This is a very interesting challenge at work. Reconciliation work is hard especially if data is not that organized. It is best to fix things at the earliest stage than just rely on reconciliation which happens at the farther end of the pipeline. I think one of the things that can be improved here is on how the collection process (input data) can be automated. This will help clean your data better. Though I understand the work right now is more on cleaning up data on historical years (2019-2021). Good luck and thanks for sharing!
You are viewing a single comment's thread from: