do i need an xml sitemap? what the list of urls does, and when it earns its keep

seo · Mar 18, 2026 · 5 min read

an xml sitemap is a file — usually mybusiness.com/sitemap.xml — that lists the urls of your site in machine-readable form, each with its last-updated date. it is not a ranking trick and not a map for visitors. it is the index page of a book handed to the librarian: here is every page i have, and here is which one changed most recently.

what it actually does

without a sitemap, google discovers pages by crawling links — your menu, your internal links, links from other sites. that works, eventually. the sitemap shortens the path for three cases where crawling alone is slow or blind:

  • a new site with no backlinks yet — nothing points at you, so discovery by crawling is slow. the sitemap is the announcement. this is exactly the launch phase of how do i get my website on google.

  • pages deep in the structure — the blog post reachable only from an archive page three clicks down gets announced directly.

  • large or fast-changing sites — the lastmod dates tell the crawler which of the thousand pages to re-check first.

    do you need one? the honest answer

  • five-page builder site: you will survive without it. google crawls five pages from the menu. having one costs nothing — every builder ships it — so the real answer is "you do not need to think about it".

  • small site with a blog: yes, mildly. the posts are the deep pages, and the sitemap plus the lastmod dates get new posts seen faster.

  • any site that publishes regularly: yes. this is the case where it earns its keep daily.

this site generates its sitemap at build time, with a generator that fails the build if any post in the code is missing from it — the only mechanism i found that prevents orphan pages, a failure i have lived through before. the sitemap is where "nothing is forgotten" becomes a machine-checked property instead of a hope.

how to hand it over

one step, after the file exists: submit the url in google search console — sitemaps section, paste, done. from then on the report shows how many of the listed urls google has indexed, which quietly turns the sitemap into a diagnostic: a sitemap of fifty urls and an index of twenty is a finding, not a statistic — and the finding has a name (thin pages, canonical mistakes, or plain newness), explained in what is duplicate content.

pair it with the one-line Sitemap: reference in what is robots.txt so crawlers that never touched search console still find the list.

FAQ: the questions i actually get

does a sitemap improve my ranking?

no — it improves discovery speed, not position. faster indexing of new content has a ranking effect over time, because the content can start earning sooner. that is the whole mechanism, and it is why the honest expectations in how long does seo take never change with tools.

html sitemap too — the page for humans?

obsolete for most sites. the menu and search do that job. the xml file is the one that matters now.

how often does it need updating?

it should be generated, not hand-written — most platforms regenerate it on publish. if yours is a static file edited by hand, that is the real problem to fix.