Methods for Improving Data Quality of Material Data

Data quality is essential across the entire business footprint. Between the quality of the material data and the accuracy and confidence of the inventory data, most companies can either find or spend a small fortune. Since much of what I have done my past relates to procurement and master data, I’d like to share some insight into improving the data quality of your material data.

At one point, I was a main driver to continuously improve the e-procurement and business intelligence platforms from which some bad inputs were generated. We knew we had to analyze the material data to find the problems, but we had to break the data down into logical categories or other manageable chunks. We started by breaking materials into direct and indirect materials. We found that when we looked at the direct material space that it was a more simplified path. It had a limited supply base and a limited set of SKUs (Stock Keeping Units for the uninitiated) to deal with. When it comes to something more manageable, you can tackle that with people first, then start digging in with data tools. If you throw a good number of skilled analysts working together, you can see what needs to be done with the differing descriptions and material numberings. You can put some people governance process to consistently review the data on a regular basis to get the data clean and drive synergies. That works for manageable data sets.

Then there was the indirect space or what is typically referred to as MRO. It’s literally impossible to use the same methods because it’s far too complex.