EVALUATING THE EFFECTIVENESS OF LOW-CODE PLATFORMS IN HIGHER EDUCATION: A CASE STUDY

ABSTRACT

In today's rapidly evolving technological environment, universities need to adapt and integrate digital solutions into educational processes. However, conventional traditional development requires significant resources, while low-code platforms allow to develop at minimal cost. The purpose of this article is to analyze the current level of awareness and experience of using low-code platforms, evaluate their effectiveness in use at universities in Kazakhstan.  As part of the research, an online survey was conducted and a prototype was developed using a Mendix to demonstrate effectiveness of low-code platform in solving the identified problem. Despite the active development and distribution of low-level platforms abroad, this is still at an early stage in Kazakhstan, and many users are unaware of their use, even when interacting with BPMN-based systems, some of which are automated using low-code platforms. The study highlights the need for further research and wider adoption of low-code platforms.

АННОТАЦИЯ

В сегодняшней быстро развивающейся технологической среде университетам необходимо адаптировать и интегрировать цифровые решения в образовательные процессы. Однако традиционная разработка требует значительных ресурсов, в то время как платформы с низким уровнем кодирования позволяют разрабатывать с минимальными затратами. Целью данной статьи является анализ текущего уровня осведомленности и опыта использования платформ с низким уровнем кода, оценка их эффективности при использовании в университетах Казахстана.  В рамках исследования был проведен онлайн-опрос и разработан прототип с использованием Mendix, чтобы продемонстрировать эффективность платформы с низким уровнем кода в решении выявленной проблемы. Несмотря на активную разработку и распространение низкоуровневых платформ за рубежом, в Казахстане это все еще находится на ранней стадии, и многие пользователи не знают об их использовании, даже при взаимодействии с системами на базе BPMN, некоторые из которых автоматизированы с использованием платформ с низким уровнем кода. В исследовании подчеркивается необходимость дальнейших исследований и более широкого внедрения платформ с низким уровнем кодирования.

Keywords: low-code platforms, Low-Code Development Platforms (LCDP), productivity evaluation, higher education, prototype development, case study.

Ключевые слова: low-code платформы, платформы разработки с низким кодом (LCDP), оценка эффективности, высшее образование, разработка прототипа, практическое исследование.

1. Introduction

In today's world, where technologies and information systems are developing rapidly, universities need to adapt to changes and implement them in educational and administrative processes. However, conventional traditional development requires a lot of resources like time, technical knowledge, and money. And low-code platforms allow you to develop applications and systems at minimal cost, increasing efficiency and effectiveness, and also allowing users to participate in the process [1], [2], [3].

After OutSystems released a free version of the LCNC platform for developers in 2014, minimizing the amount of code and reducing development costs became a trend. The CEO of the Github platform noted in 2017, that the future of programming is the absence of programming at all. Modern CNC platforms have a wide range of functionality, which makes them more convenient than traditional development methods [4].

The purpose of this article is to analyze the current level of awareness and experience of using low-code platforms in Kazakhstan, as well as, based on international research and experience, also on a local practical experiment, to study their effectiveness and evaluate the feasibility of using LCNC platforms in the educational environment, especially in universities.

As part of the study, online surveys were conducted among students, teachers, and company employees. Based on the obtained data, it is possible to assess the level of awareness of low-code platforms, familiarity with the tools, as well as to identify barriers and expectations, and form an idea of the current perception of low-code platforms. In addition, the effectiveness of low-code platforms was demonstrated through the development of a prototype designed to address the identified problem after conducting a survey among students.

The article consists of four parts: the first one provides an overview of international experience in using low-code platforms, the second one analyzes the results of the conducted surveys, the third describes a local practical experiment involving the use of a low-code platform to solve the identified problem, and the final section presents conclusions and recommendations on the use of LCNC platforms at universities in Kazakhstan based on an analysis of the data obtained and international experience.

2. Background

In 2014 an American market research company Forrester defined the terminology "Low code development platform (LCDP)". Since then, low-code platforms have been developing and industry giants like Google, Microsoft, Siemens and others are starting to release their solutions. For example, in 2016 Microsoft released Power Platform, in 2018 Siemens released Mendix and, finally, in 2020 Google released AppSheet. During the Covid-19 pandemic, the process of adoption LCNC platforms accelerated significantly, and since then, low-code platforms have been expanding and developing even more, offering more opportunities to users [4].

