Fix Your Sitemap in 20 Minutes: What Google Actually Reads (and What It Ignores)

Somewhere in your website there's a file called sitemap.xml. You've never opened it. Your web designer hasn't opened it since launch day.

Google opens it all the time.

And on Oct 1, 2026, Google's own Search Central team (Martin Splitt and John Mueller) sat down and explained exactly what they read in it. Spoiler: it's way less than you think, and most sites still get it wrong.

Quick words (one line each):

• Sitemap: a file that lists the pages you want search engines to find.

• lastmod: the "last changed" date next to each page in the sitemap.

• Canonical URL: the one clean, main address of a page (no tracking tags tacked on).

• RSS feed: a file that lists your newest posts, the thing podcast apps and news readers use.

• Search Console: Google's free dashboard that shows how your site does in Google.

Step 1: Make sure you have one (yes, even small sites)

Mueller's advice: keep a sitemap on, even for a small site. There's no downside, and most site builders make one for you automatically (0:06:20-0:07:20).

Open yoursite.com/sitemap.xml in your browser. If you see a list of addresses, you're in business. If you see a 404, turn the sitemap on in your site builder or SEO plugin.

Step 2: Stop fussing over priority and change frequency

Those <priority> and <changefreq> tags someone told you to tune in 2014? Google ignores them now. It mainly uses two things: the page address and the last-changed date (0:04:40-0:06:00).

So stop tweaking them. Spend that energy on Step 3.

Step 3: Make your lastmod dates honest

This is the big one. lastmod should be the date of the last real change to that page (0:05:40-0:06:20).

If every page in your sitemap says "changed today," every day, Google learns your dates mean nothing and ignores them. It isn't treated as spam, it just stops trusting you. If your dates look reasonable, Google uses them (0:08:40-0:09:40).

Real test: on Oct 30, 2026 we checked two live small-business sitemaps. One listed 45 pages with zero lastmod dates. The other listed 221 pages with just 3. That's one of the only two things Google reads, missing.

What to do:

1. Open your sitemap and search for <lastmod>.

2. None at all? Look in your site builder or SEO plugin for a "last modified" setting and turn it on.

3. All the same date? Your setup stamps "today" on everything. Fix the setting so it uses each page's real edit date.

Step 4: Shops, this one's for you

Google said shops gain the most here. A price change on a product page gets noticed faster when the sitemap says that page changed (0:08:00-0:08:40).

So if your store updates prices or marks items sold, make sure that page's lastmod moves to that day.

Step 5: List only clean addresses

Only the clean, main version of each page belongs in the sitemap. No ?utm= tracking tags, no date stamps in the address (0:09:40-0:11:00). Being in the sitemap nudges Google to pick that address as the main one, so feed it the right one.

While you're in there, delete junk: thank-you pages, test pages, a /404 page.

Step 6: Use the plain name and say where it is

Call it sitemap.xml. If you give it a secret name, you have to tell Google and Bing yourself, and AI crawlers have no console to tell. Mueller said he has seen AI crawlers read his sitemap and RSS feed in his own server logs (0:14:00-0:16:40).

Then add one line to your robots.txt file so every crawler finds it:

Sitemap: https://yoursite.com/sitemap.xml

Step 7: Turn your RSS feed on

An RSS feed works like a mini sitemap for your newest pages, and you can submit it in Search Console too (0:11:20-0:12:40). If your blog has one, keep it on. If it doesn't, switch it on.

Step 8: Know the size limits

One sitemap file holds up to 50,000 addresses and 50 MB (uncompressed). Bigger site? Use a sitemap index, which is a sitemap of sitemaps, and list it in robots.txt (0:12:40-0:14:00).

Step 9: Don't confuse the HTML sitemap with the real one

That "Sitemap" page in your footer is for people. It doesn't replace the XML file (0:17:00-0:18:40). Keep both if you like, but the XML file is the one Google reads.

Step 10: Don't panic at "Couldn't fetch"

Search Console says "Couldn't fetch" on a sitemap that opens fine in your browser? Google said this is often not a bug. Google is busy, or doesn't want more from the site right now, and that demand is often based on how good Google thinks the site is (0:22:00-0:24:00).

So check the file opens. If it does, the fix is better pages, not more code.

And llms.txt?

The trendy "llms.txt" file does not replace a sitemap. Mueller's words: "currently, none of this happens" and "I would not treat that as a priority" (0:19:20-0:21:40). Sitemap first.

Do this today

1. Open yoursite.com/sitemap.xml.

2. Search for <lastmod>. Missing or all the same date? Fix the setting.

3. Delete tracking-tag addresses and junk pages from the list.

4. Add Sitemap: https://yoursite.com/sitemap.xml to robots.txt.

5. Make sure your blog's RSS feed is on.

Related reading

• How to install siteprobe to check every page in your sitemap

• Robots.txt rules for AI bots that actually work

• Technical SEO for AI search: keep, start, stop

Sources

• Google Search Central (Martin Splitt and John Mueller) - Do sitemaps still matter? (2026-10-01)

• GitHub: bartTC/siteprobe

Don't worry if this seems like it's a pain in the ass for someone who doesn't build websites. That's because it is. The good news: just click the link below, pull the trigger, and we'll take care of the rest.