Finally decided to take on all required full stack technologies, and this is a great way to play around with it.

Does it need to sync with the blockchain again every time you make an update or is it just for this update?

This should be the last full sync. After that only when major changes are made to what data is being stored for analysing. When that is the case it should not be a problem to only update that data, not all.

what your process of creating the database model was like?

This process was quite straight forward. I started out with just deciding on what data to store and keep data separated in different tables that can be separated for quick access. The biggest problem is going over the huge amount of operations for analyses. I did some testing with 10 000 000 blocks of data and this resulted in adding indexes tot the timestamps.

For the future if I am looking to do more account specific data analysis adding indexes to the accounts is most likely also a must. The indexes do become quite large. But they can always be added later on. It is mostly trail and error.

The next update sounds pretty complicated from a design point-of-view (I'm not that good at it myself, so 31 different buttons sounds pretty intimidating).

Haha yes, will be fun... I did some preparation already, the API is already set up. I just need to connect the Django database models to the correct key and indeed create 31 buttons.

Do you have any other future plans, or something like a general roadmap?

Yes the idea is to switch out the Chart.js library for something more fancy that allows for zooming and is better equipped to deal with many data points. This should allow users to select any operation type and for any custom period. I would also like to see the ability to combine different charts together.

After that is set and done I will be looking into different analytics. Like tracking which accounts account for most of the volume per operation over longer periods of time. And maybe user specific analytics. This is looking to be quite a challenge due to the large amounts of data involved.