AirGarageSEO index coverageTargeted diagnosis
No broad technical blocker appeared in the reviewed sample
Wave 1 compared not-indexed URLs with indexed pages. Phase 9 then tested family, URL-shape, text-similarity, discovery, and crawl-timing ideas without claiming Google’s cause.
120Wave 1 requests
Combined result
No obvious current technical difference appeared in the limited Wave 1 comparison.
The deeper analysis found actionable inventory and architecture patterns, but neither wave reveals Google’s causal reason for individual index states.
How Wave 1 checked
A small, controlled comparison
The 78 reviewed exclusions reconcile to 73 nonterminal comparison pages plus five historical redirect/404 exceptions. They were selected before results were checked; 40 indexed pages served as comparisons. The 118 page checks plus robots and sitemap checks were sequential, did not follow redirects, and were not retried.
78excluded pages reviewed in total
73nonterminal comparison pages
5historical redirect / 404 exceptions
40indexed comparison pages
Nonterminal not-indexed comparison cohort
72/ 73
returned a current initial HTTP 200 response; Hempstead was the one current redirect in this 73-page cohort
Indexed comparison pages
40/ 40
returned a current initial HTTP 200 response
Shared signals
Current-200 pages looked alike on six basic checks
These direct observations narrow the likely problem space but do not establish content equivalence or Google’s reasoning.
| Signal | Nonterminal not-indexed cohort | Indexed comparison pages |
|---|---|---|
| Initial HTTP response | 72 of 73 returned 200 | 40 of 40 returned 200 |
| Canonical | 72 of 72 self-canonical | 40 of 40 self-canonical |
| Robots directives | No blocking directives observed | No blocking directives observed |
| Sitemap | 72 of 72 present | 40 of 40 present |
| H1 | 72 of 72 had one H1 | 40 of 40 had one H1 |
| Public robots response | Allowed by observed response | Allowed by observed response |
HTML summary measures
Descriptive medians, not quality scores
Server-returned HTML only—not browser-rendered pages. Overlapping medians are a null observation, not proof that content is equivalent.
What Phase 9 strengthened
Family is a useful planning axis—but not an isolated cause
Across 922 exact city slugs shared by all three families, indexed share differed consistently.
Interpret carefully. Matching on city does not isolate template, intent, launch history, internal links, or business value. Make separate retain/consolidate/retire decisions by family rather than applying one generic city-page fix.
Text similarity
Not evidence of thin or duplicate content
Sampled full-page server transport text was highly repetitive in several families, but it included navigation, footer, and boilerplate—and indexed controls were equally similar.
Do not mass-rewrite pages from this signal. Rendered, boilerplate-stripped main-content and user-purpose analysis would be required first.
Recorded crawl timing
Association with reverse causation risk
24.9days · indexed median
147.6days · crawled-not-indexed median
Older recorded crawls may be a consequence of continued exclusion. This snapshot does not establish a crawl-prioritization mechanism.
What this does—and does not—mean. The evidence does not support a blanket robots, canonical, sitemap, heading, content, or initial-response fix. Google does not expose a causal reason for discovered/crawled exclusions.