Expert Information - 29.10.19

Data Quality: Protecting the Achilles' Heel of Digitalization

10 Questions and Answers about Data Quality and Master Files Management

The quality of data in digital processes directly influences process efficiency and the success of the company as human revision becomes obsolete. This is why experts urge companies to improve data quality. But does this actually pay off? And if so, where to start? In this article, ERP provider proALPHA has compiled frequently asked questions and their answers.

  1. Can data quality even be measured and evaluated properly?
    Experts consider up to 15 theoretical dimensions of data quality. The practice is much simpler: Automated processes work with complete, current and – most of all – unique data. Efficiency falls short where duplicates in the part master files or the customer data are concerned. If there are two records for one customer, for instance, he might end up being treated as a C customer even though he actually is a B customer.

    Not every information exhibits the same level of importance. Different spelling of an address does not have the same effect on process efficiency as a missing list of terms for a supplier. That is why the evaluation of the detected errors is an essential part of measuring data quality. In addition, particularly critical cases demand an escalation workflow to eradicate errors as soon as possible.

  2. Is it worth the effort? Isn't everything going to be just as messy as before shortly afterwards?
    Everyone who has already tried to get a grip on data quality with several projects will feel like Sisyphus from the Greek myth. You finally made it to the top, but then the boulder rolls back down the hill and you have to start all over again. Experience has actually shown that the effect of project-based cleaning wears off again after a certain amount of time. Just as with running, it is important to keep at it – and to implement a program for data quality.
  3. Where to start?
    Ideally, a company starts at a point where better data provide the fastest added value. For instance, this could be in purchasing, since supplier addresses, terms and replenishment times considerably speed up operational procurement. Starting in production or logistics can also help to clean part master files. Parts are then completely assigned to their groups and all required weight measurements are available for shipping. Depending on the industry and company concerned, sales and service can also benefit immensely from up-to-date address data and contract terms.
  4. Can a data quality program also be established without the help from analysis gurus or Excel specialists?
    Modern analysis programs don't require programming skills anymore. Defining the rules is not rocket science – not for a user who is somewhat familiar with the system. Once the rules have been defined, the employees of the respective departments will be notified of the data that need to be updated. Ideally, they can just click their way through to the record concerned, which saves a lot of time. This process also brings about a quick learning effect. Consequently, the frequency of errors will decrease.
  5. How fast can these rules be adjusted to new requirements?
    Nowadays, changes can be implemented at short notice. Modern analysis tools can be operated without the programming skills of a software provider or IT expert. However, companies need to make sure that rule changes don't contradict each other or lead to other problems elsewhere. Data governance is a must here.
  6. Do all data have to be in one system for an ongoing data check and clearing?
    If you believe that, then you're far from reality. Nowadays, the great majority of companies work with more than one system. Verification software, so-called data quality managers, effortlessly integrate data of several sources and check it collectively.
  7. How to deal with the topic on an international level?
    There is no alternative to master files management here. The most important thing are clearly established responsibilities: Who is responsible for what data, who can and who must change data and where, i.e. in which system. Master files management regulates which data are updated centrally and which ones locally, and takes care of the necessary synchronization.
  8. How can progress be measured and documented reliably?
    The reports for regularly conducted analyses cannot be limited to showing individual errors only. They must also allow for controlling that pays attention to the "state of the data". This status report will prove to the departments and the management that data cleaning pays off and that the efforts bear fruit, even sustainably. This might even lead to a healthy competition among the different areas.

  9. How does a program for continuous data quality work?
    Data quality manager software checks the previously extracted part master files, customer files or other data against a set of rules. For instance, ZIP codes in Great Britain are alphanumeric, whereas the German and Austrian ones only contain digits. It is also possible to refer to external databases to check the plausibility of ZIP codes and streets. The software does not just detect errors, but also categorizes them by serious deficiency or minor flaw. The detected errors will then be transferred to the target system along with the evaluation. In most cases, this is an ERP system. There, the employees can correct the files directly. If a specific case turns out to be an exception, it is added to the set of rules. Nowadays, this can all be accomplished without having an employee or consultant program even one line of code.
  10. How often should the data be checked?
    There are no definite specifications. The frequency depends a lot on the respective company, its processes, and data. Like every other workout program, this needs to be adjusted to the individual goals and performance parameters. The crucial factor here are continuous and regular checks and progress measuring.

Meanwhile, most companies are aware of their Achilles' heel and ready to work on their data quality. The ones who have already started were able to observe a double effect: On the one hand, data quality management provides for more production and process security, and consequently for informed decisions within the company. On the other hand, reliable information about delivery dates and availabilities increase the satisfaction of customers as well as suppliers and speed up the collaboration.

Recommend Site:

Press Contact

Susanne Koerber-Wilhelm will be happy to answer your questions and provide you with further information about proALPHA.

+49 89 92306841-445

Weitere Fachinformationen