XML Sitemap Generator

What a sitemap is for

A sitemap is a list of the pages you consider worth indexing. It is a discovery aid, not a ranking factor: submitting one does not make a page rank, and leaving one out does not hide a page that is linked from elsewhere. What it does is make sure nothing is missed on a site where some pages are reachable only through a search box, a filter or a paginated archive that a crawler gives up on.

That means the list should contain canonical, indexable, 200-status URLs and nothing else. A redirect, a page that canonicals to a different address, a page carrying noindex, and a URL that 404s are all worse than absent, because every one of them teaches a crawler that this file is not reliable. If a page is not supposed to be in search results, leaving it out of the sitemap is the easy half of the job.

Which fields are worth filling in

The format allows four elements per entry and only two are worth your time. <loc> is the URL and is required. <lastmod> is the date the content genuinely changed, and Google has said it uses it, with a large caveat: it is trusted only while it is accurate. A site that stamps today onto every entry on every deploy teaches a crawler to ignore the field entirely, which is worse than never having set it.

<changefreq> and <priority> are ignored by Google outright and by most other engines in practice. This generator does not emit them, which is not an omission: adding fields nobody reads makes the file bigger and invites somebody to spend an afternoon tuning numbers that do nothing.

Size limits and how a crawler finds it

One sitemap file holds at most 50,000 URLs and 50 MB uncompressed. Past either limit you split the list across several files and add a sitemap index that points at them, which is an ordinary XML file listing the sitemaps rather than the pages. Most sites never come close, and a static site almost never does.

Once the file is live, tell crawlers about it in two places: a Sitemap: line in your robots.txt, which is the one that works for every engine, and your search console account if you have one. Both want an absolute URL, and both are happier if the sitemap sits at the root of the same host as the pages it lists.

One thing a sitemap will not do is get a page indexed that a search engine has looked at and decided against. Submitting a URL is a request to crawl, not a request to include, so a page that keeps failing to appear is usually telling you something about the page rather than about the file it is listed in. Check that it returns 200, that it canonicals to itself, and that it is not carrying a noindex it inherited from a template.