In traditional development, that is, in code-based technology, it is necessary to write code manually, thereby requiring good technical skills from the developer, more resources, like time and money. Unlike traditional development, low-code platforms allow professionals without an IT background to design and build complex systems using their domain knowledge. They offer a graphical interface that allows you to drag and drop components to automatically generate full-featured software products, resulting in faster development and higher productivity [5]. They are based on the principles of model-driven design(MDD) and visual programming, and almost anyone can learn this [6].

According to the results of the study, nowadays more than 200 types of LCDPs are currently available [7]. Low-code platforms are expanding more actively in the market every year. Forrester estimates that the total market for low-code development platforms is growing by 50\% every year (see Fig. 1). And Gartner predicts, due to continued demand for applications and a shortage of skilled developers, that low-code development tools will be used for most application development by 2024 [6]. Forrester also predicts that by 2028 the low-code market may reach \$50 Billion [8] .

Figure 1. Low-code platforms forecast (US$ billions)

Thus, the rapid development and spread of low-code platforms indicates that the market is increasingly striving to simplify digital transformation.

3. Research Methodology

The methodological approach aims to assess the current level of awareness and use of low-code platforms, also to evaluate their potential effectiveness in the educational environment.

Due to the lack of ready-made datasets on the use of low-code platforms in universities in Kazakhstan, primary data collection was conducted through our own questionnaire surveys among students, teachers, and university staff. The survey was conducted online using structured questions, which allowed us to group the answers received. The sample included more than 100 respondents from different fields, which provided a variety of data. The collected data made it possible to analyze and identify the current level of awareness about low-code platforms, experience in their use, as well as identify needs and barriers to implementation.

In addition, a separate survey was conducted among more than 30 students in order to identify certain problems in the educational process. When analyzing the results, a common problem was identified that was related to system overload during registration for courses, when university learning platforms often experience performance problems due to high user traffic. This conclusion was used as a basis for further practical experiments.

To solve the identified problem, a prototype solution was developed using the low-code platform Mendix. The prototype implemented basic functions such as course selection and schedule formation. As a result, it was revealed that the development process by low-code platform required significantly less time and technical effort compared to traditional approaches.

Thus, combining all the approaches described above made it possible to comprehensively analyze the topic, draw conclusions, and justify recommendations for further development.

4. International Use Cases Of Low-Code Platforms

This chapter examines international examples of the use of low-code platforms, as well as justifies their effectiveness.

LCDPs are used for different purposes all over the world. Due to the fact that deep IT-skills are not required to work with them and they require fewer resources, such platforms are used even by scientific researchers, who do not have programming skills, but need technical tools to conduct research. The meta-research, which covered more than 70 scientific papers, identified the main areas in the world, where low-code platforms are actively used:

• Software development and application development
• Manufacturing industry and logistics systems
• Business process in manufacturing
• Digitization of process
• Disaster management (tornadoes, landslides, etc.)
• Social media analysis
• Global news analysis
• Industrial engineering education
• Supply chain management
• Education for students [4]

The digital transformation in education began a long time ago, but we can say that the Covid-19 pandemic was a turning point and had a negative impact on the education system, forcing the termination of full-time education. When the COVID-19 pandemic began worldwide, about 1.8 billion students were affected by the closure of educational institutions and the cessation of full-time education. And in the field of education, during the digitalization process the low-code platforms were useful and used in many educational institutions such as Harvard, University of Pennsylvania, Georgia Tech Institute, Bentley University, Florida Gulf University, Boston College and so on [9].

Research has also been conducted, that examines the use-cases of low-code platform implementation in educational processes. One example is a project implemented at Georgia Tech University, which combined application development and design in order to provide students with the opportunity to gain practical experience. A team of students developed a software solution to support Georgia Tech's Interdisciplinary Capstone Design (IDCD) program. These applications were created to simplify the formation of teams, the selection of projects, the organization of the final exhibition, providing transparent communication and automating the main stages.  In 2013, these solutions was designed using Django, a Python-based web framework, and Ruby-on-Rails, a web framework written in Ruby. And in 2019, development is moving to a low-code platform named Blue Platform. According to the results of this study, low-code platforms provided an average of 76% faster development. Also, as a result, the program returns information from the database 10 times faster, and provided users with 43.8% more features [10].

