Dashboard week at The Data School is an intensive and challenging week of creating a new dashboard from scratch on either Tableau or Power BI each day. At the start of the day, we are given the project details and dataset or guidelines to finding a dataset. In addition to creating a dashboard, we are also required to write a blog discussing our process during dashboard week, any challenges and roadblocks we experienced, and how we overcame them. During this week, I learned a few things about how to best approach dashboard week. This blog post will take you through three thing I learned from dashboard week.


1. Understanding the data

When cleaning and preparing the data for analysis, tools such as Alteryx are very powerful and easy to use. During dashboard week, I found that understanding the data is critical to cleaning and preparing the data well. If we understand the data, then we know the necessary steps required to get the data in right shape for analysis and visualisation. During day 1 of dashboard week, I struggled to find a story with only 6 fields of data. To overcome this, I came up with an Insights Matrix that help me understand the data better and come up with questions I wanted to answer.


2. Caching Alteryx workflows

Caching is a quick and easy way to save a portion of a workflow in Alteryx so that it does not run again. The primary advantage of caching is that is allows the workflow to run faster. A good use case for this tool is when we are calling APIs because there is often a limit placed on a number of requests that can be made. If we cache and run the workflow, we limit the number of requests we make. To illustrate, this portion of my workflow was used in day 4 of dashboard week, which was about APIs and spatial data. In this workflow, we cache the workflow at the Download data tool so that the portion of the workflow that calls the API will not run again and it will only run the workflow from the output of the Download tool.

In case we reach the limit of requests we can make, you can use a phone hotspot to override this issue. This is because our phone hotspot uses a different IP address. This allows us to continue to be able to call the API. However it is still best practice to cache your workflow.


3. Dashboard designs

When designing dashboards multiple days in a row, it is important to be able to continue designing creative dashboards that best fit the data. During dashboard week, I often found myself on Tableau Public trying to find new ideas and using Andy Kriebel’s Visual Vocabulary to find interesting ways to visualise data. In addition to this, I also looked at different dashboards on Tableau Public’s Viz of the Day. To keep this interesting, I continued to play around with colours, parameters, sets and actions in Tableau to add different interactivity for users and create more aesthetic dashboards.


4. Finding a story and insights

Finding a good story from a dataset can be very challenging. Depending on the dataset, one way to do this is to focus on a specific person, country or a brand. By only telling a story from the perspective of someone, something, or somewhere, it is much easier to find a story. For dashboard day 2, we were required to create a dashboard using a sports data set. The data set I chose looked at the UFC bout results, UFC fighters statistics and UFC top rankings by division. To find a story, I looked into the world of UFC to understand who were the dominant fighters and who were the underdogs. From here, I was able to tell the story about how Alexander Volkanovski fought his way to challenge the top fighter in UFC from just entering the scene.

The journey of dashboard week has been both rewarding and enlightening. Dashboard week at The Data School truly tests your skills and reinforce what you have learned through training. This blog explored some of the learnings I took away from the week. To summarise, it is important to understand the data, cache your Alteryx workflows especially when calling APIs, continue to find inspiration online for innovative dashboard designs and how to find a good story in the data.

The Data School
Author: The Data School