Data analysis and authorization management

Why data and authorization management? Answers from practice.

Hand holding modern tablet or mobile device with analytics dashboard for sales, marketing, accounting, controlling department to check revenue, sales and business KPIsSince switching from file folders to digital storage, companies have often been driven by the motto that everything is stored and stored.

Since then, this mentality has been consistently in the heads and is being passed on to the company.

Consequently, it's now considered a constant for file server data that the volume grows by 20% to 40% every year. The data and its structures have often remained unchanged for many years. Therefore, the file server needs to be cleaned up at some point if you don't want to lose track of everything completely. Because in addition to the data that is still actually used productively, the portion of data that has no productive value for the company is growing significantly.

The Data Mountain: 2 / 3 Dark Data,
much of the remainder of RED Data.

In the past, strategies were regularly developed to get more and more data onto storage systems in less time. The problem is that, to this day, there are rarely any real "exit" strategies for the data— that is, rules that define how and, above all, when data leaves the system. As a result, companies have accumulated an average of between 5.000 and up to 25.000 files per employee, creating a veritable mountain of data.

Analyzing this unstructured data, so-called dark data (data whose potential benefit to the company is completely unknown) typically makes up two-thirds of the total data. Of the remaining third of labeled or qualified data, approximately half can be classified as ROT data, i.e., data that is " Redundant , Obsolete , or Trivial ".

Who wants to clean up the file server today has two main problems: Lack of transparency and unclear responsibilities

Create transparency as a basis for decision-making
The biggest problem with the unstructured data on the file server is the missing information. While every dataset can be tracked and evaluated in a database, that keeps Filesystem only very few helpful information about the data located there. In contrast to this, the file system is particularly easy to use and almost "random". It is usually only limited by the finite storage space. So how do you find out which data is relevant?

Create Responsibilities: Integrate Data Owner
In the full directories usually also no employee knows so well. It would be helpful, the Data Owner to ask for support, but that is not known. So what to do?

To solve these problems, data must first be collected and clearly evaluated for an analytical assessment of the current situation. Based on our extensive experience with numerous projects, we know which parameters are relevant for optimally preparing the subsequent steps. These include, for example, user activity data, which allows us to identify and eliminate inactive shares . Other parameters can be used to locate a potential data owner for specific directories or at least to identify users who access them regularly.

Relevant information is the prerequisite for making meaningful decisions when dealing with data.

In addition, the specialist departments must be involved. Projects often fail because the coordination process is poor. Simply asking specialist departments to design something new and clean up old structures is bound to fail if they lack the necessary data to make informed decisions. Besides providing information, specialist departments should also be given a user-friendly interface that allows them to make changes directly within their respective areas.

Of course, a one-time cleaning up makes only limited sense, if not at the same time the course is set for the future to handle the data differently. These include the topics:

More on this topic: