The goal of this project is for you to practice what you have learned in the Business Intelligence chapter of this program. For this project, you will choose a data set, explore the it using Tableau, and put together a Story for presentation showing the insights you have derived from the data. You should demonstrate your proficiency using Tableau and the concepts you have learned throughout the chapter. The workbook should be saved to Tableau Public and a link to the workbook should be provided.
You will be working in pairs for this project
The technical requirements for this project are as follows:
- You must construct a Tableau Story consisting of at least 5 story points for the data set you have chosen.
- You must use Story features such as captions and annotations.
- You must demonstrate all the concepts we covered in the chapter (sorting, filtering, different visualizations types, aggregations, etc.).
- Your Tableau workbook consisting of at least 5 visualizations and 1 Story should be saved to Tableau Public.
- You should create a Github repo for this project, and your data should be saved to that repo in a folder named data.
- You should also include a README.md file that describes the steps you took, your thought process as you built your visualizations and Story in Tableau, and a link to your workbook on Tableau Public.
The following deliverables should be pushed to your Github repo for this chapter.
- A Tableau workbook uploaded to Tableau Public that contains the visualizations and Story you created from your data set.
- An data folder containing the data set you used for your project.
- A
README.mdfile containing a detailed explanation of your approach and code for constructing visualizations and organizing them into a Story as well as your results, obstacles encountered, lessons learned, and a link to your completed Tableau workbook.
- Find a data set to process - As great places to start looking we recommend Kaggle, Pordata, and EuroStat.
- Explore the data set and come up with a variety of visualizations that you can potentially include in your story.
- Break the project down into different steps - identify the entities/dimensions in your data set, explore them each individually, and then progress to analyzing different combinations of them.
- Use the tools in your tool kit - the concepts and methods you have learned in the business intelligence chapter as well as some of the things you've learned in previous chapters. This is a great way to start tying everything you've learned together!
- Work through the lessons in class & ask questions when you need to!
- Commit early, commit often, don’t be afraid of doing something incorrectly because you can always roll back to a previous version.
- Consult documentation and resources provided to better understand the tools you are using and how to accomplish what you want.
Dataset
- You identify clearly the origin of your data
- Clear explanation of your a priori data transformation steps and why you did those
- Mixing/enriching dataset is priority, data should talk by itself before moving to the visualization
- Good data profilling should be done before you jump to visual analysis - get to know the integrity of your data, how it behaves across all variables, clear statement of different relations
Problem formulation
- After understanding the data, make sure your formulate a good problem/hypothesis
- Place yourself as the final users of your visual analysis
- Do not answer to more than 4 questions on the same dashboard/scream - users will be confused
- Keep it simple - maximize the answers using the minimum number of data points
Data Visualization
- Data Visualization
- Do simple chart, correlate them if needed (brushing, direct filtering)
- Describe on the documentation how you visualy encode your data (e.g. Sales are encoded on the size of bars, and each bar represents a different Country. It is possible to filter - facet - the entire chart by Year)
Text
- Your titles must be relevant
- Subtitle and/or annotations provide additional information
- Text is hierarchical in size, readable and horizontal
- Data is labeled directly and labels are used sparingly
Arrangement
- Data is intentionally ordered
- Is your chart displacement representing the correct data granularity changes (e.g. more aggregated on the top)
- If your charts correlate with each other in the dashboard, place them accordingly (e.g. do not filter from the bottom to the top)
Color
- Color scheme is intentional, used to highlight key patterns, readable when printed in black & white, and sufficiently contrasts with background
Lines
- Axes do not have unnecessary tick marks, and graph has one horizontal and one vertical axis
Overall
- Graphs highlights significant finding or conclusions
- No addition of unnecessary graphs
- Individual chart elements work together to reinforce the overarching takeaway message
Feedback
- Your feedback is also a part of this project!
- When other groups are presenting, put yourself on the shoes of the final user and comment accordingly with collaborative and constructive feedback