MVRX1
We're building a marketing tool, that enables vibe coders who put out projects to assess the effectiveness of their influencer campaigns.
Core trick
Runs GPT-4o mini over sampled video frames for product detection, then combines them with transcript and retention data into a score.
Description
We're building a marketing tool, that enables vibe coders who put out projects to assess the effectiveness of their influencer campaigns.
README excerpt
# Product Placement Analysis Tool
A comprehensive toolkit for analyzing product placement effectiveness in YouTube videos.
## Features
- **Visual Analysis**: Frame-by-frame product detection using GPT-4o mini
- **Transcript Analysis**: Product mention detection and sentiment analysis
- **YouTube Analytics**: Retention data and demographic analysis
- **Effectiveness Scoring**: Multi-component scoring system with correlations
- **Standardized Outputs**: JSON and CSV outputs for further analysis
## Installation
```bash
# Clone the repository
git clone <repository-url>
cd product-placement-analysis
# Create and activate a virtual environment (recommended)
python -m venv .venv
source .venv/bin/activate # On Windows: .venv\Scripts\activate
# Install dependencies
pip install -r requirements.txt
```
## Environment Setup
Create a `.env` file with your API keys:
```
OPENAI_API_KEY=your_openai_api_key
```
You'll also need a Google Auth Client Secret file for YouTube API access.
## Usage
### Quick Start
```bash
# Run complete product placement analysis
python run_complete_analysis.py
```
### Custom Analysis
```python
from data_collectors import collect_all_data
from effectiveness_analyzer import analyze_effectiveness
# Define your product configuration
product_config = {
"product_name": "red glasses",
"product_type": "accessories",
"product_variants": ["red glasses", "red sunglasses", "red specs"],
"visual_attributes": {
"colors": ["red"],