Do you remember the early days of Dreamweaver? To improve web development efficiency, Dreamweaver introduced a visual drag-and-drop capability to generate webpage code. Thus, it is evident that the exploration and development of low-code and no-code approaches actually began quite early.

In recent years, the term "low-code" has suddenly regained popularity, with related startups emerging. The sudden surge is still driven by the demands of enterprise digital transformation. Under the vast need for software development, existing software productivity has struggled to keep up, and low-code technology offers a way to break through this impasse.

01 About Low-Code

1 What Is Low-Code?

Compared with conventional software development, low-code and no-code essentially provide a more abstract development language. As shown in Figure 1, they are built on modeling languages. Modeling languages offer a higher level of abstraction than assembly and high-level programming languages, lowering the barrier to development. Once users understand the graphical elements, they can visually drag and drop them to build applications that implement business logic. BPMN is a mature process-modeling language specifically designed for business-process scenarios. This helps explain why many low-code products succeed in process-management-oriented applications: in addition to market demand, they are supported by a mature theoretical and technical foundation.

There is another approach that relies on accumulating a large number of basic scenario components and then building scenarios in a "building blocks" manner. In my opinion, this approach has a low probability of success. No matter how rich the component library is, it is difficult to adapt to ever-changing business scenarios. Only the ability to describe scenarios using a "language" can truly achieve flexibility and adaptability. Therefore, given the current maturity of modeling languages, business scenarios in the process management domain are more suitable for leveraging the value of low-code technology. Beyond market demand, mature theoretical support at the technical level is also crucial.

Figure 1

2 Why Low-Code?

After digitizing their core business processes, enterprises often turn to improving internal operational efficiency, which is also a core source of competitiveness. However, internal operational support systems are generally less critical than core business systems, so enterprises cannot justify excessive software development investment in them. Nor can a feature-rich system such as ERP or OA cover every peripheral, fragmented, and highly customized management scenario. This creates a sharp increase in development demand. If IT departments continue to rely on traditional development methods, whether in-house or outsourced, their productivity will be unable to keep pace with these frequently changing, fragmented, and personalized needs. This is where low-code demonstrates its value. One of its most important benefits is enabling citizen development, in which everyone can become a developer. When effectively adopted and operated, low-code can increase software delivery productivity exponentially. Thus, the value of low-code becomes prominent. Its key advantages is enabling "citizen development", empowering everyone to become a developer. If effectively implemented, software productivity can increase exponentially.

Process Modeling Language

Since low-code is built around modeling languages, we can use process modeling—the most common example—to understand how such languages work. ITSM supports an IT O&M and IT O&M process-management framework in which process-centric business scenarios are prevalent. Focusing on the process domain therefore helps explain the underlying logic of low-code and provides a foundation for understanding its application to ITSM. A well-designed ITSM platform leverages low-code engines to deliver such flexibility. A well-designed ITSM platform leverages low-code engines to deliver such flexibility.

In the field of process management, the most widely used modeling languages are standards, such as BPMN, DMN, and CMMN released by the OMG. Leading open-source process engines and rule engines, such as Activiti and JBPM, are designed based on these standards.

Figure 2

Of course, these engines do not fully implement the three standards, which is influenced by their development focus and historical evolution.

1 BPMN

In the design of management mechanisms, a crucial aspect is the design of the "workflow." Generally speaking, the more mature the management process, the more solidified the workflow becomes. This means that certain tasks no longer require much creativity and can achieve ideal outcomes by following a fixed workflow. The BPMN standard is particularly suitable for such scenarios, as it is used to describe predefined, fixed-sequence workflows.

Figure 3：Standardized Workflow

Key elements in the BPMN language include activities, gateways, and events. These graphical symbols are used to describe a standardized workflow. (For more detailed information on graphical symbols, please refer to official documentation, which will not be elaborated in this section.)

2 CMMN

Fully standardized workflows represent an ideal state. In practice, an organization’s management maturity evolves gradually from disorder to a defined level. Before the organization identifies the practices best suited to it, many issues still depend on individual initiative and expert judgment. For example, when an IT O&M incident occurs, an experienced engineer may need to make an ad hoc judgment, break the work into several key tasks, determine who should perform them, and define their sequence. Such uncertain and non-predefined work is difficult to model with BPMN, whereas CMMN is better suited to these scenarios.

