The biggest render-and-architecture leak is on product pages: 14 of 15 sampled products need JS for ~50% of their content + internal links, which is fragile under Googlebot's two-pass render budget. Crawl-budget bloat is the secondary issue — 8 parameterised URLs are crawled but non-indexable, and 9 indexable URLs have zero GSC clicks (likely thin / duplicate).
| Severity | Finding | Fix required |
|---|---|---|
| HIGH render_diff | 3 of 6 sampled URLs require JS to render content On these pages, Chrome adds >=30% more visible words or 5+ additional internal links compared to the raw HTML. Googlebot does render JS but with a delay and a budget — render failures, hydration errors or blocked scripts will silently strip this content from the index. | Server-side render (SSR) or pre-render the affected templates so the critical content + internal links exist in the raw HTML response. |
| MEDIUM render_diff | Page title changes after JS execution on 1 page The `<title>` differs between raw HTML and rendered DOM. Googlebot may index either version — SERP-visible titles depend on which version Google captures, which makes title-tag experiments and meta updates unreliable. | Render the final title server-side. Don't let client-side routing libraries set the title. |
| MEDIUM link_architecture | 6 of 28 sampled URLs have no internal inlinks Orphan URLs reach Googlebot only through the sitemap. They get crawled rarely, rank poorly, and consolidate no PageRank from the rest of the site. | Audit the orphans and either: link them from the nearest topical hub page, remove them with a 410 if stale, or 301 to a stronger sibling. |
| MEDIUM link_architecture | 24% of sampled URLs sit more than 3 clicks from the homepage Pages deeper than 3 clicks from home are crawled less frequently and rank harder. The industry rule is <20% at depth 4+; above that, architecture is bottlenecking commercial pages. | Add hub pages that surface deep product clusters; consider facet/view-all patterns to flatten. |
| MEDIUM crawl_budget | 29% of sampled URLs carry query parameters Parameter sprawl (8 sort variants + 4 paginated copies of /category/anvils) explodes crawl budget. Without deliberate handling, the long tail of variants dilutes PageRank and slows discovery. | Disallow `?sort=` in robots.txt; canonicalise paginated URLs to the bare category. Don't rely on canonicals alone. |
| MEDIUM crawl_budget | 41% of indexable URLs have zero GSC clicks 9 of 22 indexable URLs generate no organic clicks (cross-referenced against gsc-audit/latest). On a healthy site this share sits under ~30%. | Triage: consolidate thin product variants, refresh stale content, or noindex genuine archive pages. |