Adventures in Data Mapping

ArgonDigital - enterprise automation experts

Share This Post

I’ve done a lot of data mapping lately. Seriously. In the past eight months or so, my work life has been taken over by a seemingly endless stream of data mapping between interfaces for a credit adjudication system. This was an overwhelming amount of work for me at first. As I grew more comfortable mapping data, I developed a series of strategies to cope with the work.


One overarching tool for all mapping that I use is a Pomodoro timer. This keeps me focused on the task at hand, which can be quite tedious at times. When mapping, I work in 25-minute intervals spaced with 5-minute breaks in between. After every third 25-minute work period, I take a 15-minute break. Before I used my Pomodoro timer, I would max out at about 3 hours of intensive mapping. My bleary eyes really couldn’t handle much more. With my Pomodoro timer, I can work on data mapping for an entire workday.


Keeping focused on the task at hand is important, but just as important is making sure you are working on the correct task. To make sure I kept working on the highest-priority mapping items, I kept a Kanban board specifically for data mapping. I had columns for the following states.: to-do, doing, needs review, and done. I also made sure that I broke my Kanban tasks down to manageable chunks. With the Kanban board and the Pomodoro timer together, my productivity jumped dramatically.


In addition to becoming more efficient with data mapping, I became more precise. I built in validations to my mapping documents to help avoid incongruences between the interfaces. One common area of trouble is data type. Incompatible data types can be a nightmare, so I list data types when mapping up front.  Discovering a mismatch in the early stages of a sprint is certainly preferable to discovering an issue while testing. The same can be said for data field lengths. This can become a hairy issue when database storage is involved. The more information you can present, the better.


In a similar vain, attempt to list minimum and maximum occurrences for each data object. This can avoid validation errors in testing and save a tremendous amount of time and effort while testing.  As with data types and data field lengths, the idea is to present the development team with as much information as possible and communicate when there are potential issues. This may seem like extra work, but it can save days of frustration for everyone on the project team.


Another common data mapping issue occurs when there is a cardinality mismatch with complex data objects. Communicating mismatches is critical to ensure data integrity, especially when customer data is involved.

More To Explore