19 lines
2.8 KiB
Plaintext
19 lines
2.8 KiB
Plaintext
GOAL - W13 Deepen 10 vs-competitor / "best/review" pages for AI-search citations (odoo-expertos). Branch: fix/w13-ai-search.
|
|
Repo: C:\Users\eugen\.cursor\projects\Odoo Expertos\odoo-expertos. Work ONLY in this repo.
|
|
Context: AI engines (ChatGPT, Perplexity, Google AIO, Bing Copilot) most reliably cite "X vs Y" and "best of" pages with direct-answer passages, structured data, comparison tables, and verifiable sources. You ALREADY have 13 such slugs (odoo-vs-sap-hosting-2026, odoo-vs-zoho-books-hosting-2025, cloudpepper-vs-odoo-sh-2026, alternativas-odoo-sh-2026, odoo-community-vs-enterprise-hosting-2026, odoo-gratis-vs-pago-diferencias-2025, comparativa-precios-odoo-2026, mejor-hosting-odoo-2026, mejores-proveedores-hosting-odoo-2025, mejores-proveedores-hosting-odoo-2026, opiniones-hosting-odoo-2026, odoo4projects-hosting-review-2026, alternativas-odoo-ia-comparativas-2025). Pick the top 10.
|
|
TASK PER PAGE: apply the /blog geo skill principles (read ~/.claude/skills/blog-geo/SKILL.md):
|
|
1. Open the page in opening with a 60-150 word "direct answer" paragraph that an LLM can quote verbatim (lead with the answer, then the qualifier).
|
|
2. Each H2 section starts with a 40-80 word direct-answer passage before any detail.
|
|
3. Add or expand FAQPage JSON-LD to >=8 high-intent Q&A reflecting common AI prompts (e.g., "Cual es mejor: X o Y para empresa pequena?").
|
|
4. Add an HTML comparison <table> with rows for: pricing, hosting requirements, learning curve, scalability ceiling, TCO 5-year, ideal company size. Every numeric cell must include a citation marker linking to the source (footnote or inline <a>).
|
|
5. Replace fabricated "Dr. Maria Elena Rodriguez" byline + Article JSON-LD author with "Equipo Odoo Expertos" (or a real author if one exists).
|
|
6. Add Review/AggregateRating JSON-LD ONLY if real reviews exist on the page (no fabrication).
|
|
CONSTRAINTS: only edit body + JSON-LD on the 10 selected pages. Preserve design, D3.js refs, hreflang, canonical, nav.
|
|
PROVE each turn: run powershell -File seo-audit-2026-05-27\REVIEW.ps1 -Only W13 and paste full output.
|
|
DONE WHEN: that output contains "W13: PASS" (10 target pages each have: >=8 FAQ items, comparison <table>, direct-answer opening, no fabricated author) AND committed on branch fix/w13-ai-search (show git log -1 --oneline). Stop after 30 turns if blocked.
|
|
|
|
MARKER (required for REVIEW): you may add EXACTLY ONE meta tag to each updated page's <head>: <meta name="content-version" content="W13-ai-search">. Do not change any other <head> tags. REVIEW.ps1 -Only W13 counts pages with this marker.
|
|
|
|
|
|
CONCRETE TARGETS: read seo-audit-2026-05-27\OPPORTUNITIES-2026-05-28.md section 1 (top 10 AI-search pages, ranked with current weakness + concrete fix) and section 4 (cluster grouping). Apply both: deepen the 10 pages with the per-page fixes; add the pillar/spoke internal links per the cluster structure.
|