/seo-content-product-page

name: seo-content-product-page
description: "Spec, audit, competitor benchmark and build for e-commerce product pages (PDPs), so they rank in Google and get recommended by the AI assistants that now link products directly. Input: a product page URL or a product brief (Shopify, WooCommerce, Magento, BigCommerce, PrestaShop, Wix, Webflow or custom). Runs four phases: the PDP spec with stable IDs on top of the common page spec, an audit with seo_audit.py and section_audit.py --type product, a benchmark against the same product sold elsewhere, marketplace listings and the ranking brands with page_benchmark.py, then the build: unique benefits-first copy, spec and comparison tables, FAQ and definitions blocks, dated rating with count, real photos, price as HTML text, title, H1, slug and meta, out-of-stock and variant handling, Product, Offer and BreadcrumbList schema, the ChatGPT Shopping feed, and a scripted acceptance test. Use to write, rewrite or audit a product page, or when PDPs get no traffic or traffic without sales."
license: MIT
metadata:
  author: "Sorank (https://sorank.com)"
  version: "2.0.0"

Product Page SEO and GEO

A product page (PDP) has two jobs: win the transactional query ("buy X", "X price", the product name) in Google, and be the page an AI assistant quotes when it recommends where to buy. Most PDPs fail both the same way: the manufacturer description every reseller also publishes, supplier photos, a price and reviews injected by JavaScript, a rating buried at the bottom, and markup that contradicts the page. Nothing on the page is unique, so Google picks a stronger domain and the assistant fills the gaps from a competitor.

This skill runs the four phases of the page skill standard (`docs/page-skill-standard.md`): spec, audit, benchmark, build. Rules come from 115+ real agency audits (labeled field heuristics) combined with sourced 2026 practices (labeled measured or documented, with the source).

Company knowledge first (Obsidian)

If the working environment contains an Obsidian vault or any local knowledge base (a folder of .md notes, often with a .obsidian directory), read the relevant notes before acting: brand and product facts, target keywords, competitors, and the SEO action log of what was already tried. Ground every recommendation in that context instead of asking the user for facts the vault already holds. At the end of the session, append the actions taken to the vault's SEO action log so the next session starts informed.

When to use

• Writing or rewriting a product description or a full product page
• Auditing a PDP with no rankings, no traffic, or traffic without sales
• Getting products recommended by ChatGPT, Perplexity, Claude, Gemini or Google AI Overviews
• Handling out-of-stock, seasonal, discontinued or variant product URLs
• Reviewing PDP titles, H1s, slugs, meta descriptions, images, reviews, trust badges or Product schema
• Any storefront: Shopify, WooCommerce, Magento, BigCommerce, PrestaShop, Wix, Webflow Ecommerce, custom

| Case | Skill | |---|---| | Category, collection or listing pages (PLPs) | seo-content-collection-page | | Service or offer pages outside e-commerce | seo-content-service-page | | "X vs Y", alternatives and best-for pages | seo-content-comparison-page | | Supporting blog content for a product | seo-content-blog | | Keyword and intent mapping | seo-keyword-research | | Whole-site audit | seo-geo-audit | | Rendering, crawl access, AI crawler behavior, image weight | seo-technical | | JSON-LD implementation details | seo-schema-markup | | Passage-level citability writing | geo-visibility | | Measuring AI citations and referrals | geo-tracking |

Phase 1. The spec

Common requirements C-01 to C-27: skills/seo-geo-audit/references/common-page-spec.md. Every C row applies to a PDP; the rows below are the product-specific ones. "Verified by" is a `seo_audit.py` finding code, a `section_audit.py --type product` block, or "manual".

| ID | Requirement | Threshold | Verified by | Why | |---|---|---|---|---| | PDP-01 | Transactional target query | Product name, "buy X", "X price"; the live SERP shows product pages, not guides | manual (SERP check) | Google types result pages by intent; a PDP almost never ranks for an informational query whatever its quality (documented behavior) | | PDP-02 | Title tag pattern | {Product name} - {Category} \| {Brand}, with the attribute buyers search (material, size, color), unique, about 60 characters (C-01) | title_short, title_long, manual | Standard practice; Google rewrites titles it judges poor (Google title link docs) | | PDP-03 | H1 is the product name | Exactly one H1 | h1_missing, h1_multiple | Standard practice (C-03) | | PDP-04 | Descriptive slug | /products/{primary-keyword}, descriptive words, hyphens, no SKU codes or internal IDs | url_hygiene, manual | Measured, correlational: 89.78% of pages ChatGPT cites have descriptive slugs vs 81.11% in the comparison set (Ahrefs) | | PDP-05 | Handwritten meta description | Benefit + differentiator + one offer fact (shipping or returns), about 150-160 characters (C-02) | meta_missing, meta_short, meta_long, manual | Field heuristic (115+ audits): auto-generated metas repeat the first sentence; offer facts win the click | | PDP-06 | Unique description, benefits first | 150-200 words minimum, exists nowhere else on the web, never the manufacturer boilerplate; a floor, not a target to pad toward | thin_content, manual (exact-sentence search in quotes) | Field heuristic (115+ audits); Google clusters duplicates and shows one version, usually the strongest domain or the brand | | PDP-07 | Benefit bullets near the buy box | 3-5 one-line outcomes | manual | Skimmers decide here; assistants lift these lines verbatim | | PDP-08 | Spec table | Every measurable attribute, compatibility, materials, care, in an HTML table | section_audit spec_table, any_table | Extraction material for assistants and comparison shopping | | PDP-09 | Compari