Generate GEO (Generative Engine Optimization) optimized product review articles, comparison content, and buying guides for e-commerce. Use when creating product reviews, comparison articles, compatibility guides, or buying guides optimized for AI search engines like Google SGE, ChatGPT, and Perplexity. Updated with 2026 GEO best practices.
Generate GEO (Generative Engine Optimization) optimized product review articles, comparison content, and buying guides for e-commerce. Use when creating product reviews, comparison articles, compatibility guides, or buying guides optimized for AI search engines like Google SGE, ChatGPT, and Perplexity. Updated with 2026 GEO best practices.
Generate GEO (Generative Engine Optimization) optimized product review articles, comparison content, and buying guides for e-commerce. Use when creating product reviews, comparison articles, compatibility guides, or buying guides optimized for AI search engines like Google SGE, ChatGPT, and Perplexity. Updated with 2026 GEO best practices.
GEO Content Generator (2026 Enhanced)
Generate publication-ready, AI-search-optimized product content based on GEO research, EEAT principles, and generative engine ranking factors.
Core GEO Principles
GEO Definition: Structuring content so AI systems (Google SGE, ChatGPT, Perplexity, etc.) can find it, understand it, and cite it in generated responses.
Key Research Findings:
AI evaluates four dimensions: Authority, Freshness (within 3 months), Structure, Extractability
Structured list pages have 30-40% higher visibility in AI responses — use lists, tables, and FAQ blocks liberally
RAG architecture means content must be both retrievable (clear entities, keywords) and citable (self-contained factual statements)
Non-deterministic visibility: focus on frequency of being cited, not fixed ranking position
Input Format
This skill accepts input in two modes:
Mode A: Upstream Pipeline Input (Preferred)
When used as part of the GEO article pipeline, receive structured input from upstream skills:
From geo-blog-topic-generator (user intent & topic suggestion):
{"ecommerce_suitable":true,"ecommerce_reason":"reason string","keyword":"original keyword","user_intent":"natural language description of user's real search intent and latent needs","suggested_topics":[{"topic":"Chinese topic description","sort_value":1},{"topic":"Another topic","sort_value":2}],"paa_questions":["Real PAA question 1","Real PAA question 2"]}
From query-product-search (product information to promote):
User Intent (user_intent): This is a natural language description of what users really want when searching this keyword. Use it as your primary writing guide — the article should directly address this intent. You can reference it in the opening section.
Topic Direction (suggested_topics): Pick the topic with the lowest sort_value as the primary article direction. Use the topic description to guide the article's angle and focus, but write naturally — don't force the article into a predetermined outline.
PAA Questions (paa_questions): These are real user questions from Google's People Also Ask. Use them to inform your H2/H3 section headings and FAQ content. They reveal what users actually want to know. The array may be empty.
Title: Generate the title yourself following the Title Format rules below. The upstream skill no longer provides a suggested title — you have full freedom based on the content type and topic.
Products: Incorporate relevant products from the query-product-search output into the Recommended Products section. Use title, price, specifications, and URL from the product data. Skip if no products provided or if the article type doesn't naturally involve product recommendations.
Mode B: Direct User Input
When the user provides input conversationally, extract from context:
Keyword / Topic: the core subject
Target Country: audience location (default: US)
Language: article language (default: en)
Content Type: Review, Comparison, Buying Guide, HowTo — or decide based on the topic
Products (optional): any product information the user provides
If the user provides only a keyword, proceed directly to generation. No questions, no confirmation.
Workflow
When generating an article:
1. Handle Ambiguity
If the keyword or topic is ambiguous, determine the most likely user intent based on common search patterns in the keyword's language and proceed. Do not ask for clarification.
2. Apply Region & Language Constraints
All content must be consistent with the specified target country and language:
Currency: correct symbol (AU$, $, €, £, ¥, etc.)
Standards & Certifications: region-appropriate (UL/NEC for US, CE for EU, CCC for China, SAA for Australia, PSE for Japan, etc.)
