each with its own round trip.
  • The DNS lookup, if the host is not already resolved.
  • The TCP connection and the TLS handshake.
  • The server’s own work: routing, database calls, rendering.
  • The first byte travelling back to the browser.

The thresholds and where to read them

TTFBReading
Under 0.8 sGood
0.8 s to 1.8 sNeeds improvement
Over 1.8 sPoor

PageSpeed Insights shows it in the field section when your site has enough traffic, and in the diagnostics of the lab run as the server response time. A browser’s network panel shows it per request under Waiting.

What improves it

  1. Cache or pre-render the HTML

    A page built at request time on every visit pays for that work every visit. Static generation or a cached response removes it.

  2. Put a CDN in front

    Most of a slow first byte for a distant visitor is distance. A CDN answers from near them.

  3. Remove the redirects

    Point the first URL straight at the final one. Each hop removed is a round trip saved.

  4. Turn on HTTP/2 or HTTP/3

    Both remove connection overhead; they are set at the server or the CDN and cost nothing to the page.

  5. Make the server work less

    The database call in the layout that runs on every page is the usual culprit on a small site.

Questions

Check my site, free

The free check measures how your home page answers and paints on a phone, and names the first thing to fix, in about thirty seconds.

  • Free check, no card
  • Read-only, your own accounts
  • Readable by your agent

Read next