I started with date, project, activity and the number of hours.

I used Pandas to convert the date to a proper datetime object then add the day number and day name against each entry, and the month name against the first day of each month.

This isn't essential for the analysis, but it gives me a way to differentiate weekdays from weekends in the final result, as well as placing labels for the months.

Now, on to the code. Normally in my analysis I aim to work backwards from the questions I want to answer, but in this case I was partly inspired by this post by Giuseppe Solazzo showing his fitness tracker steps for 2024, and he was inspired in turn by Dominic Royé's work plotting climate circles. In both of these examples, the visualisation and calculations are all done in R.

On the other hand, I tend to work a lot more in Python. Luckily for me Tomás Capretto had already looked into coverting a very similar R visualisation by Cédric Scherer into Python.

Here is Tomás's example which is already very close to what I wanted to do. The data in this example deals with the amount of time it takes to crack certain categories and formats of passwords, but all the necessary elements are there for me to deconstruct and tweak for my purposes.

• The length of each line represents the amount of time (seconds, minutes, hours, days or years) needed to crack each password in the dataset. I can use this for the number of 'writing hours' per day.
• The angle of the line in relation to the starting point represents the position of the password entry in the dataset's index, so I can order my data in date order and use the same process to make sure everything shows up in the right place on the radial chart.
• This example uses a custom colour map of ten different shades, but I only need two - one for weekdays, and another for weekends.

Next I need to decide what questions I'm trying to answer with this data. More specifically, what questions do I think would be well-answered by the radial chart I'm planning to create?

Here are the obvious ones:

• How many hours does it take to create a novel?
• What proportion of that is writing, compared to all the other tasks?
• Do I get more done during the week, or at weekends? What about holidays, or Christmas?

Let's start not at the beginning, but with the second question. This one is better answered with a simple bar chart, at least to begin with.

Plotting, outlining, typesetting, cover designs, proof-reading, uploading files, and marketing are all part of the process if you're self-published. The bar chart to the right shows the breakdown of total hours by category.

It is interesting to see the numbers, and there are a couple of surprises. Drafting and editing make up the majority of the hours, as I would expect. Marketing - which includes social media posts, designing ads, and editing photographs - looks tiny. In fact this is an almost daily activity, but usually only for a few minutes at a time. If I reproduce this diagram for 2025 later on, I expect that number will be significantly higher.

I am shocked by how little time I apparently spend outlining and plotting, possibly something I need to address!

Proportion of time spent on different tasks