Low-code platforms allow you not only to create tools from scratch, but also to integrate with existing systems, expanding their functionality. The Covid-19 pandemic has had a significant impact on medical education, especially the study of pathology. Since students in the field of medical education study a lot of confidential data, that cannot be freely distributed, and their study format differs significantly from the classical one, which is why standard solutions did not meet all the requirements. In one of the cases, the low-code platform named PowerApps was used to expand the capabilities of Microsoft Teams, as a result of which they were able to create a learning environment that fully meets the needs of medical students during the pandemic [11].

Also, one of the numerous studies analyzed data collected from platforms like Reddit and Stack Overflow, which identified the advantages of LCDP. The advantages of fast development, ease of learning and use, and resource savings were repeated the most. And according to the report of the survey conducted by The State of Application Development (OutSystems, 2019), where more than 3,300 IT professionals were interviewed, the main advantages were noted as acceleration of digitalization/automation and increased efficiency, while a lack of knowledge and limited functionality were identified as barriers to implementation [12].

If we look at experiments comparing the productivity of low-code platform development with traditional software, we can see that low-code platforms outperform, since traditional development required 888 hours of work, and LCDPs required only 47.5 hours of work, reducing the required effort by 94.63%. As for complexity, 12310 lines of code were written in traditional development, while the LCD approach resulted in 1723 lines of code, which is 86% less. This is very effective and relevant for a business organization [13].

Thus, based on the examples given, it is possible to identify key areas of application of low-code platforms, cases of successful solutions to digitalization problems, and the proven effectiveness of such solutions in the educational and scientific environment.

5. Analysis of Survey Results

This section presents an experimental stage of the study - the analysis of data collected by conducting an author's questionnaire surveys.

The first questionnaire was structured in 4 key areas:

• Having practical experience working with low-code platforms
• Perception of their advantages
• Existing barriers to use
• Willingness to use or implement them.

Based on the first survey results, it turned out that many are familiar with the concept of low-code platforms, but only a few have used them in practice (32.4%). The respondents faced the problem of automation and digitalization, especially during the Covid-19 pandemic, like the rest of the world. Back then, in most cases, it was solved using ready-made online services or tools, such as Google Forms, Microsoft Teams, etc. Tilda, Make and BPM systems turned out to be the most widely used low-code platforms in companies. It is worth noting that the level of programming skill that the respondent attributed to himself correlates with their level of awareness about low-code platforms (see Fig. 2). At the same time, this correlation did not affect the levels of practical application of this platform.

Figure 2. Correlation of programming skill (x-axis) with level of awareness about LCDP (y-axis)

It is also worth noting that the majority of survey participants are aware of the possibilities and recognize the barriers to using and integrating low-code platforms, but despite these limitations (limited functionality, scaling difficulties, security problems), 77.2% of respondents are ready to master these tools and apply them in the future.

Based on the collected data, it can also be concluded that many survey participants, although they use low-code platforms, often do not realize this. This conclusion was confirmed by additional analysis. In particular, many large companies in Kazakhstan actively use BPM systems, which is a fundamental concept in business process automation, and one of the tools of this system is BPMN, which is used to model business processes. At the same time, many BPMN models are implemented by low-code platforms that allow dragging components and elements to automate processes without requiring programming. We can see this in platforms like Camunda, Kissflow, Creation [14], [15].

Therefore, we can conclude that many business companies in Kazakhstan are already implementing low-code platforms to automate their business processes, but many users do not realize this, even if they use it. Moreover, as a result of a search on the use of low-code platforms in Kazakhstan, both in the field of education and in the field of business management, only a couple of articles were identified, which means that little or no research was conducted. This indicates that the topic remains insufficiently studied and is not widely used in the domestic scientific and practical environment.

Thus, it can be concluded that low-code technologies in Kazakhstan are still at the stage of initial development, and it is necessary to both expand and explore in depth. It is also important to raise awareness among users and organizations about the advantages and opportunities of low-code platforms in order to encourage their wider and more informed use.

The second questionnaire was structured in 2 key areas:

• Identification of challenges in the educational process
• Assessment of the performance and usability of existing university systems  

Also, the second survey was conducted to identify some of the difficulties that arise when using the university's educational platform. Many students who used popular learning platforms like Moodle, Platonus, etc. replied that they did not encounter any difficulties when using it. However, many students whose universities have developed their own local learning platform have indicated some difficulties that may arise when using it. The most frequent of them was revealed as the system performance degradation during course registration and schedule formation, caused by a high number of concurrent users. And many students prefer to transfer this logic to another site to unload the system as a solution to the problem (see Fig. 5 and Fig. 6).