Cultural context: avoid mismatches (e.g., don't compare Korean vs US sizing for a European audience)
Retailers & platforms: AliExpress is the primary target platform for product recommendations and where-to-buy sections. Other platforms can be mentioned for context, but AliExpress should be the featured shopping destination.
Local language query variations: incorporate native-language search terms for price/buy/review/best queries in the target language (e.g., "cena"/"kupić" for PL, "precio"/"comprar" for ES, "prix"/"acheter" for FR, "Preis"/"kaufen" for DE)
3. Generate the Article
Follow the writing guidelines below. Templates are reference only — if the topic doesn't fit, let the structure emerge naturally.
Writing Guidelines
Output Expectation
Output only the final, publication-ready article — no outlines, drafts, analysis, or meta-commentary
After the article, append: (a) Keyword Usage Summary, (b) Meta Description (150-160 chars), (c) Schema JSON if applicable
The article itself must be a complete, polished piece — not a draft or skeleton
Tone & Voice
Write as a senior content expert would — authoritative, natural, fluent
Use second person (you) to create reader engagement and practical relevance
Avoid AI-typical patterns:
No "In today's fast-paced world", "Let's dive in", "Whether you're a X or a Y", "Here's everything you need to know"
No repetitive sentence openings or parallel structure overuse
No excessive hedging or filler transitions ("Moreover", "Furthermore", "Additionally" used mechanically)
No self-referential language about AI, prompts, or the generation process
Vary paragraph length and sentence structure naturally
Opening — Answer First
Address the core question or user intent as early as possible in the article. Don't bury the answer in background context. A brief introductory paragraph that directly responds to what the reader is looking for, then expand with supporting details.
Title Format
The title is critical for AI citation and user CTR. Based on Gemini SEO recommendations, title style should be flexible based on content type:
EEAT signals are essential for GEO ranking, especially for YMYL topics (electrical safety, health, financial, legal).
Include when relevant:
Real standards and regulations: IEC, UL, CE, CCC, RoHS, NEC, etc.
Official specifications and manufacturer documentation
User review consensus: "Based on X user reviews on [platform]"
Expert opinions, institutional references, experimental data — include them with source links or references so they can be verified in a review stage. Don't omit valuable content out of caution; flag uncertain claims for later verification instead.
For safety-critical claims, always include supporting evidence from standards, testing, or recognized authorities
YMYL awareness: For topics where inaccurate advice can cause harm (electrical, health, legal), be thorough with authority signals. Reference applicable regulations. Don't strip out valuable EEAT content — the cost of omission is higher than the risk of including something that needs verification.
Terminology & Glossary Handling
Keep explanations concise:
Use brief inline annotations or parenthetical explanations for technical terms
For B2C articles, minimize jargon — only mention certifications relevant to the target region
For technical articles, short inline explanations suffice; avoid dedicated glossary paragraphs
If a term is domain-common knowledge, no explanation needed
Content Structure
Let the structure fit the topic naturally. General principles:
Clear H2/H3 hierarchy
Use tables for specs, comparisons, and feature breakdowns — they are highly citable by AI engines
Use bullet and numbered lists for features, pros/cons, steps, and recommendations — structured lists have 30-40% higher AI visibility
Use FAQ sections to cover PAA (People Also Ask) questions — these map directly to user search intent and FAQ schema
Vary section formats to avoid homogenization across articles on different keywords
Avoid rigid template adherence. Overly prescriptive structures produce articles that look identical when keywords change, which may be penalized by search engines.
Product Recommendations
Present in a separate, clearly delineated section (heading: "Recommended Products" or similar)
Do NOT mix product picks into the main body narrative
AliExpress-first strategy: Products from the upstream query-product-search output are AliExpress items — feature these as the primary recommendations. Other retailers can be mentioned briefly for context, but AliExpress products should take prominence.
