Redirect Latency

Redirect latency is the delay a redirect adds between a click and the moment the destination page starts loading. It covers the request to the redirecting URL, the server’s 3xx response, and any new DNS lookup, connection, and TLS handshake needed to reach the next URL. Every hop adds directly to the final page’s Time to First Byte.
What Causes Redirect Latency
A link redirect forces the browser to make one full network trip before it can request the real page. That trip has several parts:
- DNS lookup. The browser resolves the redirecting hostname, unless it is already cached.
- Connection and TLS handshake. A new hostname needs its own TCP connection and HTTPS negotiation. This is usually the largest cost.
- Server processing. The redirect server looks up the destination. Link shorteners may also log the click and apply device or geo rules.
- Distance. A redirect served from a far-away data center pays a long round trip at every step above.
The destination then needs its own DNS lookup and connection if it lives on a different hostname. Google’s web.dev guidance lists “redirect time” as the first component of TTFB and rates a TTFB of 0.8 seconds or less as good. Every millisecond spent on redirects comes out of that budget.
How Much Delay a Redirect Adds
The cost depends mostly on whether the hop needs a new connection. A 10-run curl test on 21 November 2026, from a client in India, timed two real 301 redirects on the same website against a direct request:
| Request | Median total time | Added by the redirect |
|---|---|---|
| Direct HTTPS request, no redirect | 163 ms | None |
301 from http:// to https:// on the same hostname |
213 ms | about 50 ms |
| 301 to a different hostname on another network | 568 ms | about 400 ms |
The same-host hop was cheap. The cross-host hop cost 8 times as much, because the browser needed a fresh DNS lookup, connection, and TLS handshake to a server on a different network. Chrome’s Lighthouse documentation gives the same range, warning that each extra trip “can delay the loading of the resource by hundreds of milliseconds.” Stacked hops add up, which is why a redirect chain is the most common cause of slow links.
How to Measure Redirect Latency
Measure the redirect separately from the page itself, so you know which one is slow.
- Use curl’s timers. The
time_redirectvariable reports the time spent on all redirect steps before the final request starts. - Check Chrome DevTools. Open the Network panel, enable “Preserve log”, and click the 3xx row. The Timing tab breaks the hop into DNS, connection, SSL, and waiting time.
- Run Lighthouse. Its “Avoid multiple page redirects” audit fails when a page has two or more redirects. Lighthouse 13 moved it into the “Document request latency” insight.
- Count the hops. A redirect checker lists every hop and status code between a link and its destination.
curl -s -o /dev/null -L -w "redirects: %{num_redirects}\nredirect time: %{time_redirect}s\ntotal: %{time_total}s\n" http://example.com/
Field data can under-report redirects. MDN notes that browser timing APIs return 0 for cross-origin redirects unless the redirecting server sends a Timing-Allow-Origin header.
How to Reduce Redirect Delay
- Link to the final URL. The fastest redirect is the one that never runs. Update old links in emails, ads, and internal pages.
- Keep every link to one hop. A short link should point straight at the tagged destination, not at another short link or an
http://URL. - Serve redirects close to users. A redirect answered at a CDN edge avoids a long round trip to a single origin server.
- Avoid client-side redirects. A meta refresh redirect or JavaScript redirect must download a page before it can send the visitor on, which is slower than a server-side 3xx.
- Use HSTS for HTTPS upgrades. Once a browser has seen the
Strict-Transport-Securityheader, it upgradeshttp://tohttps://itself, with no network hop.
Frequently Asked Questions
Does redirect latency affect SEO?
Indirectly. Google’s Gary Illyes confirmed in 2016 that 3xx redirects no longer lose PageRank, so a single redirect does not cost ranking signals. The delay still adds to TTFB and Largest Contentful Paint, which shape page experience. The URL shortener SEO breakdown covers where that matters.
Do link shorteners slow down links?
A shortener adds one hop, so it adds some delay. On a well-placed server that hop typically costs tens of milliseconds. It becomes noticeable when the short link’s server is far from the visitor, or when the short link points at another redirect instead of the final URL.
Is a 301 redirect faster than a 302?
On the first visit, no. Both cost the same network trip. The difference is caching: RFC 9110 makes a 301 heuristically cacheable, so a browser may skip the request on later visits, while a 302 is not. A cached 301 never reaches the server, so any click tracking on that hop is skipped too.
What is a good redirect speed?
No standard sets a threshold for redirects alone. The practical budget is Google’s 0.8-second TTFB target, which redirects share with DNS, connection, and server time. A same-host hop that adds about 50 ms leaves most of that budget intact. A hop costing several hundred milliseconds does not.
To see how many hops sit between your links and their destinations, run them through the free redirect checker at linkutm.