Figure 5. Registration issues depending on system type by analyzing results

Figure 6. Most preferred solutions based on registration issues

Based on the analysis of the collected data, we came to the conclusion that the website freezes when a large stream of students enter during the registration period for disciplines and the formation of schedules is one of the most common problems for students. It is worth noting that this was indicated only by those students who form their own schedule at the university and the educational platform for this was developed locally by the university. Of the 31 students surveyed, 41.9% replied that they use a ready-made platform like Moodle or Platonus, while the remaining 58.1% indicated that the educational platform was developed locally (see Fig. 7). Although the sample size is limited and results may vary with a larger dataset, this study identified a clear problem related to system performance during peak usage periods. Based on these findings, a prototype solution was developed using the Mendix low-code platform. The prototype aims to address the identified issue and was used to evaluate the effectiveness of low-code development compared to traditional approaches.

Figure 7. Local university systems vs ready-made educational platforms

6. Practical Experiment Using a Low-Code Platform

According to the survey results, one of the key problems identified was the overload of the system during registration for courses and scheduling. When a large flow of students arrives during scheduling, the system becomes overloaded and productivity may decrease, after which it was decided that transferring to another site to unload the main site would be the right approach. Therefore, the Mendix was used to create a prototype, that includes basic functionality such as course selection and schedule formation. It was designed as a lightweight solution that could potentially reduce the load on existing university systems and demonstrate the effectiveness of low-code platforms (see Fig. 8).

Figure 8. Development process of the solution prototype in Mendix. (a) List of courses page in development. (b) MySchedule page in development. (c) Microflows developments. (d) Settings of the components. (e) After publication of process

The Figure 8 shows that in Mendix, it is possible to develop without hard effort and without requiring a certain technical level, as well as a simple prototype or a full-fledged application. Now let's see how long it takes in traditional development and compare them.

Table 1.

Comparison of the time of traditional development and development on low-code platforms

As a result of this experiment low-code development was approximately 3-4 times faster than the traditional approach (see Table 1).

7. Discussion

Based on the results of the survey and data analysis, it is possible to conclude about the trends in the use of low-code platforms in Kazakhstan. Although low-code platforms are increasingly capturing the market in the world, the experience of using them in Kazakhstan is limited. We can also see that despite the fact that many BPM systems have integrated low-code platforms for automating business processes, there is limited research on this topic and it is not widespread among users as expected.

However, it is important to note that many of the respondents were well acquainted with the capabilities of low-code platforms and indicated barriers to implementation, as well as difficulties. So 72.2% of respondents are ready to use and implement them.

In the chapter of practical experiment, it was observed that development using a low-code platform can be more efficient than traditional development in certain situations and can solve certain problems. Developing on low-code platforms makes it possible not only to develop simple prototypes or integrate with an existing application, but also to create full-fledged applications. Low-code platforms can effectively solve existing problems at universities and can make life easier not only for students, but also for university staff.

In the chapter of collected examples of international application experience, we can see that this is a common experience abroad, and according to the 2020 Grandview Research, LCNC platforms are expected to grow by 22.7% in CAGR from 2020 to 2027 and break through $21B in 2021. It is predicted that in 2027 there will be a $86.9B in industry with close to even split between On Premise and Cloud based LCNC platforms [16]. This is how we can justify the effectiveness of using low-code platforms and disseminate their research on the platform. Also it is important to highlight the need to conduct more research and increase local datasets.

It is worth noting that low-code platforms do not replace traditional development, but complement it, and each development method has its own advantages.

8. Conclusion

In the context of the rapid development of digital technologies, universities need to adapt to changes and implement them in educational processes. However, conventional traditional development requires a lot of resources such as time, technical knowledge and money, while low-code platforms allow you to develop applications and systems at minimal cost, increasing efficiency and allowing users to participate in the process  [1], [2], [3], [17].

Modern low-code platforms have extensive functionality, which makes them more convenient than traditional development methods. This includes cloud-based development and a user-friendly interface where you can drag and drop components and create your own program [4].

The purpose of this article is to analyze the current level of awareness and experience of using low-code platforms, as well as, based on international research and experience, to study their effectiveness and evaluate the feasibility of using low-code platforms in the educational environment of Kazakhstan, especially in universities. 

