Site Inventory Build + Duplicate Page Geographic Differentiation Strategy (April 6, 2026)

Session Overview

Sam asked to run the site-inventory-updater skill against the live site (aguiarinjurylawyers.com) to build a complete URL inventory. The session evolved into three major workstreams:

Building the full site inventory from live sitemaps (499 URLs categorized)
Investigating 7 flagged duplicate/cannibalization URL pairs by querying the WordPress database directly via SSH
Designing and beginning execution of a geographic differentiation strategy: keep the stronger page in each pair targeting Louisville, retarget the weaker page toward Lexington and Eastern Kentucky

Sam also uploaded 12 skill/template reference files (website-rules, sa-template-reference with 9 component references, wp-content-deployer) and asked that they be saved and that conflicting skills in the directory be updated.

A secondary question came up mid-session: whether the Screaming Frog headless crawl script (run-crawl.sh) could be automated as a Claude Code scheduled task.

What Was Accomplished

1. Full Site Inventory (Complete)

All 4 sub-sitemaps were fetched and parsed via SSH into the Cloudways server: pages, posts, categories, and video sitemaps. 499 total live URLs were categorized into types (Homepage, Core Page, Practice Area Hub/Spoke, Injury Page, Truck Talk, Location Page, Lexington Page, Blog/Resource, Category, Video, etc.).

Three reference documents were created from scratch:

page-inventory.md: Complete URL inventory with all 499 URLs categorized by type. This is the foundation document for all future content and SEO work.
topical-map.md: Hub-and-spoke content cluster mapping for all major practice areas (Car Accidents, Truck Accidents, Injuries, Motorcycle, Rideshare, Delivery Vehicle, Pedestrian, Wrongful Death, Premises Liability, Insurance/Legal, Locations, Lexington). Identifies related standalone pages that should link to hubs.
internal-linking.md: Structural orphan candidates (18 pages), potential duplicate/cannibalization pairs (7), and a hub-to-spoke link audit checklist. Notes that Screaming Frog crawl data is needed for real inlink counts.

All three files are in: /Projects/Website_project/site-map/

2. Duplicate Content Investigation (Complete)

SSH'd into the server and queried the WordPress database directly for content lengths, RankMath focus keywords, SEO titles, meta descriptions, and robots directives on all 16 pages across the 7 flagged pairs. Key finding: most pairs are different content targeting similar keywords (true cannibalization), not actual duplicates.

The 7 pairs investigated:

Pair	Page A (Stronger)	Page B (Weaker)	Finding
Rollovers	/rollover-accident/ (ID 45346, 18K chars)	/rollovers/ (ID 8285, 8K chars)	Different content, same keyword area
Burn injuries	/burn-injuries/ (ID 6888, 23K chars)	/burn-injury/ (ID 6916, 6K chars)	Different content + shell page ID 25391 (1K chars)
Pedestrians	/pedestrian-lawyers/ (ID 6901, 12K chars)	/pedestrians/ (ID 7061, 4K chars)	Lexington version already exists at /lex-cases/lexington-pedestrian-accident/
Rain accidents	Two pages targeting rain/wet road crashes	Need to decide which gets Louisville	Different angles on same topic
Brain/TBI	Two pages on brain injuries	Borderline, may coexist	Different depth levels
Wrongful death	Main page + shell page (ID 43538, 1K chars)	Shell needs content or redirect	Shell page is thin
Car accident lawyer	Two posts on same slug	Slug collision needs resolution	Potential WordPress conflict

Session Overview

What Was Accomplished

1. Full Site Inventory (Complete)

2. Duplicate Content Investigation (Complete)

3. Geographic Differentiation Strategy (Approved, Not Yet Executed)