Deciding what deserves to be indexed at all. On a large catalog, the most valuable technical decision is usually subtraction rather than addition: fewer, stronger URLs that Googlebot can reach quickly and that a searcher would actually want to land on.
A site with two hundred thousand indexed URLs and eight thousand that anyone would want to land on is not a strong site. It is a site spreading its crawl and its internal links across pages that will never rank and never convert.
Pagination that goes forty pages deep, tag pages generated automatically, near-identical location pages, filter combinations nobody searches for. Each was cheap to create and each carries an ongoing cost in crawl requests and diluted signals.
Deciding what should not be indexed is unglamorous work. The indexed-page count goes down before traffic goes up, which is why it tends to be postponed, and why it is so often the highest-leverage work available.
One documented rule for how canonical URLs are chosen and how paginated lists are handled, applied the same way on every template instead of decided feature by feature by whoever built it. We check the rule against the crawl, list every template that breaks it, and note where Google has chosen a different canonical from the one you declared, because that disagreement is usually a symptom of something else.
We build the internal link graph from the crawl and measure how links, and the authority they carry, flow through it. Commercially important pages are often five or six clicks from the home page while an auto-generated tag archive sits in the main navigation. The output is a short list of pages to promote, pages to demote, and the template changes that do it at scale.
Sitemaps split by template and by freshness, so indexing can be measured per segment in Search Console. A single sitemap file set with two hundred thousand URLs tells you nothing when coverage drops; ten segments tell you which template dropped and roughly when. We specify the segments and the update rules, and your developers generate them.
Where structure has to change, every legacy URL is mapped to its new destination before anything moves, including old redirects that would otherwise turn into chains. After launch we check the map against a fresh crawl and, on Retainer and Embedded, against the logs, rather than assuming it is correct. Migration planning support is part of the Embedded plan.
A full crawl compared against your raw server logs: where Googlebot spends its requests, which templates it starves, and which parameter is generating thousands of duplicate URLs.
Raw HTML against the rendered page for every template, a rendering decision per route, and Core Web Vitals work measured on field data rather than a lab score.
$160, fixed price, two weeks. Findings ranked by impact with the evidence attached, and the fee credited against your first month if you continue.
Get in touch