As part of the study, online surveys were conducted among students, teachers, and company employees.

A prototype solution was also developed to demonstrate the effectiveness of development on a low-code platform.

Despite the active development and distribution of low-code platforms abroad, it is still at an early stage in Kazakhstan.

Despite the fact that companies use BPMN models, some of which are automated using low-code platforms, many do not realize due to lack of awareness and dissemination of information. We came to the conclusion that it is necessary to conduct more research and spread the use of low-code platforms in order to follow trends and gain more efficiency in development.

However, it is worth noting that low-code platforms do not replace traditional development.

The limitation of this work is the narrow range of respondents, as well as the lack of local datasets and experiments.

References:

1. S. Shridhar, “Analysis of low code-no code development platforms in comparison with traditional development methodologies,” International Journal for Research in Applied Science and Engineering Technology, vol. 9, 2021.
2. C.H. Wang and K. C. Wu, “Interdisciplinary learning of low-code development platform programming with dual coding theory-a case study of agilepoint nx,” Journal of ICT, Design, Engineering and Technological Science, vol. 6, 2022.
3. D. Gao and F. Fagerholm, “Measuring end-user developers’ episodic experience of a low-code development platform a preliminary study,” E-Informatica Software Engineering Journal, vol. 18, 2024.
4. F. Sufi, “Algorithms in low-code-no-code for research applications: A practical review,” 2023.
5. A. Trigo, J. Varajao, and M. Almeida, “Low-code versus code-based software development: Which wins the productivity game?” IT Professional, vol. 24, 2022.
6. B. Lefort and V. Costa, “Benefits of low code development environments on large scale control systems,” International Conference on Accelerator and Large Experimental Physics Control Systems, 2019.
7. S. Hinrichsen, A. Nikolenko, K. L. Becker, and B. Adrian, “How to select and implement a suitable low-code development platform,” in Human Systems Engineering and Design (IHSED 2023): Future Trends and Applications, vol. 112, 2023.
8. Forrester. (2024) The low-code market could approach $50 billion by 2028. Accessed: 2025-05-11. [Online]. Available: https://www.forrester.com/blogs/the-low-code-market-could-approach-50-billion-by-2028/
9. F. Khosrojerdi, H. Motaghi, M. Yusof, A. Razak, and U. T. A. Razak, “The application of no-code/low-code platforms in the sector of education,” Tech. Rep., 2021.
10. G. A. Ellis, N. Mulka, and A. S. Jariwala, “Rapid development of software solutions to enhance course infrastructure and the educational experience of student developers,” in ASEE Annual Conference and Exposition, Conference Proceedings, 2021.
11. A. Rajaram, C. Olory, V. Leduc, G. Evaristo, K. Cot´e, J. Isenberg, J. S. Isenberg, D. L. Dai, J. Karamchandani, M. F. Chen, C. Maedler-Kron, and P. O. Fiset, “An integrated virtual pathology education platform developed using microsoft power apps and microsoft teams,” Journal of Pathology Informatics, vol. 13, 2022.
12. P. M. Gomes and M. A. Brito, “Low-code development platforms: A descriptive study,” in Iberian Conference on Information Systems and Technologies, CISTI, vol. 2022-June, 2022.
13. D. Aveiro, V. Freitas, ´Erica Cunha, F. Quintal, and Y. Almeida, “Traditional vs. low-code development: Comparing needed effort and system complexity in the nexusbrant experiment,” in Proceedings - 2023 IEEE 25th Conference on Business Informatics, CBI 2023, 2023.
14. G. S. N. Artyk, “Improving the efficiency of banking operations through automation: A modeling approach,” 2024. [Online]. Available: https://journal.iitu.edu.kz
15. Y. S. Dharmawan, G. G. Divinagracia, E. Woods, and B. Kwong, “Interdependencies on bpm maturity model capability factors in deriving bpm roadmap,” in Procedia Computer Science, vol. 161, 2019.
16. M. Kulkarni, “Deciphering low-code/no-code hype – study of trends, overview of platforms, and rapid application development suitability,” International Journal of Scientific and Research Publications (IJSRP), vol. 11, 2021.
17. K. Rokis and M. Kirikova, “Exploring low-code development: A comprehensive literature review,” Complex Systems Informatics and Modeling Quarterly, vol. 2023, 2023.