Content Category Popularity with Accenture
Assignment
This data analytics job simulation is a mock project via The Forage for Accenture, a professional services company. It involves analyzing data and presenting findings on behalf of a fictional client, called Social Buzz, in the social media industry. The goal of this project was to determine the five most popular categories on Social Buzz's website, visualize the findings, and present them. This information would help the client better understand how it’s succeeding with its userbase and make informed decisions about how to market itself to a wider audience.
For this assignment, I received three unclean datasets from The Forage. I deleted blanks, reformatted string data for consistency, and combined the sheets into one to facilitate analysis. Using tools such as pivot tables and PowerBI, I was able to discover not just how popular each content category was but also what implications this could hold for how Social Buzz markets itself to a growing audience.
Findings
The most popular categories indicate an interest in preserving health, keeping up with technology trends, and pets. They make up 36% of the content and user reactions on Social Buzz. This suggests that people use Social Buzz largely for informative benefits. This includes the “animals” category, which outperforms “dogs” by 22K points, suggesting that the animals category includes animals other than pets.
The score is determined by the number of positive reactions and the total number of reactions. Positive reactions lead to a higher score, as do more reactions in general. On the same note, negative sentiments and fewer total reactions result in a lower score.
The five most popular categories also have the highest post volume, suggesting that post volume is a predictor of score.
There is a strong relationship between the amount of posts in a given category and its resulting score. This suggests that more posts would raise the score of a category by a predictable amount.
Categories have similar proportions of positive, neutral, and negative sentiments. Switching categories does not guarantee a higher proportion of positive reactions. Instead, a higher post volume is more likely to increase the number of positive reactions.
The range for total reactions per month ranged from 1.9K to 2.15K. Post volume in this range is unpredictable for any given month because there is no pattern regarding seasonality.
PowerPoint Format
I organized and presented my findings using a template supplied to me by The Forage, which is linked below.
View the presentation slides here.
Recommendations
Using the insights above, I made the following recommendations to the Accenture marketing team about how to handle its growth.
Encourage high post volume from users regardless of category
Focus on animals, science, and technology when promoting Social Buzz to prospective users
Frame Social Buzz as an informative site using more popular categories when promoting the website
All templates and datasets used in this project belong to The Forage and Accenture.