visibly-seo-status-quo

SEO Status-Quo Analysis

Establish, with live data only, where a domain stands organically in a target market today. This is the foundation every later phase builds on — so every fact here must be verified, not assumed.

Tier note. Steps 1–2 use Visibly AI's Google tools (query_search_console, list_projects), which need a Visibly key (pro tier — GSC/GA run at 0 credits once Google is connected). No key? Skip to Step 3: export Search Console (Performance → Queries → CSV) yourself and feed that CSV to the Python template in Step 4 — the cross-reference, classification and quick-win logic all run locally and keyless. Setup tiers: docs/setup.md §3.

Step 1 — Discover what's wired

1. mcp__visiblyai__list_projects — find the project matching the domain.
2. mcp__visiblyai__get_google_connections — confirm Search Console (and GA) are connected.

If no project/connection exists, stop and tell the user what to connect first.

Step 2 — Pull live GSC performance

3. mcp__visiblyai__query_search_console — dimension=query, target-country filter, limit=500. This is the ground truth: clicks, impressions, CTR, average position.
4. Repeat with dimension=page to find URLs with impressions but weak clicks (underperforming pages = on-page/intent-mismatch candidates).

Step 3 — Load the client's target keywords

5. Read the client's keyword file (*.xlsx / *.csv) with pandas. These are the keywords the business cares about — often different from what it actually ranks for. That gap is where the strategy lives.

Step 4 — Cross-reference and classify

6. Map each target keyword onto the live GSC row (clicks, impressions, CTR, position).
7. Classify on two axes:
   • Type: Brand · Generic · Competitor
   • Ranking bucket: Top 3 · Page 1 (4-10) · Page 2 (11-20) · Weak (21-50) · Not Ranking (50+/none)

Don't do this join by hand. Save the GSC export (dimension=query) as CSV/XLSX next to the client's keyword file and run the Python template — it does the cross-reference, both classifications and the quick-win flag deterministically:

.\claude_tools_venv\Scripts\python.exe -m claude_tools.status_quo `
    --keywords "clients/<domain>/_knowledge/keywords.xlsx" `
    --gsc      "clients/<domain>/<date>_Status-Quo/gsc_query.csv" `
    --brand-terms <brand> --competitor-terms <comp1>,<comp2> `
    --out      "clients/<domain>/<date>_Status-Quo/status_quo_<date>.xlsx"

It writes a 3-sheet xlsx (Status-Quo · Quick wins · Summary) with the exact columns the output template expects. Columns are auto-detected (DE/EN headers); first-time setup: .\claude_tools\setup.ps1. See claude_tools/README.md.

Step 5 — Flag quick wins

8. A quick win = high impressions + CTR below ~5 % + position in the 4-15 "striking distance" band. These convert fastest: the demand and partial visibility already exist; you're closing a gap, not creating one. List them explicitly with the lever (title/meta, on-page, internal links).

Step 6 — Competitor benchmark

9. mcp__visiblyai__get_competitors and mcp__visiblyai__get_keywords to compare positions on shared keywords. Show where the client is out-ranked and by whom.

Output

Fill in the ready skeleton at templates/status-quo-template.md — it already has the visibility counts, classification split, cross-referenced keyword table, underperforming pages, quick wins and competitor benchmark. Save under clients/<domain>/YYYY-MM-DD_Status-Quo/:

• status_quo_<date>.md — summary (copied from the template)
• status_quo_<date>.xlsx — full cross-referenced table (all rows)
• quick_wins.md — prioritised, actionable (specific lever per keyword)

Only live-verified facts go into any deliverable. Slash command: /visibly-seo-status-quo <domain>. Methodology: docs/workflows.md.