SEO

TTFB

Also called Time to First Byte, server response time

The wait between a browser requesting a page and the first byte of the server's response arriving.

Quick facts: TTFB

Category
SEO
Also called
Time to First Byte, server response time
Level
Intermediate
Affects
First Contentful Paint, Largest Contentful Paint, crawl efficiency, hosting choice
Where to see it
Chrome DevTools Network panel, PageSpeed Insights, Search Console crawl stats
In this article4
  1. What TTFB measures
  2. Why TTFB matters
  3. Where TTFB goes wrong
  4. How to act on it

What TTFB measures

TTFB is the stopwatch between the moment a browser asks for a URL and the moment the first byte of the answer comes back. It is not one thing. Inside it sit the DNS lookup, the connection and encryption handshake, any redirects along the way, the time the server spends building the page, and the trip back over the network.

That mix matters, because two sites with the same number can have completely different problems. One may be running slow database queries. Another may have a fast server sitting a long way from its visitors. A third may be losing the time to a chain of redirects before the real page is even requested.

Why TTFB matters

Nothing can be drawn before the first byte arrives, so TTFB sets a floor under every other loading metric. First Contentful Paint and Largest Contentful Paint can only be as good as the wait that precedes them, and no amount of image or script work will push them below it.

It matters twice over when visitors sit far from the server. A business in Kathmandu hosted in the United States, or a Nepali company selling into Australia, pays a distance cost on every single request. Search engine crawlers feel it too: a slow-responding server is crawled more cautiously, which slows down how quickly new pages are found.

Where TTFB goes wrong

The usual cause on small business sites is oversubscribed shared hosting, where your response waits behind other people’s traffic. Next comes a page being rebuilt from scratch on every request because no full-page caching is in place, which is the normal state of an unconfigured content management system carrying a stack of plugins.

Measurement goes wrong just as often. Testing from your own office, in the same city as the server, produces a flattering figure that no overseas visitor will ever see. A logged-in test is misleading in the other direction, because it bypasses the cache and reports the slow path rather than the one real visitors get.

How to act on it

Work outwards from the server. Put full-page caching in front of it so most requests are answered from a stored copy, then examine what the uncached path is doing: slow queries, bloated plugins, an outdated language runtime. Once the origin itself is quick, distance is the remaining cost, and a content delivery network or a host in the right region answers it.

Then remove the avoidable hops — redirect chains, stray internal links that bounce through an old address, third-party lookups that happen before the page can respond. If the ceiling stays high after all of that, the platform is the constraint, and a hosting migration is the honest fix rather than another optimisation plugin.

Do and do not

Do

  • Measure from the countries your visitors actually browse from
  • Cache full pages so the server repeats less work
  • Cut redirects on the URLs people land on

Do not

  • Blame the theme before checking the server response
  • Buy the cheapest shared hosting for a commercial site
  • Treat a fast local test as proof for overseas visitors

Questions people ask about this

What counts as a good Time to First Byte?

There is no single figure that applies everywhere, because it depends on where your visitors are and how much the page has to build. The useful test is comparative: measure the same page from the countries your customers browse from, then compare a cached page against an uncached one. A large gap between them points at the server rather than the network.

Will a CDN fix a slow TTFB?

Only the part caused by distance. A content delivery network keeps copies of your pages close to visitors, which removes travel time and helps a great deal when your server is far away. If the delay is the server itself thinking — slow queries, heavy plugins, no cache — the network was never the problem and the figure will barely move.

Does TTFB affect SEO directly?

Not as a ranking factor in its own right. It matters because it delays everything the visitor sees, which drags down the loading metrics Google does assess, and because a server that responds slowly is crawled more conservatively. Fix it for the visitor and the search benefit follows; there is no separate credit for the number itself.

Related terms

Found this useful?

Share it, or ask an AI to summarise it

Back to the glossary

Knowing the term is the easy part

Applying it to your own site and budget is the work. Book a call and I will tell you what actually applies to you.