A sitemap should list only pages that exist, return successfully, and are meant to be indexed. Remove deleted pages, redirected addresses, and anything you have blocked.

What a sitemap is for

A sitemap is a file listing the addresses on your site that you want search engines to know about, usually with a date showing when each was last changed.

It is a suggestion rather than an instruction: it helps things get found, particularly on larger sites or pages that are not well linked internally, and it does not guarantee anything is indexed.

Which means its value is modest and real, and its cost is that an inaccurate one actively misinforms.

Most small business sitemaps are generated automatically and are broadly fine. The ones worth checking are the ones nobody has looked at since the site changed.

What should not be in it

The third and fourth are the contradictions worth catching: telling a search engine to look at a page while also telling it not to is a conflicting instruction, and the outcome is unpredictable.

Why it drifts

Sitemaps go stale in a small number of predictable ways.

A site is rebuilt and the sitemap is regenerated by the new system, while the old one is left in place at its original address and is still the one submitted.

A plugin generating the sitemap is disabled or replaced, leaving a static file that no longer updates.

Pages are removed during a content review, and the sitemap is not regenerated afterwards.

Or a section is blocked from crawling for a good reason, and nobody removes it from the list of pages being advertised.

None of that is negligence, and all of it is invisible unless somebody opens the file.

Open it and look

The check, and it needs no tools.

Your sitemap is usually at a standard address, commonly the domain followed by sitemap.xml, and it is readable in a browser.

Some systems produce an index that points at several smaller sitemaps, which is normal and means opening the ones it lists.

Read the addresses. You will recognise pages that no longer exist faster than any tool will.

Then click a sample of ten or fifteen, spread across the list, and confirm each loads rather than erroring or redirecting.

That is five minutes and it finds almost everything worth finding.

A worked example

A business rebuilt their site and the new platform generated a sitemap at a slightly different address.

The old file remained on the server, and it was the one that had been submitted years earlier.

So for eighteen months they were advertising a list of addresses from the previous version of the site, most of which now redirected and several of which had been removed entirely.

Their new pages were being found through ordinary crawling and were not in any sitemap at all.

The fix took ten minutes: delete the old file, submit the new one, and confirm the content matched the site.

The lesson they took was that nobody had opened the file in eighteen months, and that opening it would have been the first thing to reveal the problem.

Check what the tools report

The other half, which is free and shows what was actually made of the file.

Search Console reports how many addresses were submitted, how many were indexed, and errors it encountered, which is the feedback that turns a guess into a fact.

A large gap between submitted and indexed is worth understanding rather than panicking about, since not every page will or should be indexed.

Errors naming specific addresses are the actionable part, and they usually correspond exactly to the stale entries above.

Resubmit after making changes, which prompts a re-read rather than waiting for the next visit.

Keep it small and honest

A principle rather than a rule, and it applies particularly to small sites.

A sitemap listing your twenty real pages is more useful than one listing four hundred addresses including every tag, archive, and filtered view a platform generates.

Many platforms include those by default, and most of them are pages you would not want anybody to land on.

Configure the generator to exclude tag and archive pages unless they are genuinely useful, which for a small business site they rarely are.

The last modified dates should also be true. A generator that stamps every page with today's date whenever anything changes is providing noise rather than information.

Point to it from robots

A small step that helps and is frequently missed.

The robots file can name the location of your sitemap, which means anything reading it can find the sitemap without being told separately.

That is one line and it covers search engines you have not submitted to directly.

While you are there, confirm the sitemap address in that line is the current one, since it is another place a stale address hides after a rebuild.

Then submit it directly as well, in the tools that offer it, since the two routes are complementary rather than alternatives.

The counter-case

Sitemaps matter less than the attention they attract.

For a small, well-linked site where every page is reachable within a couple of clicks from the homepage, a sitemap adds very little, and a perfect one will not improve anything measurable.

Time spent tuning it is usually time that would be better spent on the pages themselves.

There is also a habit of treating sitemap errors in reporting tools as urgent problems, when many are informational and reflect deliberate decisions.

Make it accurate once, exclude the automatic clutter, check it after any rebuild, and then leave it alone.

The check

  1. Open your sitemap in a browser and read it.
  2. Click ten or fifteen entries at random.
  3. Remove deleted and redirected addresses.
  4. Remove anything you have blocked or marked not to index.
  5. Exclude tag and archive pages.
  6. Check for an old file left over from a rebuild.
  7. Name it in robots and resubmit.

Step six is the one that catches the situation above, where the file being advertised is not the file the site now generates.

Deciding which pages should exist is covered in which pages earned their place.

Frequently asked questions

What is a sitemap for?

It lists the addresses you want search engines to know about. It helps things be found, particularly on larger or poorly linked sites, and guarantees nothing.

What should not be in it?

Deleted pages, addresses that now redirect, anything blocked from crawling or marked not to index, duplicate versions, and staging addresses.

Why do sitemaps go stale?

A rebuild generates a new one while the old file remains and stays submitted, a plugin is replaced leaving a static file, or pages are removed without regenerating.

How do I check mine?

Open it in a browser, usually at your domain followed by sitemap.xml, read the addresses, and click ten or fifteen at random to confirm they load rather than erroring.

Should it include every page a platform generates?

No. Tag pages, archives, and filtered views are usually pages you would not want anybody to land on. Configure the generator to exclude them.

Does a sitemap matter much?

Less than the attention it attracts. For a small well-linked site it adds little. Make it accurate once, check after any rebuild, then leave it alone.

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.

Rebuilt your site recently?

Open your sitemap and check it matches the site you have now. An old file at the old address is the usual finding.

Start a Conversation