Low Code/No Code or No Go?

An In-Depth Examination of Low Code/No Code and Artificial Intelligence

About the Author: Jason is an Investment Fellow at Thornapple. He brings both operating and investing experience through his tenure at several tech companies as a data scientist and across a variety of venture capital firms as an investor. Prior to this, he was an MBA VC Associate at Sweater Ventures, where he invested in early-stage consumer-facing startups. Jason earned his MBA/Ai from the Kellogg School of Business and McCormick School of Engineering at Northwestern University. Personally, Jason is a big fan of basketball, and his NBA team is the Denver Nuggets.

Abstract

For years, businesses followed one of two paths for application development: buy them ready-made from a third party or build them from scratch. Each of these has its own downsides as well from costs, inefficiencies, and time. Today the advent of low code and no code (LC/NC) development and solutions are replacing the traditional software development paradigm. These solutions no longer require an outsized team of technical engineers to create software tools after months of development. LC/NC solutions, however, have their own set of downsides that have prevented them from universal adoption. That very much might change with advances in artificial intelligence that will then lead to the proliferation of these solutions. Because of this, founders and investors should look to LC/NC solutions for SMBs as these businesses will lead the first wave of adoption.

How big is the LC/NC market?

As shown below, the market for LC/NC solutions is expected to continue to grow at 19% and 24% respectively.

As Grand View Research states, the market size for low-code solutions will hit $35B from $6.78B by 2030. While Emergen Research predicts that no-code solutions will grow from $12.17B to $68.05B by 2028. Lastly, Base10 has listed out several current players in both spaces:

What is even low code or no code?

Given that less than 1% of the world knows how to code but business demands and software needs continue to increase, companies look to fill these gaps in more efficient ways. They have thus turned to low code and no code solutions. Gartner even estimates that 70% of application development will be done through LC/NC platforms by 2025.

So first off, I’ll explain what exactly low code and no code entail and compare them to full-code, or traditional coding.

Low Code

Definition: In this type of development, low code solutions utilize intuitive graphical interfaces that allow users with some degree of technical coding knowledge to be able to make applications easier and faster.

Benefits:

• Timesaving: They speed up the timeline for application/software development.
• Productivity gains: Engineers have more availability to focus on technical and pressing engineering matters.
• Cost reductions: Low code solutions allow companies to work with the staff they have in a shorter amount of development time.

No Code

Definition: Meanwhile, no code solutions are more drag-and-drop user interfaces that allow any user, regardless of their level of coding knowledge, to develop applications. Due to their plug-and-play nature, no code created applications tend to take less time to develop than low code created applications.

Benefits: In addition to the benefits that low code solutions have; no code solutions also have the following other benefits.

• Fill talent gaps: They allow for “Citizen developers” or non-technical employees to be able to also create applications without having to rely on engineering teams.
• Increased Functionality: Developers have more ability to fine-tune or configure the plug-and-play functionality for their no code solutions.

Full Code

Definition: In traditional software development, full code entails software developers and engineers with deep technical knowledge utilizing specific programming languages to create software applications after a long development cycle.

Benefits:

• Full Control and Customization: Due to this, the engineering teams have full control and customization over the solutions they build from scratch.
• Security: With full control of the code and no third-party involvement, there is less likelihood of vulnerabilities in the platform.

For the full code benefits, however, speed is often lost. For example, most applications take at least 4–6 months to develop alone. So, you’ll see that LC/NC solutions offer time gains and cost savings, but as research from the No-Code Census found, they also can increase productivity by about 4.6 times as much as full code solutions.

Now that you’re familiar with software development, I’ll answer when to use one over the other. When full control, customization, and security are needed, businesses should continue with the traditional development approach. If, however, businesses are more resource constrained, look to NC/LC solutions. Between those, the target users and specific use cases are important considerations. For more integration-heavy use cases, low code solutions are generally recommended over no code solutions.

The Case for No Go

Now you may be wondering, “All that sounds great so then why have NC/LC solutions not been universally adopted?” Well, currently, there are some headwinds and core issues that have slowed adoption.

