Website SEO Services

The test is intent, not keyword overlap. Two pages can share vocabulary and both be legitimate if a searcher would be satisfied by different things — a page explaining what a service costs and a page selling the service target genuinely different moments. Cannibalisation is when a single searcher would be equally served by either page, which shows up in Search Console as both URLs collecting impressions for the same query, often alternating as the ranking URL week to week. That flip-flopping is the clearest diagnostic available, because it is Google visibly failing to decide. When we find it, the decision is not which page is better written but which already holds the backlinks and impression history, since that is the equity you consolidate toward rather than throw away.

Consolidate when the page has accumulated anything worth keeping — external links, meaningful impression history, or ranking positions on any query — because a 301 passes that equity to the target while a deletion discards it. Noindex is the right tool when a page needs to remain accessible to users but has no business in search results, such as thank-you pages, internal search results or thin filtered views. Straight deletion with a 410 is appropriate only for genuinely worthless URLs with no links and no history, where you actively want Google to stop crawling them quickly. The mistake we see most often is 301-ing dozens of unrelated pages to the homepage, which Google treats as a soft 404 and ignores entirely, so the redirect target must be a genuinely equivalent page or the equity evaporates.

Because most sites already have content that cannot be found. Authority enters your site at a small number of pages — usually the homepage and a couple of link-earning posts — and internal links are the only mechanism distributing it inward. A well-written page with no internal links pointing at it is discovered late, crawled infrequently, and given no contextual signal about what it should rank for. Restructuring links costs a fraction of what new content costs and acts on assets you have already paid for, which is why we sequence it before any writing. It also compounds: every new page shipped into a properly structured cluster inherits relevance from its parent instead of starting from zero.

Word count is a symptom, never a target. When we analysed the top-ranking service and location pages in our own HQ market, all five were running five to eight thousand words with heavy H2 and H3 subdivision — but that depth is table stakes in that market, not the reason they rank, and writing eight thousand thin words will not replicate it. The measure that matters is subtopic coverage relative to the pages currently ranking: pull the top results, list every distinct question each one answers, and find what yours omits. Sometimes that means adding three thousand words; sometimes it means cutting two thousand words of preamble so the answer arrives in the first screen. Depth without specificity is just length, and length has never been a ranking factor.

Index bloat is having far more URLs indexed than you have genuinely distinct pages, typically generated by faceted navigation, tag archives, pagination, session parameters or printer-friendly duplicates. The clearest signal is a large gap between the page count you would estimate and what Search Console reports as indexed, particularly when the extra URLs are variations of the same content. It hurts in two ways: crawl budget is finite, so time spent on worthless URLs is time not spent recrawling pages you have just improved, and a large volume of near-duplicate thin URLs affects how Google assesses overall site quality. The fix is diagnostic rather than uniform — canonical tags for genuine duplicates, noindex for pages users need but search does not, and robots or parameter rules for crawl traps.

Most of it is usually recoverable, because the underlying loss is almost always structural rather than qualitative. The recoverable causes, in the order we check them: missing or chained redirects from old URLs, changed URL patterns without mapping, removed content that was ranking, and an internal linking structure rebuilt around a new visual design with no regard for the previous link graph. We reconstruct the old URL inventory from Search Console exports and archive crawls, restore the redirect map, and reinstate the content and links that disappeared. What is genuinely hard to recover is external links pointing at URLs that were deleted and redirected to irrelevant targets long enough ago that the referring sites have since dropped them, which is why speed matters once a drop is identified.