Startup Runway Estimator

This project was built as part of the Data-Driven VC Hackathon organized by Red River West & Bivwak! by BNP Paribas

Description

The Startup Runway Estimator helps venture capitalists (VCs) project a startup’s financial runway by analyzing revenue and costs month by month. This tool leverages data from external sources, including People Data Labs, SimilarWeb, and Harmonic, combined with salary benchmarks, to provide actionable insights into a startup’s financial health and projected time until cash depletion.

---

Prerequisites

Environment

• Python 3.9+
• Docker (for MongoDB and Mongo Express)

API Keys

To run the project, you’ll need API keys for the following services:

• People Data Labs (PDL): To fetch headcount data.
• Harmonic: To filter and select target companies.
• GPT API: To estimate revenue using website traffic and employee data.

Libraries

Install the required Python libraries by running:

pip install -r requirements.txt

---

Setup

1. Clone the Repository

git clone https://github.com/YOUR_REPO_URL
cd YOUR_PROJECT_NAME

2. Set Up the Environment

Create a .env file in the project root and include the following variables:

HARMONIC_API_KEY=your_harmonic_api_key
PDL_API_KEY=your_pdl_api_key
GPT_API_KEY=your_gpt_api_key

3. Start the Database

Use Docker to start the MongoDB and Mongo Express services:

docker-compose up

This will:

• Start MongoDB on localhost:27017
• FastAPI on localhost:8000

4. Populate the Database

Load initial data from Harmonic by clicking the "synchronise database" button on the application:

python main.py

This fetches company data from Harmonic and populates the MongoDB database.

---

Usage

Running the API

Start the FastAPI server:

uvicorn backend:app --reload

• API Endpoints:
  • GET /: Returns a welcome message.
  • POST /data: Placeholder for future data endpoints.

Calculating Runway

The main.py script calculates and updates each company’s Estimated Time of Death (ETOD) in the MongoDB database. It uses:

• Revenue estimates from the GPT API (using website traffic and employee data).
• Cost projections based on salary benchmarks and headcount data.
• Funding data from Harmonic.

---

Customization

Modifying Salary Benchmarks

The salary benchmarks used in the calculation can be updated in get_monthly_salaries.py:

salaries = {
    "engineering": 4160,
    "sales": 4000,
    "marketing": 3660,
    # Add or modify as needed
}

Adjusting Multipliers

Adjust salary ratios for roles and seniority in the same file:

salary_ratios = {
    "entry": 0.7,
    "senior": 1.3,
    "manager": 1.5,
    # Add or modify as needed
}

GPT Prompt Adjustments

Update the prompts used for revenue calculation in the retrieve_sw function within main.py.

---

Limitations

• Revenue Estimation: The model relies on GPT and website traffic data, which may not accurately reflect B2B SaaS revenue.
• Assumptions: Salary data and multipliers are based on benchmarks for France and may not generalize globally.
• Data Sources: SimilarWeb data relevance varies, especially for B2C startups.

---

License

This project is licensed under the MIT License. See the LICENSE file for details.

---

Contributing

Contributions are welcome! Please open an issue or submit a pull request for bugs, improvements, or new features.