MARVEL MOVIES INTERACTIVE DASHBOARD
by Abigail Gonzalez Bejarano
Current Project
This project aims to create an informative interactive dashboard in excel using a Marvel Movies dataset.
The datasets were obtained from Kaggle: https://www.kaggle.com/datasets/minisam/marvel-movie-dataset
Here is a preview of what the final dashboard looks like ->
Cleaning the Data
The datasets originally are two separate data sets. I combined the two datasets in Excel using the "Combine Queries" function, performing a "Full Outer Join" (combining all rows from both data sets by "Title". The first tab you see below is the unclean merged data sets, the second tab named "MARVEL_CLEAN_TABLE" is the cleaned final dataset. Please feel free to look at both for comparison.

Below are some of the main/major steps I completed to clean the dataset:
Remove all duplicate entries.
Use the PROPER function in excel to standardize all columns with text.
I added filters to all columns to begin making edits.
The first edit was in the Distributor(s) column. I combined all "20th Century Fox" and "20th Century Studios" to "20th Century Studios". This is because after Disney acquired "20th Century Fox" it was renamed "20th Century Studios".
Instead of keeping the "Release Date" column I created a "Release Year" column. To do this I used the =RIGHT function to only extract the year.
For "Budget (millions)" I multiplied each row in that column by 1 million in order to show the numbers in millions. This is because the remaining rows reporting money were in millions.
"Budget (millions)", " Opening weekend (North America)", "North America", "Other territories", and "Worldwide" were all rounded and standardized as currency (USD).
I added a column afterward, "Grand Profit" which simply subtracts "Worldwide (Rounded)" from "Budget in Millions". (i.e. profit = price - cost).
For the movie review columns, "reviews.Rotten Tomatoes", "reviews.Metacritic", and "reviews.CinemaScore" I used the =LEFT function in order to extract the review rating from the remaining information provided.
After extracting the ratings from each of those columns I created new columns that categorized each rating by the respective reviewer. I used the =IF function to create these new categories. (i.e. for Rotten Tomatoes a rating <60% is Rotten, >=60% is Fresh, and >=75% is Certified Fresh).
There were several other minor things I did, but the final and clean dataset is in the second tab named "MARVEL_CLEAN_TABLE".
Creating the Interactive Dashboard
With the clean dataset it was time to make pivot tables and charts! You can see here some of the charts/tables I made. The movie dropdown section that I included in the first section was made using the =FILTER function and creating sub-tables from the original dataset.
The Finished Product (Video)
