25 January 2016 by Chris McDonald
As a business, you need to be able to track all of your sustainability information, sometimes this is done using spreadsheets, but as you grow as a business, the need for a system becomes apparent. The business gets a new system and all of the staff use it to input data, but the business is still growing and eventually, you realise you need a new system so you can add more sites, more users and more functionality.
It is uneconomical to run two systems concurrently for the same data so how do you keep all of your old data in a readable and usable form?
Here at SustainIt we deal with this exact question on a daily basis and the answer is data migration. Although this process isn’t quite as simple as it sounds, let’s delve into the nuts and bolts of the process itself.
Each system you can build or purchase needs a database to store the information the users submit, this is where the first problem lies. For almost all systems these databases are very different so there is no simple way to directly push information from one system to another.
Jumping quickly to the second obvious problem off the back of the previous point, each system’s unique database will have a certain way of accepting new data via various methods i.e. manual input, direct imports or automated imports.
The third problem we encounter, once you know where the data is coming from and where it is going, you have the manipulation. The ability to change large datasets from one configuration to another takes a keen eye and an intrinsic understanding of how the data is supposed to fit together from a standard and relational perspective.
So, from a top down perspective it is a very simple process; Export, Manipulate, Import. During my time with SustainIt I have often wished it was as simple as that, but alas, the puzzle is what makes data migrations interesting.