top of page
Filming a Music Video

Data Analysis on 
YouTube Trending Videos

Extracted valuable insights for stakeholders by using MySQL, Python and PowerBI to analyze YouTube dataset of 50k trending videos. Generated a Wordcloud of popular video tags, and analyzed popular channels to provide suggested video contents.

SQL Project

Business Objective

Dataset

Actions

Result

To extract data-driven business insights and come up with suggestions to key stakeholders (aspiring YouTubers and companies that wish to collaborate with influencers).

Data on 50,000+ daily trending top 200 YouTube videos with 14 columns including category, tags, views, and so on.

1. Cleaned the Dataset

My team first figured out there are emoticons and links in the dataset, causing problems when importing data. We first used Python to drop unrelated columns and duplicated rows, and imported the dataset into MySQL.

 

2. Thought as Stakeholders and Defined Business Problems

Before I started to analyze the data, I put myself in the stakeholders' shoes and came up with several questions they might encounter. If I were an aspiring YouTuber, I would like to know which video category was most popular and what strategies could help make a viral video. On the other hand, if I were a marketing coordinator who tried to collaborate with a video influencer, I would like to have a list of suggested YouTubers who could help promote my products efficiently.

 

3. Extracted Insights and Told the Story

I used MySQL to analyze data and figured out potential categories with great amount of views and steady growth rate. In addition, I used qualitative method to analyze video contents made by popular YouTubers in order to generate more insights.

The project gave a general idea of most growth video categories and popular video tags in two different time periods, before COVID and Post-COVID. In addition, the project generated valuable insights and data-driven strategies for stakeholders.

Results of Analysis

Below are the parts that I was responsible for the team project. 

Heading 3

Suggestions for Stakeholders

  • For a new YouTuber who wants to open a personal channel, I suggested that he open a People & Blogs channel since the category has a steady amount of total views and most channels are individual content creators.

  • I further digged into the top20 YouTube channels with most views in People & Blogs Category by MySQL. According to my analysis, 55% of these content creators are male and top3 channels usually released funny and interesting videos.

Heading 3

Suggested Video Contents
for Stakeholders

  • I used quantitative method to analyze top20 YouTube channels with most views based on their video contents. I put some tags, such as funny, challenges and makeup, regarding their video contents.

  • I discovered that challenges, daily vlog and family are the most popular contents among these trending videos. 

  • These three video contents are the suggested contents in order to make a trending video

  • LinkedIn

© 2022 by Peggy Liang. Proudly created with Wix.com

bottom of page