Data Maturity — What You Can Run When
Not all sites have the same data available. A brand-new site has no GSC performance history. A site mid-migration has incomplete data. Use this table to decide what’s appropriate before starting.
Site stages
| Stage | GSC data available | GA4 data available | What to run |
|---|---|---|---|
| New site / just launched (0–4 weeks) | Coverage only (indexing status, 404s) | None or unreliable | Squirrel Scan + PageSpeed only. Fix code-level issues. Hold skills. |
| Post-migration (4–8 weeks) | Early performance data (low volume, noisy) | Partial | Squirrel Scan + PageSpeed. Run /schema-markup-better-than-corey-haines, /site-architecture, /programmatic-seo (strategy-level). Hold /seo-audit. |
| Establishing (2–4 months) | Meaningful performance data (queries, CTR, impressions by page) | Reliable channel + landing page data | Full Phase 1 data collection + all 6 skills. This is the minimum for a useful /seo-audit run. |
| Mature site (4+ months, no migration) | 12-month history available | Full funnel data | Full audit. Year-over-year comparisons possible. /seo-audit most specific here. |
Which skills need which data
| Skill | Minimum data required | Can run without GSC? |
|---|---|---|
| Squirrel Scan | None — code-level crawl | ✅ Always run first |
| PageSpeed Insights | None | ✅ Always run |
/programmatic-seo | Site structure + business context | ✅ Yes — strategy output is data-independent |
/site-architecture | Existing page list + content structure | ✅ Yes |
/schema-markup-better-than-corey-haines | Squirrel Scan output | ✅ Yes |
/competitors | Competitor names (ask client — don’t guess) | ✅ Yes, once competitors confirmed |
/ai-seo-better-than-corey-haines | Top commercial queries helpful but not required | ⚠️ Generic without query data |
/seo-audit | 4+ weeks GSC performance data (queries + pages CSV) | ❌ Output is generic without it |
| Remove AI slop without removing CTAs and good marketing copy | Identified AI-written page text, links, and conversion goal | ✅ Yes — review audit copies, even on a new site |
| trend-scan | GSC Dates CSV for direction; Pages/Queries CSVs for concentration; comparable prior/recent exports for movers | ❌ Record unavailable or insufficient history rather than inventing a trend |
| claim-check | A draft report or other text containing factual claims; source access for verification | ✅ Yes — does not require traffic history |
The three public playbook helpers have different data needs from the core analysis skills. Holding traffic-dependent analysis on a new site does not prevent editorial review or claim checking. During an audit, record recommended fixes and work on evidence copies; implementation follows the separate action runbook.
What’s always actionable regardless of data age
Squirrel Scan findings are code-level — they don’t depend on traffic history. Fix these immediately on any site at any stage:
- Schema validation errors (e.g.
Organization.logotype mismatch) - Missing H1 tags
- Orphan pages (internal linking gaps)
- Oversized images
- Missing or duplicate meta descriptions
- Missing alt text
- Security headers (CSP, X-Frame-Options)
- E-E-A-T gaps (no Contact page, no Privacy Policy)
These should be fixed before Google does more recrawling — not after you’ve collected more data.