Figure 4：CMMN scenarios

3. DMN

Another important activity in the management process is decision-making. Different decision outcomes can affect subsequent workflows. For example, whether to execute a specific change requires human judgment of the risk level of the change to determine the subsequent handling strategy. For such decision-making scenarios, the best practice is to describe them using the DMN standard. The comparison in the Figure 5 clearly illustrates the differences in decision-making scenarios with and without DMN.

Figure 5

4. Best Practices

As mentioned in the previous sections, different standards apply to different scenarios in process modeling. So, how should one determine the appropriate combination of standards when designing a complete process? The industry also provides best practice principles for reference:

Figure 6: Best Practices

• If a business process involves a series of gateways, consider replacing them with a DMN decision engine.

• If a business process involves a large number of events, prioritize using the CMMN standard.

• If a business process includes a series of additional approvals or flexible workflows, prioritize using the CMMN standard.

• If elements within a CMMN case have strict execution orders, prioritize using the BPMN standard.

02 Low-Code Engine Design

引擎模块体系

要通过低代码完整地实现一个管理类应用的闭环构建，通常需要五大引擎能力进行配合。

Engine Module Architecture

To fully realize the closed-loop construction of a management application through low-code, it usually requires the collaboration of five core engine capabilities.

Engine Function Design

The Figure 8 provides references for the positioning and functional design of each engine module.

Figure 8

03 Application of Low-Code in ITSM

Ticketing System Setup

Figure 9

The design of operation tickets best reflects the business logic of IT O&M management. Details such as the definition of event priority or problem categories can impact operations and even affect regulatory compliance. Therefore, even after the initial implementation phase, subtle adjustments to defined forms may be required during operations. Thus, the flexible ITSM forms are crucial. Based on a form engine, visual modeling of O&M tickets can be performed, including definitions for form fields, data sources of form fields, interaction rules between fields, and data linkage rules among fields, to accommodate frequently changing form scenarios. Common application scenarios include:

• Customizing process forms for incidents, problems, changes, etc.

• Customizing scenario-specific process forms, such as resource requests or access permission applications

Construction of IT O&M Workflows

Figure 10

Operation workflows reflect the collaborative processes in operations. In IT O&M management, collaboration is not only between people but also between people and systems or between systems. Based on workflow engine, end-to-end collaborative modeling of operation tasks can be achieved, including definitions for activities, branches, approvals, and events within the workflow. Application scenarios in ITSM include:

• Approval Scenarios: Multi‑level Approval, Multi‑person Approval, and Additional Approver Approval, etc.

• Collaboration Scenarios: Circulation between first and second lines of incident handling, ticket reassignment, task allocation, etc.

• Integration Scenarios: End-to-end integration between workflows and automated workflows.

Construction of IT O&M Management Strategies

Figure 11

In practical IT O&M work, many management-policy rules are involved, such as incident assignment strategies, priority matrices, risk assessments, and approval policies. Although these rules may appear simple, their impact on efficiency and cost can be significant because they are executed frequently. For example, service desk personnel may perform hundreds of assignment actions each day. With a rules engine, these rules can be codified flexibly through decision tables, decision trees, and other methods, replacing repetitive manual operations.

Construction of IT O&M Reports

Measurement is a prerequisite for the continuous improvement of IT O&M management. It involves two key aspects: designing metrics and obtaining accurate measurement data. Measurement reports are not static; they evolve as management maturity increases, and requirements vary by stage and manager. During the early stages of process implementation, the focus may be on whether the process has been adopted, so metrics related to ticket volume receive greater attention. As the process matures, teams begin to focus on outcomes such as closure rates and average processing time. As management requirements advance further, SLA compliance may also need to be measured. A lightweight reporting engine can respond flexibly and dynamically to these needs. Through data integration, the definition of measurement dimensions and metrics, and dashboard orchestration, IT O&M measurement reports can be generated quickly.

Construction of IT O&M Portal

ITSM serves a wide range of users, including not only operations engineers but also end-users and operations managers. Different roles have different focuses. To deliver an enhanced user experience, user-oriented views may be highly personalized. Based on a view engine, it is possible to define view content, layout, and styles to meet personalized demands for interaction and visual presentation.

04 Summary

Low-code is essentially a type of "development language," not merely the assembly of pre-built components. Low-code technology in process management is relatively mature, supported by well-established modeling languages. Therefore, low-code technology can effectively achieve its value within the ITSM domain. It helps address challenges such as the surge in process volumes and frequent changes in management demands, thereby better supporting the digital transformation of IT service management. As a flexible ITSM tool, low-code technology can significantly reduce the cost of customizing IT service management processes. As a flexible ITSM tool, low-code technology can significantly reduce the cost of customizing IT service management processes.