CONTENT & CMS RECONCILIATION
Multilingual publication records
Reconciled duplicate records, translation relationships, public destinations, draft states and redirect verification. Kept routing failures distinct from successful record changes.
SELECTED WORK / TECHNICAL INVESTIGATION & QA
I follow the evidence, not the easiest explanation.
The most useful part of an investigation is knowing what the evidence actually establishes—and what it doesn't. I trace symptoms across search data, website output, content records, redirects and change histories, then document decisions and validation.
01 / SEARCH BEHAVIOR & PAGE IDENTITY
When the wrong page appears for the right query
A metropolitan law-firm practice page and its parent hub competed for search exposure.
The question
Was Google favoring a general personal-injury hub over a more specific car-accident page because of a page-identity defect—or because of a broader relevance problem?
My investigative approach
Compared exact queries and URLs across dates, segmented devices and markets, checked indexed canonicals and current page content, and reviewed internal links and change timing before recommending an action.
Observed: hub shown more oftenVerified: both URLs independently indexedUnproven: the hub caused weak rankings
What the evidence established
The search-data preference was reproducible; both pages remained indexed with distinct self-canonicals and intended content. Earlier records showed the specialized page was already weak before the observed switch.
Decision & status
Retain both pages. A targeted contextual link was proposed as a measured test; no redirect, merge, or blanket rewrite was justified. Search-performance causality remains open.
02 / MULTI-PAGE CLUSTER INVESTIGATION
Auditing a city-wide content cluster
Separating actual defects from weak performance signals across a large page group.
The challenge
A city hub and its related service pages had uneven search visibility, complex internal-link paths and questions about whether Google had recognized newer page versions.
My contribution
Organized the investigation by page and evidence source, reviewed destination intent and internal links, distinguished incorrect links from ranking observations, and coordinated bounded, approved corrections with verification records.
Inventory pages→Compare saved & public output→Classify findings→Approve targeted edits→Verify & record
Confirmed findings
The November re-audit reported all ten core pages indexed with matching canonicals, alongside wrong-intent destinations and other localized issues requiring focused handling. A later approved mechanical correction batch was saved and exact-diff checked.
Evidence limits
Some ordinary-browser, cache and mobile checks remained outstanding after the latest batch. Google recognition of the latest versions and any resulting search impact were not proven. The case remains an investigation, not a ranking-recovery claim.
03 / HISTORICAL CHANGE & CAUSAL ANALYSIS
Did an internal-link cleanup cause a decline?
Testing a plausible explanation against actual link-change receipts.
The hypothesis
A previous link cleanup may have removed inbound references to an important city-specific car-accident page, contributing to a later visibility decline.
How I tested it
Reconciled historical source-and-destination rows, distinguished direct target links from generic redirects, compared two crawl inlink exports, and preserved unknown crawl dates and incomplete historical redirect evidence.
Hypothesis: valuable inlinks were removedReceipts: four examined changes pointed to the targetOutcome: direct-removal explanation not confirmed
Key investigative finding
The examined exact-destination change-log rows documented links directed to the target page—not removals from it. Differences between crawl exports could not be interpreted as losses without reliable crawl dates and comparable scope.
Decision & status
No unsupported rollback. The investigation stayed open for historical redirect confirmation and other possible causes. The value here is ruling out an attractive but unproven explanation.
04 / PERSISTENT CRAWL & INDEXING DISCREPANCY
When a page works—but its search record stays stale
A three-page investigation across retrieval, rendering, indexing and deployment history.
The problem
Several service pages loaded and rendered successfully in current tests, yet Google’s stored crawl and indexing records remained old or excluded. The pages did not all share the same indexing status.
Investigation method
Checked current URLs and redirects, robots and canonicals, sitemap inclusion, rendered HTML, WordPress duplicates, caching, staging exposure, and dated Google inspections. Kept current retrieval separate from historical index records.
Current public retrieval: successfulStored Google records: old / excludedRoot cause: not yet established
What was established
Tested pages served their intended content and survived rendering. One page was indexed with an old reported crawl; two were excluded with older crawl records. Other pages in the same area were being crawled.
Next decision
Avoid another speculative rewrite or URL change. Complete remaining account-level checks and use the prepared evidence for an indexing escalation. The report was drafted, not recorded as submitted.
ADDITIONAL INVESTIGATIONS
The same discipline across different problems.
Short examples from the broader technical investigation program.
ENVIRONMENT & SECURITY QA
Public staging access
Investigated staging content visibility, noindex directives and historical crawl evidence. Distinguished public accessibility from permission to appear in search.
ALGORITHM & EVENT MONITORING
Update-impact watch
Maintained a watchlist around a reported search update, without assuming that timing alone proved causation. Historical records and comparative cohorts remained pending.