Master Data Management (MDM) is a thorny topic, because there is often no logical structure, especially when dealing with a network of different systems. The result is that the same type of data is stored in different places, where it is maintained in one place but not in the other. In short, information is unreliable and that is of course not possible. It is not without reason that MDM is complicated. To make it more comprehensible, we introduce four basic principles.
Principle 1: Clear Origin
One of the problems with data is that you often have to hold the same data field in several places at the same time in your information network. If you then adjust one data field, how do you ensure that the data in the other places are also adjusted and do not go out of sync?
It is obvious that you have to ensure that the number of places in the system where the same data field is minimal, but that is not always possible. You cannot simply remove data fields in applications because programs are made so that they have to retrieve the data from a certain fixed location. The files are not the same in the different systems either. Data is in other places, some data is general data but others are specific to a certain application or program. that holds most data fields.
Schematically this looks like this:
Good MDM starts by mapping such an overview. The routing must then be determined: which field in which system or file is filled first and which field is et cetera. Preferably, that routing is automatically ended. You can also see whether you can expand the main system with the data used in other systems/applications. That way you ensure that there is really only one place where the data is maintained. If that is difficult, you should see if it is possible to somehow earmark the fields from the first file in the application that follows. What you often see is that gray areas indicate that data comes from other systems.
Once the routing has been determined, it must also be clearly established who is responsible for filling the fields. This is usually a key user because they are expected to know how much data fields work within ‘his or her’ range. As indicated earlier, Excel sheets can provide good support here. With Excel, you can get a good overview of which fields have to be filled for which key fields. Excel can also be very supportive as a check whether the correct values have been entered.
The motto is, therefore, to get those topics on the table quickly after an acquisition or merger or migration of systems to see how you think you should deal with this.
Principle 2: Strive For Clarity
Ensure that the description or coding is consistent. If you want to search on a word, it must always be done in the same way, so ‘Zinc screws 20 * 30’ for example, and then not ‘iron screw 40’.
Here too integration, and system migrations often play a role. You then have to get everything back on track. Again the motto here does not delay too long. It really gets messy if you let this run too long.
Principle 3: Strive For Completeness
Ensure that all data is filled. You must, therefore, ensure that somehow it is quickly and conveniently visible where data still needs to be supplemented or improved.
Principle 4: Perform Regular Maintenance
Master Data becomes obsolete quickly and must be renewed again and again. Set up a workgroup that works on maintenance every week led by a Logistics Manager or Financial Manager. The Data Manager receives or issues commands via this group.
Conclusions And Recommendations
Don’t make the world of Master Data Management more complicated than you really need to. R.A. Jonker, F.T. Kooistra, D. Cepariu, J. van Etten and S. Swartjes, have stated the main MDM: Do’s and Dont’s that we recommend checking here.
We have discussed the four principles that should be kept in mind when trying to structure Master Data. Make sure that the file is clear. The second principle focuses more on the content of a field. Make sure that the encoding or text is unambiguous. It should be a logical layout that is consistently followed. After all, you have to be able to search for it. The third principle focuses on keeping the data fields as complete as possible. Do not wait too long to complete the data. A lot goes wrong there. The fourth principle discusses how you can best arrange maintenance. Make frequent use of Excel and functional specialists.
Want to learn more about PIM?
If you have any questions regarding Product Information Management, from PIM Selection to Implementation or how a PIM would fit in your IT landscape? Feel free to browse our Knowledge Base of articles on everything PIM related.Visit our Knowledge Base