HTTP redirects, 301 vs 302, and how to count every click
Updated Nov 8, 2026 · 3 min read
A redirect is a response that says "what you want is somewhere else". The server answers with a status in the 3xx range and a Location header holding the new URL, and the browser makes a second request to that address on its own. The status code tells the browser two things. Is the move permanent, and can it reuse the answer next time without asking?
GET /x7Kp2Qa HTTP/1.1
Host: sho.rt
HTTP/1.1 302 Found
Location: https://example.com/spring-sale?utm_source=email
Cache-Control: private, no-storeThe five statuses that matter
| Status | Meaning | Cached by default? | Keeps the method? |
|---|---|---|---|
| 301 Moved Permanently | The resource has a new home for good | Yes | Not guaranteed (old clients turn POST into GET) |
| 302 Found | It is somewhere else for now | No | Not guaranteed |
| 303 See Other | Go fetch this other thing with GET | No | No, always becomes GET |
| 307 Temporary Redirect | Like 302, but strict | No | Yes |
| 308 Permanent Redirect | Like 301, but strict | Yes | Yes |
307 and 308 exist because 301 and 302 were historically allowed to change a POST into a GET. For plain links, which are always GET, 301 and 308 behave the same, and so do 302 and 307. 303 is the usual answer after a form submission, so that refreshing the result page does not submit the form again.
Why caching decides click counting
A permanent redirect is cacheable by default. Once a browser has seen a 301 for a short link, it may skip your server entirely on later clicks and go straight to the long URL, sometimes for as long as the cache lasts, with no expiry unless you set one. That is great for speed and terrible for analytics, because the first click is counted and the rest never reach you.
A 302 is not cached unless you add headers that allow it, so the browser asks your server on every click. That costs one extra round trip per click, and it is exactly what lets you count each one.
A cached 301 is hard to take back
Because browsers may keep a 301 indefinitely, you cannot reliably change where it points later. If a customer edits the destination of a short link, or you need to disable a link that now leads to malware, visitors with the old 301 cached keep going to the old place. With a 302 the next click picks up the change immediately. For anything whose destination might change, permanent redirects are a promise you may not be able to keep.
Search engines
Search engines treat a 301 or 308 as a signal to replace the old URL with the new one in their index and carry its ranking over. A 302 or 307 tells them the old URL is still the real address. That is the right choice when a site really moves; for a short link it rarely matters, since the short URL is not what you want ranked anyway.
Redirects that are not 3xx
<meta http-equiv="refresh">in an HTML page makes the browser download and parse the page first, so it is slower, and many link-preview bots never follow it.- A JavaScript redirect (
window.location = ...) is slower still, needs scripts enabled, and invisible to crawlers and previewers that do not run JS. - A 404 or 410: says there is nothing here. Use 410 Gone for a short link you deliberately disabled.
Counting clicks without slowing the redirect
- Look up the code (cache first, then database).
- Send the 302 straight away.
- Push a small click event (code, time, referrer, country) onto a queue instead of writing a counter row inline.
- Let a background worker batch events from the queue into your analytics store.
The visitor waits only for the lookup, and a burst of clicks becomes a burst of cheap queue appends rather than a burst of database writes on the same hot row.