The third part of our series is about time pressure and possible capacity bottlenecks that can arise when implementing new requirements. We will explain how data virtualization can help to solve these problems.
“Companies are pushing an ever-increasing wave of requirements ahead of them.”
Companies nowadays have to be able to react very quickly to regulatory changes, changes in the market or other requirements. At the same time, systems and processes are becoming increasingly complex due to growing requirements. This dilemma is leading to crucial tests, particularly in IT.
Increasing Requirements and Growing Complexity
Rising or falling interest rates, company acquisitions or sales, crises, sanctions or changing accounting standards, companies are pushing an ever-increasing wave of requirements and necessary changes in front of them.
While the technical challenges caused by the changes are mostly spread across different areas (accounting, controlling, manufacturing, etc.), they all lead to new or changed requirements for IT.
The company's IT is forced to prioritize the topics, with the result that numerous requirements cannot be implemented at all or cannot be implemented in the short term. In these cases, the end user remains solely responsible for the problem and its solution and uses the means at his disposal. In most cases, this results in repetitive, time-consuming and possibly error-prone manual processes.
“Checking multiple scenarios is usually as impossible as ad-hoc requests.”
A direct consequence of this is that the employee spends most of the time collecting and preparing data. There is often not enough time for a qualitative evaluation, the interpretation of the data and the derivation of recommendations for action.
Due to the large amount of time required for each individual run, options for action usually cannot be evaluated. The user is forced to define and apply basic hypotheses in advance. A review of multiple scenarios is usually as impossible as ad-hoc requests outside of the regular reporting cycle, which makes it difficult to identify developments at an early stage.
Lack of Separation of Duties in Development
The proper storage and management of company data consists of a large number of sometimes very complex tasks that require technical understanding and experience. For this and other good reasons, corporate IT not only feels responsible for controlling these processes, but also for maintaining the quality and reliability of the data.
When interpreting the numbers and defining the business logic, the responsibility lies with the specialist and business areas that can understand and interpret their meaning.
In the classic data warehouse, these two tasks cannot exist separately. This usually leads to large projects in which both parties first define requirements together and later implement them while having a constant mutual exchange.
Complexity of Business Logic
In some departments, business requirements, reporting and business logic are so special and complex that they have to be programmed by the departments in suitable programming languages. In classic modeling, these calculations are then either integrated by IT as part of the data flow or they are included directly in the analyses by the department.
“If multiple versions of the analyses exist, there is a risk of inconsistent business logic.”
Both approaches have serious disadvantages. In the first case, changes can only be implemented by IT, even if they are defined by the department.
In the second case, the logic is located in a static form, decentralized in various analyses, for example in various Excel files, scripts or dashboards. Changes are often difficult to understand and can only be reversed if an old version of the entire analysis still exists.
If multiple versions of the analyses exist, there is a risk of inconsistent business logic.
Optimal Work Distribution With Data Virtualization
Data virtualization makes it possible to separate data storage from data usage. While IT is responsible for creating a consistent and robust database, the end user can use, transform and combine the data with other data according to their permissions. Additionally, the user can apply business logic to it and integrate the result into use cases directly.
“This separation of duties reduces the pressure on the company's IT and at the same time increases the speed of implementation of technical requirements.”
This enables fully automated data procurement and allows the end user to expand their own analyses and enrich them with data from other sources.
This separation of duties reduces the pressure on the company's IT and at the same time increases the speed of implementation of technical requirements. Additionally, content and technical specifications are not restricted in terms of quality and reliability.
Centralization and Auditability of Business Logic
Data virtualization allows logic to be maintained and provided centrally by the department, which means it can be applied consistently across the entire company.
There is the option of storing the logic in software repositories. These ensure professional, fully auditable retention and management of business logic by preserving a change history storing all details about the author, time and scope of changes.
This not only provides a complete overview of all changes, but also the possibility of undoing them at any time or restoring historical states.
The integration of the logic takes place in the virtual data room by linking the data with the corresponding logic. Individual logic components can not only be reused, but can also be orchestrated with data, virtual models and other logic components to form new virtual models.
With this approach, time pressure and complexity can be significantly reduced, IT and specialist departments can make optimal use of their expertise and thus permanently solve capacity bottlenecks.
Comments