Requirements Analysis for Enterprise Systems: How Expectations Can Meet the Reality

by Dr. Sotiris Gayialis

Business Processes and Systems’ Functionality

And the time comes when it is realized that an enterprise system needs to be upgraded or a new one needs to be implemented. What is the most appropriate software solution? Will it be adapted to the organization's specificities? To what extent should the business processes be redesigned to adapt to the new system without losing competitive advantages? Ultimately, can the gap between the system’s functionality and business processes be bridged?

Such questions are raised very early, in an information system implementation project and refer to the so-called requirements analysis. It is an important step for the success of such an IT project as it can assist in the alignment of business processes with the system's functionality.

Our relevant article entitled "A business process modeling-enabled requirements engineering framework for ERP implementation" has been published in Business Process Management Journal. The article is based on the experience gained from ERP implementation projects and the related literature. It presents a methodological framework for analyzing requirements in ERP system implementation projects, focusing on business processes, throughout the project’s lifecycle [i].

This framework adopts a hybrid process-driven and technology-driven approach for requirements engineering. It highlights the importance of the early correlation between the requirements arising from their business processes and the functionality of the ERP systems, even before the implementation begins. It also emphasizes the importance of business process modeling for effective requirements engineering, successful system implementation, and change management.

Some modern approaches for IT projects, such as agile methodology, try to address difficulties in requirements analysis by adopting cycles of gradual improvement of design and development. The success of such approaches is not guaranteed and is difficult to apply to large projects such as ERP implementations. Therefore, the ideas contained in the discussed framework are always applicable and useful. Let’s see what the framework says.

The Requirements Analysis Framework for ERP Implementation Projects

In the first phase, even before the software selection, business processes and how they are supported by existing enterprise systems are analyzed and documented using process models. These models are used to request specific functionalities from software providers, along with market research to select the most suitable software. Analyzing the existing processes (AS-IS) can lead to clear requirements within a Request for Interest (RFI). In turn, the software provider can provide equally clear answers and there can be a common understanding of the needs of the enterprise.

In the second phase, the offered by the software provider functionality of the ERP system, which is qualified as the most suitable, can be used by the enterprise to understand the changes that will occur and to redesign its business processes. So, new models of the expected processes (TO-BE) are defined. Therefore, there is better preparation for upcoming changes, as well as clear requirements for the new system and the functionality that it should offer.

In the third phase, the expected functionality based on the new models is included in the Request for Proposal (RFP) sent to potential implementers of the system, constituting the functional specifications of the system. This provides a detailed, documented, and clear understanding of what the enterprise requires and what the implementer must deliver as part of the project. A common understanding of the business needs helps in more accurate proposals from the implementers, bridging the gap between the system and the processes, and ultimately in a more efficient phase of system design and business blueprint creation.

In the fourth phase, during the implementation of the system, the modeled TO-BE processes and the offered functionality can also be utilized beyond the design phase, later in the user acceptance testing, creating real-life scenarios based on the alternative paths that the processes include. They can also be used in training actions for the formation of educational scenarios that "speak the language" of end-users and the specificities of the enterprise. Ultimately, the implementation of the system will inevitably bring changes to the redesigned processes, shaping the new AS-IS processes of the new reality which is based on the finally implemented functionality after the system go-live.

The Importance of Business Process Modeling

From the first to the last phase of the presented approach, the focus is on business processes and process models. The same modeling methods are utilized in all phases to ensure consistency in documenting the different versions of the processes. Regardless of the level of detail of process models, the methods should be decided upon from the first step of the AS-IS state and maintained throughout the whole life-cycle of the system implementation.

Therefore, a critical decision is to determine the modeling architecture from the early stages of the project, mainly referring to the different modeling views (process view, organizational view, function view, systems view, data view, system roles, decisions, reports) and the related modeling methods. For example, for modeling the process view, such methods could be BPMN (Business Process Management Notation) or the event-driven process chain (EPC), which also integrates the other modeling views.

Equally important is the decision to select a software tool for business process modeling. There are several tools in the market, such as the well-established Software AG's ARIS or constantly evolving tools like Signavio by SAP, as well as other popular tools like Adonis or IBM Blueworks.

In any case, the choice of the modeling software tool also depends on whether the company adopts it solely for documenting processes or into a broader Business Process Management (BPM) philosophy, utilizing not only process modeling, but also simulation, process orchestration, low-code process automation, and process mining.

Summarizing

The framework presented for ERP systems requirements analysis can be applied more broadly to the implementation of any enterprise information system that supports specific processes, such as an electronic procurement system, a process automation system for standardized tasks, a customer service system, or a digital transformation system for production and logistics.

This approach can be modified according to the information system project. A key component is to have a consistent modeling of processes from the outset, which is utilized throughout the system implementation project. Consistency and accuracy in process modeling and analysis will be ensured through the configuration of a modeling architecture, including selected views and methods.

It should be noted that technological advancements in business process management systems can significantly enhance the effort of process analysis and modeling, such as through the use of process mining tools to identify alternative paths and the most common instances of executed processes.

The creation of process models is proved to be extremely useful both in the phase of collecting interested software providers (using AS-IS models) and implementation service providers (using TO-BE models), helping in the common understanding of the company's operation and deviations from the best practices embedded in the information system.

In addition, process models based on the expected functionality of the system (TO-BE models) can accelerate and improve the effectiveness of fit gap analysis, business blueprint, and ultimately the whole system design phase. The design phase in an information system implementation project must have the business processes already documented and not try to explore them at that time, possibly without a uniform and consistent way with the other initiatives or projects of the company, so that the transition to parameterization and implementation can be smooth and without setbacks. Thus, a significant advantage is provided concerning the allocation of project resources, the effectiveness of implementation, and the minimization of risks from potential inadequate requirements analysis.

And to conclude jokingly, the matchmaking of the organization with the information system should not be based on ambiguities and unfulfilled expectations but on early clear requirements and common understanding, for a happy marriage.

---

[i] Panayiotou, N.A., Gayialis, S.P., Evangelopoulos, N.P. and Katimertzoglou, P.K. (2015), "A business process modeling-enabled requirements engineering framework for ERP implementation", Business Process Management Journal, Vol. 21 No. 3, pp. 628-664. https://doi.org/10.1108/BPMJ-06-2014-0051