AppGen Is an Existential Threat to the Enterprise App Business

"Nobody wants to admit that their own death is coming soon."

Low-code has been swinging the pendulum away from off-the-shelf applications and toward custom development for years.

There are good reasons for this. When practical, fit-to-purpose software is best. And the lower cost, risk, and lead time of low-code development — coupled with an expanded developer pool, easier integration, management of apps on a common platform, leveraged licensing, etc. — makes it much harder to justify off-the-shelf software licenses and vendor sprawl.

AI-powered enterprises will "build" software instead of "buy" it — and many applications in enterprise portfolios will consolidate onto low-code AppGen platforms.

Until recently, this shift was typically unplanned and organic. Even as firms scaled low-code and told us that they were buying fewer apps and building more instead, they were often surprised to realize that their practices and view of "build vs. buy" had changed. It had just sort of happened along the way.

But now, this "build first" mindset is becoming a deliberate enterprise strategy. Here is a sample of comments we've received from enterprises over the last few months:

"We're freezing all new app purchases. We start by developing [on low-code platforms]." (CIO, North American energy enterprise)

Related:Quick Reference Guide for Software Development

"For new applications, our recommendation is [low-code platform] first." (IT director, global engineering firm)

"AI tilts the floor. We see a move away from the big enterprise apps. You've got a lot of single-purpose SaaS tools that are expensive when you put them all together. Some of those will collapse into one low-code platform." (Partner, global consultancy)

This last quote hits to the heart of the trend. Advancements in low-code and development practices already made the "build first" and "platform consolidation" strategy unavoidably practical. But it's generative AI — and its killer use cases in TuringBots and low-code AppGen platforms — that has served as the accelerant for more firms to recognize these conditions and embrace them.

AI-powered AppGen platforms will drain the competitive "moat" of domain knowledge encoded in off-the-shelf business apps.

There are two benefits of genAI in software development that tip the scales: 1) even more speed and ease throughout the SDLC (self-evident) and 2) the infusion of business and industry "domain knowledge" through AI models into the development act.

This second point is monumental. The typical remaining "moat" for many business application vendors is the "domain knowledge" and "industry best practices" encoded in their off-the-shelf software. AppGen will drain this moat. Even a vanilla large language model knows what a CRM is and how it's put together, or what a truckload shipment process looks like, or what the airspeed velocity of an unladen swallow is. And AppGen platforms make this domain knowledge instantly available in the development act. This means you can ask the platform for the app you need and get it — like the gentleman we interviewed who generated an app for managed sea containers and their documentation. He marveled that the platform knew "his" industry!

Related:AI-Assisted Coding: What Software Developers Need to Know

Where's all this going? Over the next several years, these factors will lead to market consolidation as enterprises retire many of the apps in their portfolios (both off-the-shelf and custom) and replace them with bespoke, dynamic applications delivered using AI on low-code AppGen platforms.

True story: a frank conversation with an enterprise software vendor.

There are caveats to this prediction. Some specific app functionality is too high-risk and legally bound to be done custom by the typical firm (e.g., general ledger), some app vendors will become AppGen platforms themselves, some apps have legitimately differentiated technology that's not easily replicated, and so on.

But the many applications of the business world, which are basically collections of the same generic, fungible software components rearranged into different industry and use case patterns, is clearly under threat. And the vendors know it.

To illustrate: Several months ago, we interviewed a leader at a significant software vendor. This vendor's flagship product is an application in one of the major "three-letter acronym" enterprise software categories (such as ERP, CRM, HCM, etc.), which from here on we will refer to as "app." In our discussion, he said:

"Fast-forward five years. Building an [app] is going to be very easy. Half a dozen prompts, and something will work for you, and it's going to be very specialized to your use case. So what is the value of our own [app] product? Or anyone's [app] product for that matter? In 10–15 years, people won't be buying our software. We might not even be slinging [app] anymore … that product could go to zero; we're not going to be pulling money that way. People will be accessing that functionality through different mechanisms. There has to be the next-level step of where the value is going to be provided."

We agreed. Many trends in AI and software development point to it, and we had years of research backing it up. But outside our research, we'd never heard the point so boldly and clearly stated. So we asked, "Why are none of the software vendors talking about this?" His response:

"Nobody wants to admit that their own death is coming soon."

— John Bratincevic, Principal Analyst

This article originally appeared on Forrester's Featured Blogs.