Brand Watch Report - Ogilvy Entrance Test

Introduction

This repository contains the project submission for the Ogilvy Entrance Test focusing on conducting a Brand Watch report for three brands in the F&B industry: Coke, Pepsi, and Fanta. The goal of this project is to analyze their social media presence, content strategies, and performance metrics over the timeframe from February to April.

Assignment Details

Objective

To conduct a comprehensive social media audit on the following brands:

• Coca-Cola Viet Nam

• Pepsi Viet Nam

• Fanta Viet Nam

Research Scope

1. Data Collection:

• Gather data from social media channel (Facebook).

• Collect metrics such as page followers, number of interactions, comments, shares.

2. Content Analysis:

• Categorize content angles, pillars, or messages used by each brand.

• Analyze content performance based on engagement metrics.

3. Strengths and Weaknesses:

• Provide an evaluation of the strengths and weaknesses of each brand's social media presence.

4. Key Learnings:

• Summarize observations and derive actionable insights for the client to enhance their own content strategies.

Libraries

1. Data crawling:

• Selenium: simulate web surfing behavior.

• BeautifulSoup: parse HTML structure to extract potential data.

2. Data analysis:

• NumPy, Pandas, Matplotlib, seaborn: basic data analytics libraries.

• advertools: online marketing productivity and analysis tools.

3. Deployment:

• Hosting: Azure VM

• Mercury: Python notebooks to website.

• docker-compose: follow Mercury's instruction

Repository Structure

• analysis.ipynb: Jupyter Notebook file containing data analysis and visualization scripts.

• *.csv: crawled data.

• msedgedriver.exe: MS Edge Driver.

• ogilvy_test.png: Ogilvy test capture.

• README.md: This file, providing an overview of the project and instructions.

• requirements.txtL used Python libraries.

Getting Started

To replicate or further develop this project, follow these steps:

1. Clone the repository:

git clone https://github.com/dellacer2003/Brand-Watch-Report.git
cd Brand-Watch_Report

2. Set up your environment:

• Ensure Python and necessary libraries are installed.

pip install -r requirements.txt

• Consider using a virtual environment to manage dependencies.

3. Explore the analysis:

• Open and run analysis.ipynb in Jupyter Notebook.

• Review the findings and visualizations generated from the data.

Contact

For inquiries regarding this project or further collaboration opportunities, please contact:

• Nguyen Dang Khoa, Email: khoadangnguyen1810@gmail.com

Warning

Due to Facebook's policy on limiting data scraping, its HTML structure frequently changes, causing our crawling script to become outdated over time.