Business Data Analysis

1.Business Scenario & Problem :

I work for a video game company. Recently we have seen a decline in sales. I've been tasked with finding ways to maximize our sales through targeted marketing. I have a dataset with a breakdown of sales by region and genre.

Potential Questions

What countries/regions should we focus on? What genre of games should we focus on?

2. Data Life Cycle

3. Excel Analysis

For my excel analysis I created a pivot table that displayed the different regions as columns and genres as rows. From this we can see how much each genre sold in each region. Using the SUM function I calculated the total sales per genre across all regions. Using this table I created a bar chart to display the total sales across all regions. I used the INDEX & MATCH functions to create a new column to display the most profitable genre in each region.

Pivot table:

Bar Graph using Excel:

4. Python Analysis

Using pandas built in method DataFrame.Head() I printed the first 5 rows on to the console. I got the data type for each column using pandas built in method dtypes. Using pandas method DataFrame.sum() I got the sum of a specific column. In my case I wanted the sum of each regions' sales. With this new total I created a new dataframe that contained the 5 regions as columns and the total sales as a row. Lastly using matplotlib I created a bar graph that shows the total sales by region. All of this this is shown below.

Bar Graph using matplotlib:

5. Data Types : Most of my data types are numeric and quantitative. Since the sales are in decimals, it would be considered as continous data. My data was pulled from Kaggle. The data has columns for the year, publisher, genre and how much it sold in different regions. It had about 16,000 entries and the years range from 1980 to 2020. Some entries under Year and Publisher were unknown but since its irrevelant to my questions I left the data as is.

6. Conclusion Moving forward I would like to explore the corelation between the genres with the most sales and crimes in the region. For this I would need an additional data set and a way to directly compare the two sets. Three Things I learned

* How git works remotely and the importance of making sure I am only making changes in one place
* How to use matplotlib to create graphics
* How to use pandas to manipulate data from a csv file