Your XML Sitemap Is Probably Telling Google the Wrong Things

Most WordPress sitemaps quietly include things that should never be in one — noindexed pages, thin tag archives, attachment URLs, parameterised duplicates — and stamp lastmod dates that are simply untrue. Both waste the crawl budget you were trying to direct, and one of them teaches Google to stop trusting the file.

An XML sitemap should list every URL you want indexed and nothing else: canonical, indexable, 200-status pages only, with honest lastmod dates, split into a sitemap index when the site grows large, and submitted once to Google Search Console where you then monitor it rather than resubmit it. A sitemap does not make pages rank; it makes them discoverable and tells crawlers when they changed. This guide covers what belongs in a WordPress sitemap and what never does, how sitemap indexes and image entries work, why lastmod honesty matters more than most settings, and how to troubleshoot the stale-cache problems that make sitemaps quietly lie.

What a sitemap is (and is not)

An XML sitemap is a machine-readable list of URLs you are inviting search engines to crawl, optionally annotated with when each URL last changed. It solves a discovery problem: on large sites, deep pages and freshly published ones can wait a long time to be found by link-following alone. The sitemap shortcuts that.

Equally important is what it is not:

• Not a ranking boost. Inclusion in a sitemap does not make a page rank better; it makes it findable sooner.
• Not an indexing command. Google treats the sitemap as a strong hint, not an order. Listed pages can still be skipped; unlisted pages can still be indexed via links.
• Not a substitute for internal links. A page reachable only through the sitemap is an orphan, and orphans get treated accordingly. Sitemaps and internal linking work together, not interchangeably.

What belongs in your sitemap

The membership rule: a URL belongs in the sitemap if, and only if, you want it indexed and it is capable of being indexed. Concretely, every listed URL should be:

• Canonical. The one true address of the content, matching its own canonical URL, not a variant.
• Indexable. No noindex robots meta, not blocked in robots.txt.
• A 200. Live, not redirecting, not erroring.
• Worth indexing. Real content a searcher could want to land on.

For a typical WordPress site that means posts, pages, meaningful custom post types, and taxonomy archives that serve genuine browsing value. Hydrogen SEO generates XML sitemaps automatically along these lines, with exclusion controls for anything that should not be invited.

What never belongs

Every entry that fails the membership rule sends crawlers a contradictory or wasteful signal:

• Noindexed URLs. Listing a page while its robots meta says "do not index" is a direct self-contradiction, and Search Console will report it as such. Keep your robots meta settings and your sitemap telling the same story.
• Redirecting URLs. The sitemap should list final destinations. Old URLs belong in your redirect rules, not your sitemap.
• Thin and utility pages. Cart, checkout, login, thank-you pages, internal search results, paginated fragments with no standalone value. Inviting a crawler to these spends its attention on pages you would never want in results.
• Parameterized URL variants. ?sort=price, ?utm_source=…, session parameters, filtered duplicates. These are alternate doors to canonical pages; listing them invites duplicate crawling of content that already has one true URL.
• Attachment pages and other WordPress detritus. Bare media attachment pages are classic thin content most sites should exclude.

If some of these already pollute the index, exclusion from the sitemap is step one; the audit checks in site audits help surface the thin and duplicated pages worth cleaning up at the source.

Sitemap indexes: how large sites stay legible

The sitemap protocol caps a single file at 50,000 URLs and 50MB uncompressed, but the practical reason to split earlier is organization. A sitemap index is a sitemap of sitemaps: one parent file linking to child sitemaps, typically split by content type (posts, pages, each taxonomy).

This structure has a monitoring superpower: Search Console reports discovery and indexing per child sitemap. When "product pages" shows a problem, you know where to look, instead of staring at one 30,000-URL blob. Hydrogen SEO's sitemap output works this way out of the box, and developers can shape the output further with filters like hydrogen_seo_sitemap_index_xml and hydrogen_seo_sitemap_exclude_item documented in the filters reference.

Image entries: helping image search find your visuals

The sitemap protocol allows image entries attached to page URLs, telling crawlers "this page contains these images." That matters for sites where image search drives real traffic (recipes, products, portfolios, tutorials with diagrams), because images loaded in ways crawlers parse poorly can otherwise go undiscovered.

Hydrogen SEO includes image entries in its XML sitemaps automatically. Image discovery pairs naturally with image quality: an image found via the sitemap still needs descriptive alt text to be understood and ranked in image results.