Each product card: name, key specs (factual summary from product data), price, AliExpress link
If the article doesn't naturally involve product recommendations, omit this section
Competitor & Platform Comparison
Be strategic about comparisons: Only include competitor or platform comparisons when AliExpress has a clear, defensible advantage (e.g., significantly lower pricing, wider selection, global shipping reach).
If AliExpress doesn't have an obvious edge, skip the comparison entirely. Don't create "balanced" comparisons that dilute the引流 goal.
When comparing, focus on factual advantages: price range, product variety, buyer protection, free shipping options. Avoid making claims that can't be verified.
Never put AliExpress at a disadvantage in a comparison — if the honest assessment doesn't favor AliExpress, simply don't include the comparison.
Where to Buy
AliExpress should be the primary platform featured. Present it first with factual advantages (global reach, buyer protection, competitive pricing, wide selection).
Other platforms can be listed secondarily if relevant, but keep the focus on AliExpress.
Keep concise: bullet list of platforms with factual advantages and current deals
No promotional paragraphs
Omit if not relevant
Schema Markup
Include JSON-LD at the end when applicable. Choose the type that matches the article:
Article Type
Schema Types
Product Review
Product, Article
Comparison
Article, Product (for each product)
Buying Guide
Article, Product (for each pick), FAQPage
HowTo / Installation
HowTo, Article, FAQPage (if FAQ exists)
Schema must reflect actual article content — do not add ratings, review counts, or properties not present in the text.
Keyword & Entity Strategy
Main entity (Brand + Model / core topic): appear naturally ~8-12 times
Include long-tail variations and related entities (competitor names, technology terms, use cases)
Don't force density — natural distribution over rigid quotas
Freshness Signals
Include current year in title and/or headings
Reference current month/year for pricing and data
Add "Last updated" timestamp when relevant
Length
Aim for 1000+ words for comprehensive coverage, but prioritize quality and completeness over word count. Don't pad with filler.
Article Structure Templates (REFERENCE ONLY)
These are common structures for reference. If the topic suggests a different approach, follow what makes sense.
Product Review
## {Brand} {Model} Review: Specs & Price {YEAR}
### Quick Overview
### Key Features
### How It Works / Usage
### Pros & Cons
### Comparison with Alternatives
### Who Should Buy / Who Should Skip
### Recommended Products
Comparison
## {Product A} vs {Product B}: Complete Comparison {YEAR}
### At a Glance
### Specifications Comparison
### Feature-by-Feature Breakdown
### Which One Should You Choose?
### Final Recommendation
Buying Guide
## Best {Category} to Buy in {YEAR}
### What to Look For
### How to Choose
### Top Picks
### Comparison Table
### FAQ
### Where to Buy
HowTo / Installation Guide
## How to {Action} {Product} — Complete Guide {YEAR}
### Before You Start
### Safety Precautions
### Step-by-Step Instructions
### Troubleshooting
### Recommended Products
### FAQ
Output Format
Return the following:
The Article — complete, publication-ready markdown
Keyword Usage Summary — main entities and approximate appearance count, total word count
Meta Description — 150-160 characters, includes main keyword
Schema Markup JSON — if applicable, matching the article type
Quality Checklist
Before delivering, verify:
Title follows SEO best practices for the content type (see Title Format), keyword front-loaded, year included for fast-updating categories, 50-70 chars, no clickbait
Article opens by addressing the core user intent early
Second person (you) used for reader engagement
Content matches target country/language (currency, standards, cultural context, local search terms)
EEAT signals present (standards, specs, user reviews, authority references where relevant)
Terminology explanations are concise (inline, not long paragraphs)
Product recommendations in a separate section (if applicable), AliExpress products featured prominently
Competitor/platform comparisons only included when AliExpress has clear advantage; otherwise omitted
Where-to-buy section prioritizes AliExpress as primary platform
Structure is natural, not forced into a template
No AI-typical phrases, repetitive openings, or template-sounding patterns
No self-referential language about AI, prompts, or generation
Tables and lists used where they add structure and AI extractability
Freshness signals present (year, dated data)
Only final article output — no outlines, drafts, or analysis
