Time to First Byte (TTFB) is the time it takes for a user's browser to receive the first byte of page content from the server. It includes the network latency and the server's processing time.

Code Examples#

Setting Cache Headers#

// ✅ Good: Use Cache-Control to reduce TTFB for static content
res.setHeader('Cache-Control', 'public, max-age=31536000, immutable');

Optimizing Database Queries (SQL)#

-- ❌ Bad: Selecting everything
SELECT * FROM users WHERE active = true;
 
-- ✅ Good: Selecting only what's needed and using indexes
SELECT id, name, email FROM users WHERE active = true;

Using a CDN (Next.js Example)#

// next.config.js
module.exports = {
  images: {
    domains: ['cdn.example.com'],
  },
}

Why It Matters#

• Foundational Metric: Everything else in the page load process waits for the first byte to arrive.
• User Perception: A high TTFB causes users to see a blank screen or loading spinner for longer, leading to frustration.
• SEO Impact: Search engines like Google use TTFB as part of their ranking algorithms, particularly for Core Web Vitals.
• Server Load: Optimizing TTFB often means reducing the server's resource usage, which can save on infrastructure costs.

Best Practices#

✅ Leverage CDN Caching: Cache as much as possible at the edge to avoid hitting your origin server. ✅ Optimize Server Logic: Profile your backend code to find and fix slow functions. ✅ Efficient Database Access: Ensure your database queries are optimized and indexed correctly. ✅ Static Site Generation (SSG): For content that doesn't change often, use SSG to serve static files instantly.

Tools & Validation#

• Lighthouse (opens in a new tab) (check for "server-response-time")
• PageSpeed Insights (opens in a new tab)
• Browser DevTools Network tab (look at the "TTFB" in the timing breakdown)
• WebPageTest (opens in a new tab)
• KeyCDN TTFB Checker (opens in a new tab)

Verification#

Automated Checks#

• Capture a fresh request in DevTools, WebPageTest, or your APM and confirm TTFB < 800ms for key pages.
• Inspect server logs or tracing spans for slow queries, upstream API calls, or expensive render steps before the first byte is sent.
• Re-test from multiple geographic regions if you rely on a CDN or edge network.

Manual Checks#

• Compare cache-hit and cache-miss responses so you know whether the bottleneck is origin work or caching strategy.