Video and news sitemaps

Two more sitemap types exist for specific kinds of site, and most sites need neither.

A video sitemap describes the videos on your pages: title, description, thumbnail and where the video plays. It helps Google find videos it would otherwise miss, especially embeds, and show them in video results. It is worth turning on if video is a real part of your content, not for one embedded clip on an About page. Hydrogen SEO publishes one at /video-sitemap.xml, splitting into video-sitemap2.xml and onwards on large sites; see the video sitemap doc.

A Google News sitemap is different in kind: it lists only articles published in the last 48 hours, so Google News can pick up fresh stories quickly. It is useful only if your site appears in Google News, and it never helps ordinary pages get indexed; the regular sitemap does that. An empty news sitemap on a quiet day is normal. One that lists week-old articles means something is wrong. Hydrogen SEO serves it at /news-sitemap.xml; see the news sitemap doc.

Both are off by default in Hydrogen SEO. Once switched on they are listed in your sitemap index (the news sitemap whenever it has articles in it), so submitting the index in Search Console covers them.

Lastmod honesty: the setting that is actually a promise

lastmod tells crawlers when a URL's content last changed, and Google has said publicly that it uses lastmod when it has proven trustworthy on your site, and ignores it when it has not. That makes lastmod a credibility account:

• Deposit: lastmod updates when content genuinely changes, and crawlers that recrawl find real changes.
• Withdrawal: lastmod claims "today" on every URL, every day (a common misconfiguration and an occasional dishonest tactic), crawlers find nothing changed, and the field gets ignored sitewide.

The honest configuration is boring: let lastmod reflect the real modified date, do not touch-update posts to fake freshness, and accept that a page unchanged since 2023 says so. On a healthy site, honest lastmod is what gets your genuinely updated pages recrawled quickly, which is the entire point. The related priority and changefreq fields, for what it is worth, are largely ignored by Google; do not spend effort gaming them.

Submitting and monitoring in Search Console

Submission is a one-time act; monitoring is the ongoing job.

1. Find your sitemap URL (for Hydrogen SEO sites, the XML sitemaps doc shows where it lives, and the plugin can reference it from robots.txt via the robots.txt editor).
2. In Search Console, open Indexing → Sitemaps, paste the sitemap index URL, submit. The step-by-step walkthrough is in submit your sitemap to Google.
3. From then on, watch rather than resubmit. Google refetches sitemaps on its own schedule; resubmitting daily changes nothing.

What to watch monthly:

• "Couldn't fetch" or parse errors on the index or any child
• Discovered vs. indexed gaps. A child sitemap where most URLs are discovered but few are indexed points at quality or duplication problems in that content type, not a sitemap problem.
• "Sitemap contains noindexed/redirecting URLs" style findings, which mean your sitemap and your site disagree and one of them needs fixing.

Troubleshooting stale sitemap caches

Sitemaps are usually cached for speed, in the plugin, a page cache, a server cache, or a CDN, and a stale cache makes the sitemap lie: you publish a post and the sitemap does not show it, or you exclude a URL and it stubbornly remains.

Work outward through the layers:

1. Plugin cache. Hydrogen SEO caches sitemap output; clearing it regenerates the XML from current content.
2. Page cache / CDN. If the regenerated sitemap still serves old XML, a caching layer above WordPress is holding the previous copy. Purge the sitemap URLs specifically, and consider excluding sitemap paths from long-lived caching entirely; sitemaps are cheap to generate and expensive to serve stale.
3. Verify like a crawler. Fetch the sitemap URL in a private window or with curl, not the wp-admin preview, and confirm the entry you expect.
4. Check the weird cases. Wrong URLs in the sitemap after a domain or HTTPS change usually mean a stale site URL setting or cache from before the switch.

The sitemap troubleshooting guide walks these steps in order, including the multisite and staging edge cases.

The checklist

Indexable, canonical, 200-status URLs only. No noindexed, thin, redirecting, or parameterized entries. Sitemap index split by type. Image entries on for visual content. Video and news sitemaps only if video or Google News matters to you. Honest lastmod. Submitted once, monitored monthly, caches cleared when reality and XML disagree.

Hydrogen SEO does the generating, image entries, caching, and exclusions for you, free, no account required, with audits to catch the contradictions. Get started free and let your sitemap tell the truth automatically.