The practical implementation of an MDM approach is associated with special challenges. In addition to selecting the right software solution, organisational issues also come into play. The following article therefore takes a closer look at this topic.
This article is part 3 of a three-part series on the topic of "Master Data".
The three parts of the article series are:
- ERP Systems, Master Data & New Challenges
- Master Data Management in the ERP System or Across Software? (Part 2)
Important: Support From Top Management
Even if you can muddle through for a while by occasionally cleaning up the master data, in the long run there is usually no way around a systematic solution. This is especially true when a company is growing. The rapidly increasing amounts of data also favour this development. If one wants data, analyses and reports to be meaningful, correct and clean master data are indispensable. Appropriate software solutions can be quite useful in such cases. But it is a common mistake to believe that all problems will be solved with the purchase and implementation of a specific software. Rather, a number of questions should be clarified in advance and the necessary preparations made. This can prepare the ground for a successful MDM software deployment.
When introducing and subsequently operating a software solution for master data management, support from higher management plays an important role. In addition, those responsible from relevant departments should be involved as early as possible. This is the only way to adequately consider important needs and avoid costly changes later on. This step also helps to clarify where which data is located in the company and which of it is relevant. The procedure for the regulated transition from old to new systems and the protection of the latter from contamination with erroneous and superfluous data should also be clarified here.
Master Data Management: The Ignition Power Of The First Project is Decisive
When it comes to the first project, it is often recommended to choose a low-hanging fruit. Although this advice is often useful as a rough guide, following it rigidly can yield suboptimal results or even result in the failure of a project. The reason for this is that successful pilot projects depend on two other, more important issues. First, the ROI of the first project should be large enough to make the pursuit of similar projects financially worthwhile for the decision-makers involved. Secondly, a meaningful and logical path - consisting of a transferable body of knowledge and experience - should be apparent to other stakeholders that could benefit them in implementing their own projects. In this way, further persuasion is also facilitated and the prospects of securing the lasting support of the management and the departments increase.
Example of Prioritisation in Pilot Project Selection
For example, the practical implementation of the requirements of the GDPR is undoubtedly a task that should be given high priority. As important and topical as this matter is, it would perhaps be unfavourable to choose it as the initial project for comprehensive master data management. In this case, the project experiences could possibly only be transferred to other similar areas, such as the management of product master data, to a very limited extent. The focus of a master data project in the context of the GDPR would not, for example, give high priority to finding and deleting duplicates as long as they are unobjectionable from a data protection perspective. It would be much better to consider the implementation of the GDPR requirements as an essential part of a larger master data project.
Data Silos: Removing Another Stumbling Block
Another major challenge in organisations is the elimination of data silos. Because only when data flows unhindered can a uniform view of all relevant information and interrelationships in the company emerge. Even a software solution can only optimally support this process if the exchange of data between departments and their systems is unhindered. This problem can often be remedied with the use of technical solutions. In this context, for example, APIs (Application Programming Interfaces) and partly other point-to-point integrations offer an interesting possibility to automate the data flow between different systems.
Master Data Management Software: These Are The Points to Watch Out For ...
On-Premise vs. Public Cloud
The general trend towards cloud computing has not left the management of master data untouched. The debate about master data management in the cloud or on-premise is as intense here as in other areas. Even though in the case of cloud computing, issues - e.g. regarding data security and sovereignty - should be critically considered in any decision, it is also important to adequately include the benefits. Public cloud solutions give SMEs relatively cheap access to software with functionality that only large companies could afford in the past. There is also a lot to be said for a public cloud solution, especially in terms of security - provided it comes from a reputable provider. As a rule, public cloud software is always updated to the latest security standards. In this respect, it is superior to many on-premise solutions, where updates and upgrades take place at much longer intervals. In addition, the topic of data security and availability is usually of great importance for the success of cloud service providers - also because security mishaps are often publicly discussed. As a result, the investments in this area on the part of cloud providers are significantly higher compared to normal companies.
One Point to Note: Contemporary Core Functionality
There are now many software solutions for master data management on the market, so it is not always easy to get an overview and make a decision. A promising strategy here is to first familiarise oneself with the up-to-date core functionality. In the next step, one could then devote oneself to the specific requirements of one's own company. The core functionality of modern MDM software should include, for example, the ability to smoothly unify master data from multiple sources. This process can be supported and automated through the use of APIs, for example. Another aspect to consider would be innovative technologies such as machine learning or predictive analytics. Even if they are sometimes rightly criticised by critical voices as immature and overhyped, they can still often lead to fewer errors and time savings with larger amounts of data compared to manual data processing. In addition, software support for a comprehensive hierarchy of master data is also important. The reason for this is that many products and services today are being broken down into more and more sub-products and components due to the continuing trend towards variant diversity.
Other Functions: Checking Correct Data Entry & Data Standardisation
Another important function is a so-called "first-time-right" approach. Here, data are already checked for duplicates and other deviations or inconsistencies when they are entered. A search function that allows searching for as many parameters as possible within the data records is also very important. This functionality can be used in the CRM system, for example, to standardise master data that refers to the same customers. The concept is also known as entity or identity resolution. For example, if there are two or more customers in the CRM system named Peter Schmidt, they can usually only be distinguished and kept apart by including such parameters as the address or telephone number. Finally, a user-friendly interface is important. For example, it should also support functions that could be used to compile information in dashboards in a clear and concise manner.
With a software solution alone - as helpful as it can be when used correctly - the challenges of master data management cannot be overcome. Essentially, it also depends on the preparations in advance and the subsequent implementation. Special care should be taken to ensure that the first project is strong enough, both financially and in terms of experience and knowledge transfer, to provide an initial spark for further relevant MDM projects in the company. Finally, MDM software should have both up-to-date core functionality and be adaptable to specific needs.