Timeseries plots in python can tell you many important things. And the first thing that they can do is give you a sense of the direction and trend of your data. This can be critical for understanding what the future will hold.
You can supplement this with a rolling average. You can use handover to create the average across days or months and display it on your graph. This is very useful for noisy data, where the overall trend can be lost in the scattered data points (particularly if there are multiple data points per time interval). You could plug the median instead, or even 25th and 75th percentiles, which can be valuable for data where the spread changes over time as well. It’s important that you use colors that stand out nicely in this case – use matplotlib’s “alpha” parameter and the “markers size” parameter to make the raw data points fade into the background, so that the trend lines really pop out.
You can also have a horizontal line, representing the overall average. This is useful for telling the simple story that recent data have moved significantly from the historical average. It will make it extremely clear if there is a point in time where the trend really shifted, such as if you are visualizing website visits, and showing that they jumped and continue climbing after a particular intervention.
Finally, the simplest solution may be to graph your data into different colors. This is the most elegant way to show that things change from one region to another, such as before and after you launch a new ad. No need for any other markers breaking up your chart, though you can can add a legend if you want to explain what drives the differences. For this you can use MPL color perimeter, or simply work with the colormap. Here is a link to all of the default color maps.
I hope that you find this for showing the changes between different time periods on your plot. It’s one of the most basic thing that you will find yourself wanting to do with Python on time series plotting, so it’s worth getting right!