Here at Profit Point, we typically put in a fair amount of effort up front to scope out a project together with our client. This typically helps us and our client to set appropriate expectations and develop mutually agreeable deliverables. These are key to project success. But another key element to project success is getting good quality data that will allow our clients to make cost effective decisions from the analysis work we are doing or the software tool we are implementing.
Decision support models are notoriously data hogs. Whether we are working on a strategic supply chain network design analysis project or implementing a production scheduling tool or some optimization model, they all need lots and lots of data.
The first thing we do (which is usually part of our scoping effort) is identify each of the data types that will be required and what will be the source of this data. To do this we start with what decisions are trying to be made and what data is required to make them successfully. From there we identify if the data currently exists in some electronic form (such as an MRP system) or whether it will have to be collected and entered into some system (say a spreadsheet or database program) and then figure out how the data will get into the tool we are developing.
Second, we try to get sample data from each data source as early as possible. This allows us to see if the assumptions that were made as part of the scoping effort were valid. There is nothing like getting your hands on some real data to see if what you and your team were assuming is really true! Often there are some discoveries and revelations that are made by looking at real data that require design decisions to be made to be able to meet the project deliverables.
Third, to help with data validation we find it extremely helpful to be able to visualize the data in an appropriate way. This could take the form of graphs, maps, Gantt charts, etc. depending on the type of data and model we are working on. On a recent scheduling project, we had the schedulers review cycle times in a spreadsheet but it wasn’t until they saw the data in Gantt chart form that they noticed problems with the data that needed correcting.
Identifying data sources, getting data as early as possible and presenting the data in a visualized form are absolutely required to make a project successful. Omitting any of these steps will at least add to the project cost and / or duration or possibly doom the project to failure.