Core Concepts
Before diving into the features, it's helpful to understand the fundamental concepts and terminology used throughout Lociator.
Graph Fundamentals
Lociator models your website as a directed graph. This is the same mathematical structure used by search engines (like Google's PageRank) to understand how pages relate to each other.
Nodes (Pages)
Each crawled page becomes a node in the graph. A node stores:
- URL and normalized URL (for deduplication)
- Depth — click distance from the root
- Status code — HTTP response status
- Title — extracted and cleaned from the
<title>tag - In-degree / Out-degree — count of incoming and outgoing internal links
- Orphan status — whether the page has zero incoming links
- Extracted text — clean content for topic analysis (up to 2,000 chars)
- Parent page — the BFS parent that first linked to this page
Edges (Links)
Each internal link from one page to another creates a directed edge. Edges store the source page, target page, and anchor text. Self-links (a page linking to itself) are automatically filtered out.
Crawl Depth
Depth represents the minimum number of clicks needed to reach a page from the root URL, determined by BFS traversal order. The homepage has depth 0, pages directly linked from the homepage have depth 1, and so on.
In-Degree & Out-Degree
In-degree is the number of internal links pointing to a page.Out-degree is the number of internal links going from a page. These are computed during graph building and stored in the database.
| Metric | Low Value | High Value |
|---|---|---|
| In-degree | Hard to discover (potential orphan) | Well-linked, authoritative page |
| Out-degree | Distributes little link equity | Hub page connecting to many others |
Orphan Pages
An orphan page has zero incoming internal links (in-degree = 0) and is not the root page. Orphans can only be found via sitemaps or direct URLs — search engine crawlers following internal links will never discover them.
Link Equity
Link equity (or "link juice") is the concept that links pass authority and ranking signals between pages. A well-structured internal linking strategy ensures important pages receive adequate equity through both direct and indirect links.
Topic Clusters
Lociator automatically groups pages into topical clusters by analyzing their extracted text content using LLM embeddings. Pages with similar content are grouped together, revealing your site's topical structure and helping identify content gaps or misaligned linking.
Pillar Pages
A pillar page is a hub page with high out-degree (≥10 outgoing links) and moderate in-degree (≥5 incoming links). The metrics engine identifies these pages and rewards sites that have an appropriate ratio of pillar pages (roughly 1 per 20 pages).
Content Silos
Content silos are groups of related pages organized under common URL paths (e.g., /blog/,/products/). The system identifies silos using the first path segment and measures cross-silo linking — the ideal range is 15–30% of links crossing silo boundaries.