Brand Competitor Analysis Platform

Overview

This project is a full-stack web application for brand competitor research, data enrichment, social media scraping, and AI-powered email generation. It helps marketing, sales, and research teams quickly:

• Identify and summarize direct competitors across multiple channels.
• Enrich competitor data with URLs, metadata, and social media metrics.
• Scrape competitor social media profiles (Instagram, LinkedIn, TikTok, Twitter, YouTube).
• Generate personalized outreach emails using a variety of AI models.
• Manage search history and edit data via an intuitive UI.
• Switch between light/dark theme and customize email templates.

Key Features

• Brand Competitor Search
  Enter a brand name and region to research top competitors via GPT-4o-mini and GPT-4o.

• Competitor Data Enrichment

  • Edit/add competitor info (brand, region, platform, URL, etc.) in a table.
  • AI-powered URL inference and content completion.

• Social Media Scraping
  Automatically scrape competitor profiles and followers from:

  • Instagram
  • LinkedIn
  • TikTok
  • Twitter
  • YouTube
    (powered by Apify)

• History & Data Management
  View, search, and restore previous competitor analysis sessions.

• AI-Powered Email Generation

  • Support for Anthropic Claude, OpenAI GPT-4o, Google GPT, Perplexity.
  • Customizable templates with JSON data injection.
  • Live email editor with preview and light/dark mode.

• Theme Toggle
  Switch between light and dark UI themes on the fly.

• Multi-Model AI Integration

  • Anthropic Claude
  • OpenAI GPT-4o
  • Google GPT
  • Perplexity

Tech Stack

• Frontend: React (Next.js 14), Tailwind CSS 4
• Backend: Next.js API Routes, Supabase (PostgreSQL engine)
• Scraping: Apify client (@apify/client)
• AI/ML Integration:
  • Large Language Models: OpenAI GPT-4o, Anthropic Claude, Google GPT, Perplexity
  • Agents & SDKs: @anthropic-ai/sdk, @apify/client, perplexity-api
• Utilities: SWR, React Hot Toast, React Markdown
• Tooling: ESLint, Prettier, PostCSS, Autoprefixer

Code Statistics

• Total Lines of Code: 16,226 lines
• Application Code: ~8,787 lines (excluding package-lock.json)
• Main Components:
  • Frontend Pages: ~3,500 lines
  • API Routes: ~2,000 lines
  • Library & Utilities: ~500 lines
  • Configuration: ~7,500 lines (mostly dependencies)
• Version: 1.7.8

Quick Start

1. Clone the repository

   git clone <your-repo-url>
   cd brand-research

2. Install dependencies

   npm install

3. Configure environment variables
   Copy .env.example to .env.local and fill in:

   NEXT_PUBLIC_SUPABASE_URL=your_supabase_url
   NEXT_PUBLIC_SUPABASE_ANON_KEY=your_supabase_anon_key
   ANTHROPIC_API_KEY=your_anthropic_key
   OPENAI_API_KEY=your_openai_key
   GOOGLE_GPT_API_KEY=your_google_gpt_key
   PERPLEXITY_API_KEY=your_perplexity_key
   APIFY_TOKEN=your_apify_token

4. Run the development server

   npm run dev

5. Open your browser
   Visit http://localhost:3000

Folder Structure

app/ api/ anthropic/ openai/ google-gpt/ perplexity/ apify/ instagram/ linkedin/ tiktok/ twitter/ youtube/ competitor-history/ simple-mode/ history/ simple-mode/ competitor-scrape/ competitor-result/ email-editor/ ThemeToggle.tsx layout.tsx globals.css lib/ prompts.ts supabase.ts supabase-utils.ts apify-client.ts public/ favicon.ico ... README.md

Pages

• / (Home)

  • Brand competitor search entry point.
  • Input brand name and region to run deep research via GPT-4o-mini and GPT-4o.
  • Edit/add/delete competitors, save and navigate to the result page.

• /history (Search History)

  • View and group all past search records.
  • Paginate, expand/collapse each search session to see detailed tables.

• /simple-mode (Simple Mode)

  • One-click simplified workflow.
  • Step-by-step: create search session, analyze competitors, extract URLs, scrape followers, generate email.
  • Built-in progress bar and debug logs.

• /competitor-scrape (Competitor Scrape)

  • Batch scrape social media data for selected records (Instagram, LinkedIn, TikTok, Twitter, YouTube).
  • Edit mode to modify fields and save changes back to the database.

• /competitor-result (Competitor Result)

  • Display competitor search history and expand "all platform" rows into multiple platform entries.
  • Auto-fill URLs via GPT prompts, edit, and save the final data.

• /email-editor (Email Editor)

  • Select or customize email templates based on historical competitor data.
  • Insert variables, attach JSON data, and preview in Markdown.
  • Generate final email content with Anthropic Claude and view past results.

Screenshots

License

MIT