Competitive Visibility Audit

A Jupyter notebook that builds a competitive intelligence pipeline using Bright Data's Web Execution Layer APIs and OpenAI GPT for analysis.

What It Does

This notebook performs a comprehensive competitive audit across 5 sections:

1. SERP Ranking - Searches Google for industry keywords, finds your position vs competitors, and identifies top competitors
2. Deep Dive - Scrapes competitor homepages as markdown and extracts value propositions, features, pricing, and trust signals using GPT
3. AI Perception - Queries ChatGPT, Perplexity, Grok, and Gemini to see if AI engines recommend your brand
4. Deep Competitor Analysis - Scrapes deeper pages (pricing, features) and generates consolidated company profiles with competitive comparison
5. Executive Summary - GPT-powered analysis compiling all findings into actionable insights

Prerequisites

• Bright Data Account with:
  • SERP API zone configured
  • Web Unlocker zone configured
  • API token
• OpenAI API Key for GPT-powered analysis

Setup

1. Click the "Open in Colab" badge above (or this link)

2. Add your API keys to Colab Secrets:

   • Click the 🔑 key icon in the left sidebar
   • Add each secret and toggle "Notebook access" on:

   Secret Name	Description
   BRIGHTDATA_API_TOKEN	Your Bright Data API token
   BRIGHTDATA_ZONE_SERP	SERP API zone name (e.g., serp_api1)
   BRIGHTDATA_ZONE_UNLOCKER	Web Unlocker zone name (e.g., unlocker)
   OPENAI_API_KEY	Your OpenAI API key

3. Run the notebook cells in order

Bright Data APIs Used

API	Purpose
SERP API	Google search results with structured JSON output
Web Unlocker	Scrape any website as clean markdown
Web Scraper API	Pre-built scrapers for AI engines (ChatGPT, Perplexity, Grok, Gemini)

Output

The notebook generates:

• SERP ranking analysis with competitor identification
• Structured insights from competitor homepages
• AI engine perception analysis (brand mention tracking)
• Consolidated company profiles with competitive comparison
• Executive report with strengths, weaknesses, and recommended actions

License

MIT