Our sitemap generator crawls your site the same way a search engine does: it fetches your home page, follows the links it finds, and keeps going. That design has a useful side effect. When a crawl surprises you, it is almost always your site doing the talking, and the same surprise is waiting for Google. We spent years answering sitemap questions on our old support forum, and the fixes below covered nearly all of them.
The crawl stops at the first page
The classic cause, then and now, is a redirect mismatch. Suppose your site lives at www.example.com but you type example.com into the generator, or the other way around. The crawler fetches the address you gave it, gets redirected, and depending on how your site handles the move, may find a page whose links all point somewhere it was not asked to crawl. Years of forum threads that started “it only finds my index page” ended with exactly this.
Two fixes. For the crawl: submit the address your site actually settles on. Load your site in a browser, look at the address bar after it finishes, and give the generator that, https or not, www or not. For your site: pick one form and permanently redirect the other to it. Search engines see www and non-www as two different sites, and splitting your identity between them costs you.
If the address checks out, look at robots.txt. The generator honors it, so a disallow rule that covers your whole site produces a one-page crawl. And if your navigation only exists as script-driven menus with no real links in the HTML, a crawler has nothing to follow. Search engines hit the same wall.
The sitemap has fewer pages than my site
First, the honest limits: the free crawl covers up to 600 pages, three levels deep from the page you submit, within a two-minute window. If your site is bigger or deeper than that, the crawl is working as designed, and the large-site section on the generator page covers your options.
Inside those limits, missing pages usually mean missing links. A page nobody links to (an orphan) cannot be discovered by any crawler. A section you disallowed in robots.txt is skipped on purpose. And a page that only exists four clicks from the home page sits beyond the depth limit; if a page matters, link to it from somewhere prominent, which is good advice for search rankings anyway.
It found more pages than I have
The opposite complaint is just as common. One shop owner watched five real pages balloon into a hundred entries because every link carried a different tracking code on the end. Query strings, session IDs, and filter combinations can make one page look like twenty.
The durable fix is at the site level: link to each page one way, set a canonical tag on pages that answer to several addresses, and let redirects clean up the rest. That shrinks the sitemap and, more to the point, stops search engines from splitting one page’s credit twenty ways.
The XML looks strange in a browser
Two different things get reported as “strange characters,” and only one is a problem.
Escaped punctuation is correct. XML reserves a few characters, so an address containing an ampersand appears as & inside the file. Raw XML also renders as a wall of tags in a browser. Both are normal, and Google reads them fine.
Genuinely garbled accents or symbols are an encoding mismatch: the page claims one character encoding and serves another, and non-English characters pay the price. The fix is to serve your pages as UTF-8 and declare it, once, consistently. This solved every “weird characters” thread our forum ever saw, from Czech accents to Russian product names.
Google says the sitemap is invalid or not found
“Not found” (a 404) means the file is not where you told Google it was. Upload sitemap.xml to the folder that holds your home page, then prove it: load yoursite.com/sitemap.xml in a browser. If you can see it, Google can. If you cannot, fix that before touching Search Console. The generator page walks through the upload and submission steps.
“Invalid tag” or parse errors almost always mean the file Google got was not a clean sitemap: an editor “helpfully” reformatted it, an upload corrupted it, or the wrong file went up. Upload the exact sitemap.xml you downloaded, unedited. On our old desktop tool, the number one cause was people uploading a project save file instead of the exported sitemap; the format has changed, the lesson has not.
Do I need lastmod, priority, or gzip?
Less than you would think. Google has said publicly that it ignores priority and change frequency, and reads lastmod only when it proves trustworthy. Gzip compression is allowed, never required. The hard limits worth knowing: one sitemap file may hold up to 50,000 URLs and 50 MB uncompressed, and beyond that you split into multiple files listed in a sitemap index. At 600 pages and under, none of these ceilings are in play.
One closing thought from years of these questions: a sitemap is a discovery aid, not a ranking trick. It tells search engines where your pages are; it does not make them rank. The crawl is still worth reading closely, because its errors and oddities are a free preview of what Googlebot will hit. For the fuller story on what sitemaps are and are not, see our XML sitemap page.