Are your addresses helping or hurting?
A URL is read before the click and is expensive to change after it.
The one thing you cannot tidy up later
Everything else on a page can be rewritten at no cost. Change an address and you owe a redirect, and you lose some of every link pointing at the old one. So the only cheap moment is before it is published.
One page at two addresses
A capital letter or a trailing slash makes a second address for the same page. Links get split between them, and neither version gets the benefit of the other's.
A date that ages the page
A year in the path makes the page look stale the moment the year turns, whatever is in it, and it means you can never properly update the article without moving it.
Nobody looks at the list
Addresses are decided one at a time, months apart, by whoever published the page. The pattern only appears when you see them together, which nobody does.
A list in, the patterns out
Get your addresses
Your sitemap has all of them: open yoursite.com/sitemap.xml and copy them out.
Paste them here
One per line, full addresses or bare paths. It runs in your browser.
See what runs through them
Grouped by fault rather than by page, so you fix a pattern once instead of a hundred times.
What it checks
Capitals, which split a page in two
Paths are case-sensitive on most servers, so /About and /about are two pages that get crawled, linked and shared as one. Anyone typing it lowercase gets a 404.
Dates in the path
Fine on news, where the date is the point. On anything you intend to keep updating it is a promise to move the page later or to leave it looking old.
Length and depth
Search truncates long addresses in results, and what is left is what someone reads before deciding to click. Depth is measured too, as the symptom it usually is.
The same page twice
Two addresses differing only in case or a trailing slash are one page in two places. It names the pair; which to keep is yours.
Words that carry nothing
A slug spending four of its words on the, of, for and your is longer and harder to read for no gain. The title carries the sentence.
Query strings and encoded characters
Sorting and tracking parameters make many addresses for one page, and a space arrives as %20, which breaks when the link is copied out of a message.
Straight talk about what it cannot see
It reads the addresses, not the pages behind them, and every check here is about the address itself.
It does not open anything
Whether an address answers, redirects or 404s needs a request, and this makes none. It reports what the address says, not what is at the end of it.
It cannot choose for you
When two addresses are the same page it names both. Which one keeps the links, and which one redirects, is a decision about your site.
A tidy address is not a good page
This is about the part a person reads before clicking and the part that costs money to change later. It says nothing about what they find when they arrive.
Your list never leaves your machine
The whole check runs in your browser. Nothing is uploaded, which is also why there is no limit on how many you paste.
Read your addresses once, together
It is the only part of a page that gets more expensive to fix every week it stays published.