This study leverages extensive datasets to explore the profitability and predictors of sales growth among businesses in Los Angeles. By segmenting the market, identifying optimal locations for new ventures, and predicting sales growth, the project aims to provide actionable insights for business owners in the LA area.
Given the high rate of new business failures, understanding and predicting factors contributing to sustained revenue and profitability is essential. This project aims to uncover these factors within the LA business landscape, focusing on location and demographic influences on sales growth.
report.pdf: A comprehensive report detailing the study's objectives, methodology, analysis, and conclusions.presentation.pptx: Slides summarizing the project's key insights and findings.clustering.py: Python script for performing cluster analysis on demographic data to identify distinct customer groups in Los Angeles.linear.ipynb: Python notebook for conducting linear regression to select the best ZIP code for opening a new business, targeting each customer cluster identified in the cluster analysis.
-
Cluster Analysis (
clustering.py): Segments LA's market into six customer clusters based on demographic factors. This script uses K-Means clustering to organize locations into groups with similar demographic characteristics, facilitating market segmentation and targeted business strategies. -
Linear Regression (
linear.ipynb): Identifies the optimal ZIP codes for new businesses by predicting sales growth within each demographic cluster. This notebook runs linear regression and ranks ZIP codes to recommend the most promising locations for each customer cluster.
The analysis employs datasets from the National Establishment Time-Series (NETS) Database and the US Census Bureau. These datasets provide a rich overview of LA's business landscape, including sales growth, demographics, business types, and owner demographics. Access the data here.
Begin by examining the report.pdf for a thorough understanding of the research and findings. The presentation.pptx offers a visual summary suitable for stakeholders. The clustering.py script and linear.ipynb notebook detail the analytical processes used to segment the market and identify optimal business locations.
This project is a collaborative effort with Herat Devisha, Diego Estuar, Yannick Angouo Lopes, Andrew Morris, and Riley Nickel.