After our excursion into the new, eciting, sometimes confusing, often just plain wierd world of Power BI, we were back in our native Tableau for our second to last Dashboard Week Challenge.

This was another data set from the US (it’s like nothing interesting ever happens this side of the pond). Today, we would take a closer look at bike traffic in the city of Seattle, Washington, brought to us by the admittedly amazing data.gov initiative. Andy left us to get a hold of our bike traffic data and suggested we shop around for some additional pieces of information to supplement our analysis. It’s easy to lose yourself in the sheer amount of data that is available on these platforms, and that is pretty much exactly what happened to me in the first hour or so. There was location data, traffic data, weather data, policy data, data about infrastructure, pretty much everything you could possibly need for the task.

After some trial and error, I decided to go for temperature data to add to my analysis. It’s a pretty straight forward assumption that warmer weather warrants more cycling, but I wanted to know just how much the two are correlated, and whether there are trends across the seasons or hardy bike veterans who brave even the grimmest of Northwestern winters.

Both my data sets had similar levels of granularity and covered the same date ranges, so I knew that it would be possible to merge them by date. The only difference was that one of my data sets had one entry for every full hour, while the other had entries that were less regular and needed to be smoothed down to the hour. Doing this took up a good chunk of the morning, since manipulating date fields in Alteryx can be a challenge, to say the least.

In the end I managed to join my data sets with accepltable loss, meaning I now had information about the current temperature for every row in my bike counter data and was ready to start exploring in Tableau.

It was here that I hit a blockade, which was strange because I knew what I wanted to do and I had sketched out how I wanted my dashboard to look, on the fourth day of consecutive dashboarding it felt like I had hit a brick wall and things just didn’t seem to come together. I wasted time going back and forth between options I had long ago rejected. I faffed around with formatting that would need redoing anyway once the dashboard was finished.

It was long after Andy’s (admittedly soft) deadline had passed that I had something I was more or less happy with. After learning the value of planning and sketching the work ahead of time, today was a great lesson in how difficult it can become to take a step back from your own work and get a fresh perspective. I was thanks to the help and feedback of my colleagues and my partner that I realised the things that were and weren’t working, and had I asked for that help sooner I’m sure I would have been able to produce the same or better results in a much shorter time frame.

It’s a lesson that will come in very handy tomorrow, since our last day of Dashboard Week will be cut short by the fact that we’ll be presenting in the afternoon.

Find my viz on Tableau Public