Skip to content

SEO with AI

When to Use

Use AI-powered SEO automation to generate optimized meta tags, descriptions, structured data, and content that ranks well in both traditional search engines and AI-powered discovery tools (ChatGPT, Perplexity, Google SGE).

Decision

Approach Best For Automation Level Human Review
AI SEO Analyzer Analysis and recommendations Report generation Review required
Metatag AI Automator Automatic generation on save Fully automated Optional review
Metatag AI Module Generate with button in UI On-demand Required before save

Implementation Patterns

Method 1: AI SEO Analyzer (analysis and recommendations)

// Configuration at /admin/config/ai/seo
// 1. Select provider and model
// 2. Configure analysis criteria
// 3. Set permissions for SEO Analyzer tab

// Usage: Navigate to node view, click "SEO Analyzer" tab
// Generates report with:
// - Keyword density analysis
// - Readability scores
// - Meta tag recommendations
// - Content structure review
// - Improvement suggestions

Method 2: Metatag AI Automator (automatic generation)

# Configure on content type
automator_type: llm_metatag
field: metatags
trigger: [create, update]

prompt: |
  Generate SEO-optimized meta tags:

  Content: [node:title] - [node:body:summary]
  Target audience: [node:field_audience]
  Primary keyword: [node:field_keyword]

  Requirements:
  - Title: 50-60 characters, include {{keyword}}
  - Description: 150-160 characters, compelling, include {{keyword}}
  - Open Graph title: Engaging, social-optimized
  - Open Graph description: 2-3 sentences, shareable
  - Twitter Card: Optimized for Twitter

  Output as JSON:
  {
    "title": "...",
    "description": "...",
    "og_title": "...",
    "og_description": "...",
    "twitter_title": "...",
    "twitter_description": "..."
  }

E-E-A-T Implementation

E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) is critical for search ranking and AI discovery:

Experience Signals: 

# Automator prompt including experience
prompt: |
  Write content incorporating first-hand experience:

  Topic: [node:title]
  Author expertise: [node:uid:entity:field_bio]

  Include:
  - Specific examples from real usage
  - Concrete outcomes and results
  - Challenges encountered and solutions
  - Timeline and context

  Avoid: Generic advice, theoretical knowledge only

Expertise Indicators: - Author bylines with credentials - Citations and references to authoritative sources - Technical depth appropriate to topic - Industry-specific terminology used correctly

Authoritativeness: - Link to recognized industry sources - Include expert quotes and interviews - Reference original research or data - Display author credentials and affiliations

Trustworthiness: - Fact-check AI-generated claims before publishing - Add publication date and last updated date - Include sources and citations - Display editorial policies and review process

Structured Data Automation

Generate schema.org structured data with AI:

automator_type: llm_json_native_binary
field: field_structured_data
prompt: |
  Generate schema.org JSON-LD for this article:

  Title: [node:title]
  Author: [node:uid:entity:name]
  Published: [node:created:custom:c]
  Content: [node:body:summary]

  Schema types to include:
  - Article (headline, author, datePublished, description)
  - Person (author details)
  - Organization (publisher)

  Output valid JSON-LD following schema.org standards.

output_format: json_ld
validation: schema_org

SEO Best Practices

Title Tags: - 50-60 characters optimal length - Include primary keyword near beginning - Make unique per page - Match user search intent - Avoid keyword stuffing

Meta Descriptions: - 150-160 characters (Google's typical display limit) - Include primary keyword naturally - Write compelling copy to drive clicks - Unique per page, not duplicated - Call-to-action when appropriate

Open Graph & Twitter Cards: - Optimize for social sharing, not just search - Use more engaging, conversational tone than meta description - Test preview rendering in social platforms - Include relevant hashtags in Twitter description

AI Discovery Optimization: - Structure content with clear sections and headings - Use Q&A format when appropriate (FAQ schema) - Provide direct, factual answers near top of content - Include relevant context and definitions - Cite sources that AI tools can reference

Common Mistakes

  • Wrong: Generating identical metatags for all content → Right: Every page needs unique tags; configure prompts to use page-specific tokens
  • Wrong: Exceeding character limits → Right: Search engines truncate long tags; enforce strict length limits in prompts
  • Wrong: Keyword stuffing in AI prompts → Right: Modern SEO and AI penalize over-optimization; use natural language
  • Wrong: Not reviewing AI-generated tags before publishing → Right: AI may miss context or nuance; human review is essential for brand consistency
  • Wrong: Ignoring social media optimization → Right: Open Graph and Twitter Cards drive social traffic; don't auto-generate from meta description
  • Wrong: Forgetting to update metatags when content changes → Right: Set automators to trigger on update, not just create
  • Wrong: Not tracking SEO performance of AI-generated content → Right: Monitor rankings and adjust prompts based on performance data

See Also