Since I started at The Information Lab, a common question I've been asked is how I ended up working with Tableau. To me the answer to that question is a little broader, I joined the team from a design and digital marketing background with no native experience of data visualisation in a business context beyond branding, marketing and communication guidelines. So how did I end up getting into data visualisations?
It starts with LastFM
In 2006, I discovered LastFM. The idea of being able to track what music I listened to was an exciting one. I wanted to believe that there was a logic to my music taste and that by tracking it, I'd discover the secret to my listening habits. The truth is, a month later when I went back, I realised I'd only tracked 100 plays. I hadn't setup the scrobbling properly so I invested some time into that and forgot about it. A couple of years later, back I came and I produced a visualisation similar to the one below.
I was pretty excited by this and I started wondering what else I could track. Back in 2008, there wasn't much you could do. Bear in mind phones had only just started being smart and apps were a foreign concept. However in 2009 things had picked up and Runkeeper launched. Once again I found it awesome that you could track your run and see a much more granular level of detail and stats about your run.
Then came the Feltron reports
Fast forward a few years and I came across a designer by the name Nicholas Felton. Every year or sometimes every other year, he produces a report on his travels logging his trips, photos, people he spend time with, down to small details such as his most worn shoes.
Ok his reports don't exactly conform to the principals that Stephen Few sets out but nevertheless, the breadth and consistency of his tracking techniques allows him to build visualisations that paint a much richer story about his life. I started challenging myself to collect some data about my day to day activities and started wondering what other insights you could glean if you collected data about yourself.
... then came Conspic.io
Conspicio is a Latin verb, it means "to see, to notice, to understand". So like most hobbyists out there, I gave my new hobby a name bought a domain, and set up a project, a project to collect enough data about various touch points of my life to create visualisations that were easy to look at, helped you notice trends and also helped you drill down further and understand that data. I quickly discovered several barriers, the biggest being simply tracking things in a consistent and meaningful way, whether it meant writing it down, logging activities via your phone / computer or automatically using a device. I scaled the project down and looked at passive alternatives and found a happy balance with fitness trackers, my tracker of choice being the Fitbit Flex (more on that in an upcoming post)
I'd started collecting data but what about displaying it?
... Yet another barrier, I had stuff in CSV files in excel and whilst excel had charting tools, I'd always had the vision of this data being alive, refreshing via the web through APIs and generally being more than a series of cells. I played around with various tools but never found one that handled the data in a flexible way. If you're a Tableau user, you're probably shouting "what about Tableau?"... I'd tried it in its early days but it felt like a heavy handed solution for some very simple data. It had also proved difficult to connect up to some of the APIs and databases I'd setup so I parked it temporarily.
Instead I took to my design background and designed some simple web interfaces and graphics that allowed me to track a few basic things like steps and distance travelled but in short I'm still on that journey.
The final step
The truth is, that's a lie, it's journey that's still happening. This month, I was fortunate to join a team of dedicated Tableau Gurus, we even have a Tableau Zen Master on the team and since then, the journey just changed lanes and I'm being exposed to so many concepts and techniques in data visualisation that I hadn't even know was possible, and to the Tableau fans, yes I take back what I said about Tableau earlier. It turns out that with Tableau, that most of the time, the question to ask is how to do something rather than whether it can do something.