Ongoing attempts to drive innovation in the financial markets have been undertaken both within and outside of VI Company. In my previous article, I explained our contribution and how in- and extrinsic needs lead to innovation.
In this two-part article, I share the details of our fundamental research into low-code. I'll provide you with insights into our approach, experiences, and the final results. The second part of this article will be published in January 2021.
Determining the approach
During our conversations with clients, we noticed a growing interest in low-code. Some of them had already initiated low-cow experiment within their own organization. We decided that we should embrace this new development, but where to start?
Together with a group of interested colleagues, our innovation lab kicked off the initiative for research into low-code. As usual, we started by setting the goal of our initiative: “Learn if, and how, VI Company and their clients could benefit from using low-code in projects”. Having a measurable goal such as this is key to eventually conclude the success or failure of a research project.
Next, we defined a fitting approach. This helps to provide structure along with the project and monitor progress:
- Determine our version of low-code, its definition, and the scope of the research project.
- Conduct field research by gathering information and specifications for selected low-code platforms on pre-defined topics.
- Set up a scorecard by using objective & measurable criteria, and quantify the selected platforms.
- Conduct empirical research and gain experience by prototyping with a select number of alternatives.
- Gather results and draw up our final verdict.
Determining the scope
In my recent article about low-code, I mentioned that low-code is widely used and often misunderstood. Companies interchangeably use the term with “no-code”; therefore, it was important to have a clear distinction between the two.
“A low-code development platform (LCDP) is software that provides a development environment used to create application software through graphical user interfaces and configuration instead of traditional hand-coded computer programming.”
Definition as found on Wikipedia.
Alongside using this definition, we also set up some criteria to better separate low-code and no-code solutions:
- Low-code platforms oftentimes require knowledge or experience with programming. No-code platforms do not require any such thing.
- Low-code platforms often offer more options and functionality to customize applications using programming knowledge or experience. No-code platforms are restricted to using drag-and-drop interfaces and forms in order to create a solution.
We decided to focus solely on low-code platforms because our knowledge and expertise with full-code would have a complementary effect, resulting in better solutions for our clients.
Mapping the low-code landscape
After proper preparation, the time had come to finally start putting our hands to work. To gather information about the available alternatives and quantify them at a later stage, we created a list of topics upfront:
- Pricing and licensing
- Technical match (with VI Company)
- Effort and learning curve
We selected 13, more and less known, platforms that complied with our definition of low-code and started gathering information on all these topics.
The applications of low-code
It is important to know that low-code is often used for three major types of applications.
1. Process Automation
This is used to schedule or trigger background processes. Depending on the outcomes of a process, it could trigger other events or processes, eventually resulting in a circuit of connected services and applications. This often makes it easier to automate business processes.
2. Extract, Transform & Load (ETL)
Low-code platforms that serve this purpose try to make it easier to consume data from many standardized data sources, transform the data, and distribute the result (think about databases, spreadsheets, and APIs). When building custom software, the integration and maintaining of many data sources can take up a lot of time and can be fragile. For example, if anyone changes a field in a spreadsheet, it might require developers to update the application and deploy a new version.
3. Rapid Prototyping
Platforms focused on this purpose often have the functionality to aggregate datasets, design interfaces, and deploy the application to multiple platforms (Google Play Store, Apple App Store). This could significantly decrease the time-to-market of a new application, especially if you are only trying to verify your business case.
The diagram above shows the 13 low-code platforms we selected for our research, categorized per application. Some of these platforms even offer functionality for multiple applications, such as the Microsoft Power Platform.
Hopefully, you have learned more about the definition of low-code and its different applications today. In the next part of this series (to be published in January 2021), I will take you along the next stage of this project and explain how we gathered information on each of these technologies and were able to compare them objectively. So, stay tuned for more!
Are you curious to learn more about low-code? Our solution engineer Kevin Wareman is happy to tell you all about it. You can contact him via firstname.lastname@example.org.
Fundamental research into low-code pt. 2
Stay tuned! Part 2 of this article will be published in January 2021.