Use cases

What CrawlSense does — 5 jobs, one platform

Pick the use case that brought you here.

Use case 01

Detect keyword cannibalization automatically

Cannibalization happens when two or more pages on your site target the same intent. Google has to choose, your link equity gets split, and both URLs end up ranking lower than a single consolidated page would. It's one of the most common reasons a site stalls after publishing dozens of "long-tail" articles.

CrawlSense flags it for you. We embed every page with a sentence-transformer model and compare pairs by cosine similarity. When two URLs cross the 85% threshold and target overlapping queries in Search Console, they surface in the cannibalization report ranked by impact — so you start with the pair that's costing you the most clicks, not the one that simply looks similar.

Cannibalization pair 98% similar
/blog/how-to-do-keyword-research
312 impressions / week · pos. 14.2
/guides/keyword-research-guide
298 impressions / week · pos. 13.8

Sample. Your dashboard groups every pair, surfaces shared queries, and suggests merge or canonical actions.

Use case 02

Topic clusters built on semantic embeddings

Most "topic cluster" tools group pages by URL path or shared keywords. That misses the point. CrawlSense embeds the actual content of each page and runs HDBSCAN — a density-based algorithm that finds natural groupings without forcing every URL into a bucket. Pages that don't belong anywhere stay flagged as outliers, which is usually a signal worth investigating on its own.

From the cluster view we surface the Pillar Strategy: the page in each cluster that should anchor the topic, the supporting articles that already exist, and the gaps where a new article would close a hole. It's the same shape SEO consultants build manually in a spreadsheet — except you get it ten minutes after the crawl finishes.

Cluster #4

Technical SEO basics

12 pages
  • P /technical-seo-guide pillar
  • /blog/robots-txt-explained
  • /blog/sitemap-best-practices
  • Gap detected: "canonical tag examples"

Sample. HDBSCAN finds the cluster — the Pillar Strategy view ranks the anchor and lists gaps.

Use case 03

Internal linking analysis with PageRank

Internal links shape how authority flows through your site. CrawlSense runs PageRank over the internal graph after each crawl and reports the distribution: which pages concentrate authority, which earn far less than their commercial value would justify, and which sit completely orphaned with zero inbound links from anywhere on the site.

Same report flags broken internal links (404s reached from your own templates), redirect chains that waste a hop, and nofollow links going to pages you actually want indexed. The output is a ranked list of fixes, not a graph viz — because the goal is shipping the changes, not admiring the diagram.

Top authority
/pricing
PageRank 0.087 · 142 inbound
Orphan pages
23
0 internal inbound links
Broken internal
7
404s linked from templates

Sample numbers. The actual report ranks each issue by the PageRank cost of leaving it in place.

Use case 04

Core Web Vitals on mobile and desktop

LCP, CLS and TBT measured per page, on both mobile and desktop, after every crawl. We sample a representative subset (full coverage gets expensive fast) and produce a score band for each metric so you can see at a glance which templates are inside the green threshold and which slipped into "needs improvement" since the last run.

The point isn't to replace PageSpeed Insights for one-off URL checks. It's to give you a per-template trend across hundreds of pages, so you spot a regression the morning after a deploy instead of waiting for the field data in Search Console to catch up six weeks later.

Core Web Vitals — mobile
Good Needs work Poor
LCP
2.1s
CLS
0.14
TBT
180ms

Sample. Each crawl produces a score band per template — desktop and mobile reported side by side.

Use case 05

Connect GSC and GA4 in seconds

Both integrations run over OAuth, read-only. We never write to your Search Console or Analytics property and we only request the scopes we actually use. The connection takes two clicks per site — pick the property, accept the consent screen, and we start pulling data.

We keep a 30-day rolling window of organic-only traffic per URL. That feeds the cannibalization report (so we know which pairs are actually competing for the same query), the cluster view (so the pillar suggestion accounts for which page is already winning), and the internal linking report (so authority recommendations consider real downstream traffic, not just topology). Disconnect anytime from site settings — we delete the cached data within 24 hours.

Search Console
Read-only · 30-day window · organic only
Connected
Analytics 4
Read-only · sessions + engagement per URL
Connected

Sample. OAuth flow takes about 30 seconds per property and only requests read scopes.

All five, on one platform

Cannibalization, clusters, internal linking, Web Vitals and GSC + GA4 — same crawl, same dashboard, no extra tools to wire up. Plans start at €59/month with a 14-day free trial.

See pricing

Ready to put all five to work?

14-day free trial. No credit card required.