In this blog post we will continue our look at different sorts of time series visualisations with the cycle chart. These charts are, like heat maps excellent ways of visualising seasonality, but unlike heat maps they also give a sense of the trend over the period in question. Whereas heat maps are generally used to look at seasonality in an average sense cycle chars are more often used to see a combination of seasonality and trend over time. The basis of the cycle chart is the Line graph, like the one below.
Here we have a chart showing the change in a measure (Sales) over time (Year and Month). Time has been left as two discrete pills, and this is important as to create the cycle chart the first step is to swap the order of these two pills.
This will give us a modified line chart as seen below. Here rather than looking at time in a linear way we look at each month of a calendar year and how the amount of the measure has changed for that month over the years of the dataset.
To make this chart even clearer I prefer to change the mark type from automatic to area, this gives a better sense of seasonality while preserving the trend for tech month.
Now in this chart we can see clearly that there is a spike in sales over the last quarter, but that in January and august there were major jumps in sales in 2014. It is also nice to note that there are relatively few instances of months where there was a fall in sales between 2013 and 2014.
Hope you have found #TodaysTableauTip useful, and as always we appreciate any feedback you might have!