A boutique law firm came to us with a site that looked healthy on paper. Its sitemap was clean, every URL in it loaded, and nothing was broken. The trouble was everything outside the sitemap. A full crawl showed search engines could reach thousands of pages the firm never meant to put in front of them, and the pages that mattered were buried under the noise.
- 3,320crawl trap URLs taken out of search
- 191pages with the wrong or missing canonical, all fixed
- 4 → 1duplicate blog hubs folded into one each
The problem
The sitemap listed 1,486 URLs. Our crawl found 2,162 indexable pages, plus another 3,320 generated by a fault in the comment pagination. Every blog post was spinning off endless near-empty copies of itself, running to dozens of pages on a single article.
Four blog hubs lived at two addresses each, so the navigation pointed to one version while the sitemap listed the other. Roughly 3,800 pagination and archive pages were set to index, including hundreds of pages all titled as “uncategorized.”
Smaller problems were stacked on top. Navigation and footer links pointed at redirects. Dozens of posts linked to a practice page that was still a draft. And nearly two hundred speaking and publication pages carried canonical tags that pointed at someone else’s website, or at nothing at all.
What we did
- Shut down the pagination fault and set the crawl trap URLs to noindex, then confirmed in Search Console that Google saw it.
- Set pagination, attorney profile pagers and taxonomy archives to noindex, follow, so link value still flows to the real pages.
- Redirected the four duplicate hubs to a single version each and updated the sitemap to match.
- Repointed navigation, logo, footer, menu and template links at their final URLs, and removed author archive links that were hiding hundreds of redirecting links.
- Hid draft practice pages from the related practice lists, so the front end never links to a page that does not exist yet.
- Fixed canonicals, share URLs and schema on 191 speaking and publication pages so each one points to itself.
- Re-crawled the site after every change and checked the live result before moving on.
The result
Search engines now see a site that matches the one the firm actually runs. The thousands of trap pages are out of the picture, each hub has one address, and every speaking and publication page claims itself instead of pointing at another site. Re-crawls after each fix came back clean.
Begin a project