About fifteen years ago, a new family of software products emerged: no-code software. Before the rise of no-code, there was only one flavor: the Commercial-Off-The-Shelf (COTS) application. Applications are software products designed for a specific use. No-code software, on the other hand, offers end users the ability to ‘configure’ an application themselves instead of programming it. As a result, no-code made developing applications much faster and cheaper.

If you’d like to read this article in Dutch, click here.

The advantages of no-code software used to create applications are:

• Tech-savvy users can create custom solutions quickly.
• It is flexible to modify by non-IT professionals.
• Configuration costs are generally lower than traditional programming. [1]

We use no-code for applications where the client does not yet have fully developed standards for its working methods and information management, and where much is still being learned as we go along. Such a situation calls for applications that can easily adapt to changing needs.

The disadvantages of no-code software are:

• If the process is standardized, flexibility offers no added value.
• Flexibility often leads to many different applications to manage. [2]
• Not very user-friendly, because no-code is a one-size-fits-all solution.

Configuring instead of programming

Every application always consists of different ‘layers’: storage, processing logic, and a user interface (UI). These layers are programmed. But while most (COTS) applications are programmed from the bottom up, no-code applications are not. In no-code software, part of what the software needs to do is ‘configured’ by a functional administrator (see Figure 1: No-code shifts the burden of customization from programming to configuring.). Think of it as making a mortgage calculation in Excel. Excel itself is nothing more than smart graph paper and not yet an application. Only when a ‘mortgage template’ has been created can it be used for that purpose. No-code software works similarly.

Figure 1: No-code shifts the burden of customization from programming to configuring.

In doing so, the no-code software vendor saves the customer a great deal of programming work. The customer can build any application they wish. However, configuring no-code software into an application naturally also costs time and money, not for the no-code software vendor, but for the customer or their advisor. The costs of the end product are shifted from the vendor to the customer and, in the case of a tender, from a product to a service. Because the no-code software vendor does not necessarily provide the service as well, configuration is often not viewed as ‘custom work’, even though it is. After all, a customer-specific application is being created.

Configuring no-code software is, therefore, effectively custom work and should be tendered as such not least because quality requirements and information security requirements should also apply to configuration work, just as they do to programming work.

Procurement of applications and no-code

When public authorities wish to procure an application, a clear distinction must be made between the procurement of:

1. a no-code software package, AND
2. the service to develop an application, OR
3. an off-the-shelf application that requires no customization.

To tender for an application without customization, the intended use must be highly standardized (a uniform working method), the coherence of information is clearly described in an information model, and the desired functionality, including UI, is specified in advance to ensure a level playing field (see Figure 2: Procurement of no-code or Commercial Off-The-Shelf.).

Figure 2: Procurement of no-code or Commercial Off-The-Shelf.

Due to laws and regulations, governments cannot replace existing no-code software with another provider without also tendering the configuration service (the customization) (unless the contracting party does so itself, of course). What is also not possible is replacing an application built on no-code software with an off-the-shelf product AND requesting no-code functionality.

No-code-agnostic customization

When governments wish to tender for no-code software or application configuration, it is advisable to provide the market with sufficient specifications. After all, requirements and wishes must not be tied to a specific no-code software supplier or a specific configuration.

Moreover, not every organization needs no-code software. The rule of thumb is: the more mature the organization, the more standards there are in working methods and information, and the less need there is for flexibility and therefore for no-code. The speed at which customization is created can still be a reason to choose no-code software as a basis.

To avoid vendor lock-in, it is essential to organize the following components in-house:

• Standardized information models for every no-code configuration,
• Knowledge and expertise regarding various no-code software [3],
• Open Standards for exchange with other software.

Let’s get in touch

Want to know more about no-code software or need help with procurement? Then do not hesitate to contact Daan Oostinga!

[1] With the rise of AI and the steadily declining costs of development, this argument will no longer hold water.
[2] See also the article: ‘How to avoid low-code and-no-code chaos?’
[3] For no-code solutions, we use tools such as WEM, Thinkwise, Mendix, Relatics, Datastorms, and Outsystems.