Welcome to the third instalment of my experience with The Data School’s infamous dashboard week.

Going into DBW on Tuesday morning, I was feeling pretty comfortable with this whole arrangement; having an entire day to work on a small project seemed fairly straight-forward, but as you can expect…I was very wrong. 

It was on Thursday, the day that I created the dashboard below, that the fatigue from doing this several days in a row sunk in.


An Approach to Building a Dashboard

Despite the stress of the challenge bearing down on me, I was actually really happy with how I approached this particular challenge.

We started the day being given an ABS Census dataset for Victoria. It contained data for several different aspects of the population, over a number of years.

The catch with today’s project was that we couldn’t use Alteryx; which honestly had a big impact on me because it’s probably my biggest strength.

Here’s how I structured my work on this project:

  1. Start by looking at the data very closely. Understand the shape of it, and what information it contains.
  2. Look for pattern in the data. How do the measures change over time? Are all members moving in the same direction?
  3. Go on a story finding mission. Start brainstorming ideas and look for different opportunities without getting too stuck on any particular one.
  4. Start re-shaping the data if necessary to effectively visualise it.
  5. Find additional datasets to supplement your analysis.

1.Start by looking at the data very closely.

In the image above you can see an example of the dataset we were provided with. This tab, ECON, contains Economic data for Victoria including information about businesses.

As a business owner in a previous life, I was particularly interested in this subset of the data. I wanted to find a story to tell, something hiding in the numbers.

2. Look for Patterns in the Data

Here’s an example of the pattern I found. In the image above the blue line represents businesses with 1-4 employees, and the red line represents businesses with no employees.

The measure in this chart is the percent change in the amount of businesses created compared to the previous year.

You can see that the rate of new businesses is only growing in the 1-4 employees range. When I discovered that I knew there was a story in it.

3. Go on a Story Finding Mission

Here’s a look at the same measure except looking at business exits. Interestingly, not only are more businesses with 1-4 employees are being created, they’re also exiting less that non-employing businesses.

4. Start Re-Shaping the Data if Necessary

Here’s a look at the shape of the data we started with; you can see that each year has its own column. This “wide” dataset needs to be pivoted before using Tableau to create visualisations.

That’s why we have programs like Tableau Prep and Alteryx, to transform the data for analysis.

 5. Find Additional Datasets to Supplement Your Analysis

I found this dataset published by the ABS when looking for additional data to supplement my analysis and drill-down further into the story of small businesses.

The ABS counted all businesses started in 2018-2019, and then tracked the survival of those businesses over the following 3 years.

The data was broken down similarly to my first dataset; by turnover, and by employment size. This was the perfect dataset to supplement my analysis.

Wrapping Up

This dashboard was the easiest one for me to present; despite having the least preparation for it. If I had to speculate, I think that the time I spent finding a good story helped make the presentation write itself.

I’d found a clear story about micro-businesses in Victoria and New South Wales and it was a compelling one.

One day is not enough time to completely tell a story, so I hope to be able to revisit this some other time.

For now, it’s time to work on the final day of dashboard week — and it’s data from the Big Bash League!

Until next time.

Dan Lawson

Daniel Lawson
Author: Daniel Lawson

Right off the bat I can tell you that I’m not your average data analyst. I’ve spent most of my career running my own business as a photographer and videographer, with a sprinkling of Web Development and SEO work as well. My approach to life and work is very T-shaped, in that I have a small set of specific skills complemented by a very broad range of interests; I like to think of myself as a dedicated non-specialist. Data Analytics, and Programming, started as a hobby that quickly grew into a passion. The more I learned the more I looked for opportunities to pull, manipulate, and join data from disparate sources in my life. I learned to interact with REST APIs for services I used, personal data from services I use like Spotify, and health data captured by my devices. I learned SQL to create and query databases, as well as analyse SQLite files containing my iMessages and Photos data on my Mac. Every technique I learned opened up more possibilities; now I’m hooked and there’s no turning back. Learn More About Me: https://danlsn.com.au