A year of work at Datawrapper

Sometimes the simplest chart is the best

Hi everyone, here’s Ivan and for this weekly chart I’m filling in for Lisa!

Lisa took us down memory lane last week with her great overview of all the weekly charts that got published over the last year. This inspired me to consider what else we’ve been up to at Datawrapper during this time.

Every improvement or change to Datawrapper, whether a new blog post, a bug fix or a part of a new feature gets “commited” to our version control system (we use Git for this). The changes within a commit can be big or small, but the number of commits generally indicate the rate of work that’s taking place.

So I decided to take a look at our commit history over the last year to illustrate how hard we’ve all been working!

There is some variation in the number of commits on a week by week basis but overall certain trends can be picked out. During winter and summer holidays when there are fewer people in the office (especially when our CTO Gregor is away!) the commit rate drops off. But when a major project such as locator maps is about to get completed, the work rate intensifies to finish off all those important last minute changes and to ensure a smooth launch!

Chart Choices

A simple line chart is perfect for showing time series, especially with only one variable, as it makes it easy to spot trends over time.

The data is grouped and shown on a weekly basis. This enables us to spot finer trends than would be possible with monthly data, but it also does not cram in too many data points into the chart (as would be the case with daily data).

Since the commits were actually made at various points during each week shown, and not all at the same time, curved interpolation was used on the line to approximately illustrate this tendency (for an in-depth explanation of interpolation have a look at Lisa’s earlier Weekly Chart).

Finally, to supplement the chart there are some range and text annotations which highlight specific points in time and explain some of the peaks and troughs in the data.

That’s all from me, Lisa (or somebody else from the team?👀) will be back with you next week!