For instance, developer tools are “programs that allow developers to create, test, and debug software” and they’re making coding easier, faster, and more accessible. Examples of developer tools include Git for code documentation and collaboration or Visual Code Studio for code compiling. Additionally, with the rise of AI, GitHub’s Copilot autocompletes code for developers or Sourcegraph’s Cody, an in-editor coding assistant that utilizes their codebase repository. And so, AI-assisted software development will also make developers more productive and allow companies to make do with less.

Combine those headwinds with the following core issues to LC/NC solutions and you’ll see why adoption has stalled:

• The Shadow IT Phenomenon: This phenomenon occurs when citizen developers create applications from LC/NC development that end up not working well or scaling properly. In turn, company IT departments are left to fix the applications or there is no continuity if the creator leaves the company after application development. Without sufficient oversight into the creation of these applications, there also is the risk of having a parallel infrastructure.
• Scalability: In a similar vein, these applications are usually not built with scalability in mind. They often require integration with other services or data sources leading to breakage or engineers needing to be involved. That’s also seen in code errors. In terms of error rate, LC/NC solutions have higher rates than full code as research from Coimbra Business School finds. In addition to code error, there is also user error to consider. For proper usage and app creation, user training must be completed and that is difficult at an enterprise-wide level.
• Limited Customization or Flexibility: In both LC/NC solutions, often the end-to-end process is dictated by templates and pre-built drag-and-drop options. This leaves users constrained without the ability to build more complicated applications for specific use cases. Couple that with the fact that most users do not have a deep technical knowledge, these solutions are often seen as opaque black box systems.
• Security Concerns: Businesses run the risk of security vulnerabilities with the introduction of LC/NC solutions. Since the company does not maintain control of the code, they could be subject to security breaches from the LC/NC provider. As users continue to create applications and connect to other applications, they do not know the full security compliance around those other applications.

Universal Adoption of LC/NC with AI

As Jonathan Reilly points out, technology usually follows a set progression: “First, it’s used by a small core of scientists. Then the user base expands to engineers who can navigate technical nuance and jargon until, finally, it’s made user-friendly enough that almost anyone can use it.” We’re seeing this progression in AI adoption and we’re a bit further along that curve for LC/NC solutions. By combining these two technologies, AI can help to increase the LC/NC adoption progression by addressing each of the concerns mentioned above.

• The Shadow IT Phenomenon: For this core issue, AI can make it easier for users to troubleshoot and debug their own applications without having to bog down their IT department, create continuity and knowledge transfer by allowing for collaboration and documentation, and have built-in infrastructure checks to avoid over-taxation, duplication, or parallel systems.
• Scalability: On this front, AI can lead to the proliferation of citizen developers with scalable applications. AI would make applications easier to create, fix, integrate, and implement. For instance, considering that over 90% of any company’s data is unstructured or unlabeled, AI can automatically handle this for users. And on the error reduction front, AI can provide code assistance or code generation.
• Limited Customization or Flexibility: Imagine being able to enter a prompt and create an application based on that. With natural language processing, LC/NC applications would no longer be constrained by pre-built templates and could instead be prompted to create anything for any use case. We already beginning to see this with autonomous ai agents like Auto-GPT.
• Security Concerns: As I wrote about previously, AI is great at pattern recognition and can thus be used for anomaly detection or potential security breaches. Much like AI-enhanced smart contracts, AI can continuously monitor and track LC/NC applications to ensure security compliance and integrity.

With AI working on these core issues and its continued advancement, I see Gartner’s prediction being a conservative estimate and that we’ll reach universal adoption where the everyday person is empowered to create applications like never before.

Advice for Founders/Investors

Before we reach that point, there is plenty of opportunity to capture some of that value creation. To get in on the action, I think the way to reach that democratization of AI and LC/NC solutions is by first targeting SMBs. In my opinion, enterprises are not as appealing because they have dedicated teams who are knowledgeable, and they are slow to adopt enterprise-wide systems/applications. While SMBs hold actionable data, want to iterate and move fast, are time constrained, and have talent gaps or employees without deep technical knowledge.

So, if you’re a founder with a small engineering team, keep an eye out for these solutions to make your engineers more productive or to create citizen developers within your company. Or if you’re looking for your next startup idea, think about creating an AI-enhanced LC/NC solution catered to SMBs.