Can no-code be used for enterprise-level applications?

My most recent visit to my tech garage was to develop a web application for product managers based entirely on the no-code tool bubble.io and some backend tools. I wanted to test the hypothesis of whether a single founder could launch and scale a product without code and turn it into a profitable business.

While my experiment is still ongoing, I've already launched the product and a few dozen users have created accounts, some of whom use the product regularly, give me feedback, and I iterate based on their feedback. If you are curious, this is the link to the app: https://produkando.io.

I've seen firsthand how I can bring an MVP to market in a few months or less, at one-hundredth the cost - not counting the opportunity cost of my time - of bringing a product to market using the traditional approach in the past.

The fair and legitimate question

Is No Code a tool just for startups and founders, or can the benefits of faster time to market and lower development costs be leveraged in enterprises as well?

What got me curious was a 30-year-old video in which Steve Jobs at my alma mater MIT said that object-oriented programming was revolutionary at the time because it allowed companies to get products to market ten times faster than procedural software. For companies like banks, manufacturers, retailers and many others, this ability to get software to market ten times faster was a clear competitive advantage.

Thirty years later, is No Code a reinterpretation of object-oriented programming?

Some Clarifications

Before I answer the question, first an obvious clarification. No Code is a term that refers to a method of developing software applications in which no code is written. Ultimately, it is just another layer of abstraction. Many of the elements available in No-Code are the same as those available in JavaScript. The developer must enter many of the same parameters that are available in JavaScript, but does so by filling in fields instead of writing code.

This approach makes software development accessible to more people who have no experience in programming or software development. In addition, even tech-savvy people can develop applications much faster with no-code, even though they may find it difficult or strange to change the approach.

The powerful no-code tools have a steep learning curve, but compared to mastering a programming language, the effort is much less.

Let's review some relevant facts systematically

1. On the front end, a no-code tool like bubble.io offers very few limitations compared to using a programming language. The appearance of a web app is not significantly different from an app built with code, and virtually every single effect available in JavaScript (hovering, shadowing, ) is also available in bubble.io. There are countless plugins that offer pretty much everything from detecting user inactivity to integrating with Google Maps, Google Login, displaying a countdown, etc.?
2. The plugins are not limited to front-end functions. You can also integrate any kind of API, create endpoints to which you can send or retrieve data using authentication, run background tasks, integrate with an external database, use two-factor authentication, etc.
3. You can create different development versions, similar to branches, and more than one developer can work on a Bubble application at the same time. And, of course, there's a development system and a production system, and you can also set up a staging system if you want.

Of course, there are limitations. The most noticeable ones relate to scalability and performance:

1. Development is mostly screen-by-screen, but there are back-end functions that are not screen-based. That means it's harder to have centralized components or functions, or to copy functions from one screen to another. It's definitely more labor intensive than copying code.
2. Performance is slower than code-based applications, especially at the lower pricing tiers. At the higher tiers, you can have your own dedicated server, but the extra layer between no code and code implementation inevitably takes a toll.

Let's now summarize the advantages and disadvantages.

1. In terms of time to market, traditional development cannot compete with No Code. You could assign a product manager with typical product management tasks and develop and design the app at the same time. The effort is lower because the overhead is smaller, as there is no need to communicate with a larger team.
2. Cost. No code is also orders of magnitude cheaper. You need fewer people for less time. And you can iterate much faster after launch, even after very extensive QA.?
3. Improved collaboration: because there is no code, developers, designers, and business stakeholders can collaborate and get more involved in software development. This helps speed up the process and ensures that everyone is pulling in the same direction.
4. New ways to organize work: you can assign engineers to more complex or mission-critical applications, while product managers develop tools without any precious engineering time at all.

Of course, there are some drawbacks:

1. Limited customization options: Any time you have a simplified system, you also have more constraints. You may not be able to customize the application down to the smallest detail, but if you just want something very specific, that's not a big problem.
2. Scalability issues: this is the biggest issue for me. The inherent logic of no-code products makes scalability difficult, both when you have many people working on the product and when you add more screens and features.No-code tools are often designed for small projects. As a result, they may not be suitable for large enterprise applications that need to scale.
3. Security issues: No-code tools can have security vulnerabilities that can put sensitive data at risk. It is important for the enterprise to carefully consider the security implications of using no-code tools.
4. Binding to and transitioning to code: No-code vendors lock you in and never release the application's code unless they go bankrupt. Transitioning from no-code to code may not be as easy. That said, you can do it in several steps, converting the backend first and then the frontend. It's not clear to me if you can convert one screen at a time. If that were possible, that would be a big advantage of no-code.

Conclusion

My conclusion is that no-code in the enterprise is suitable for many use cases, but not all. For extremely mission-critical or large-scale applications, it may be better to stick with the traditional approach for now.

However, that should not be an excuse to reject no-xode for every type of application. Companies could use it for internal applications first and then, as they gradually gain confidence, use it for non-mission-critical external applications. No-code isn't perfect, but it can accelerate the pace of innovation at the enterprise level in much the same way that object-oriented programming did 30 years ago.