IT leaders are buying into the promise of low-code, no-code, and intelligent app-dev tech at a rapid clip. The trick, however, is making good on those investments. Credit: 1 Faced with a long-running shortage of experienced professional developers, enterprise IT leaders have been exploring fresh ways of unlocking software development talent by training up non-IT staff and deploying tools that enable even business users to build or customize applications to suit their needs. A broad spectrum of tools has arisen to facilitate software development in the enterprise, from no-code platforms like Bubble and low-code drag-and-drop tools, both stand-alone and integrated into enterprise applications, to intelligent tools that use machine learning to suggest lines of code to professional developers as they work. Sales of all three of these categories of tools are growing. IDC forecasts that sales of no-code platforms will grow at an annual rate of 13.9% through 2026, with sales of low-code platforms growing at 14.1%, and those of intelligent developer technologies roaring ahead at 31.3%. This last category has received a boost as platform vendors explore the potential of generative AI models such as ChatGPT to create boilerplate application skeletons on which developers can hang their own business logic — or even turn human-readable requirements into machine-readable code. The predictions about the future of software development are contained in IDC’s “Worldwide Low-Code, No-Code and Intelligent Developer Technologies Forecast, 2022-2026” report. Its author, Michele Rosen, says the market for intelligent developer tools has become even more interesting since she finished writing it, now that some of those tools — such as Salesforce’s Einstein GPT or Microsoft’s GPT-based Copilot — have become public, although even before that, products such as OutSystems’ AI Mentor were offering similar functions. Autocomplete on steroids “Think of them as boilerplate writers, or autocomplete on steroids,” Rosen says. “They are tools used by someone who knows how to do this themselves, who may be using them to supplement their knowledge about a technology, library, or framework they haven’t worked with before, or to avoid looking up a few lines of code on Stack Overflow.” Other uses for them might involve typing a few words as a prompt to generate the 20 lines of boilerplate needed to get a project started. “It’s really just a force multiplier, an accelerator,” she says. Low-code and no-code platforms, on the other hand, typically adopt a drag-and-drop metaphor rather than a command-line interface, and that shows up in the way line-of-business developers think about the problems they’re solving too. Users without a technical background will typically consider an app from the user interface inwards, she says: “That’s just the mentality that most people approach computing with.” If they are provided with UI components they can arrange to create the user interface, and then also components that can be assembled into business logic and even integrate with third party systems, then, in a sense, no-code and low-code development for the non-technical developer becomes a component based experience, she adds. That componentization is key, says Andrew Peterson, CTO of executive search firm Riviera Partners, a longtime user of low-code development tools. “One of the reasons I like low-code is because certain parts of your application are commoditized,” he says. “If I can get those things off-the-shelf, then I can focus on building the things that really add value, that are important to my particular business: the business logic, the innovation, the competitive advantages. Then I have a quicker time to market.” But it’s not just about making life easier for the coders, whether they’re in the IT department or elsewhere. A good low-code or no-code platform will also help the CIO, says Rosen. Governance guidelines “If I had to tell someone looking to buy a no-code or a low-code tool what to look for to tell whether that vendor is serious about helping them build a culture of low-code/no-code development, it would be controls to help them set up governance around who can use the tools and what the tools can be used for,” says Rosen. In some ways, governance around low-code tools is no different than for other software development tools, says Nick Mates, VP of operations and technology at Lendr, an online business-to-business loan platform. “We treat a low-code application as if it was a traditional code application,” he says. “It should follow the same governance lifecycles, from a business analyst’s desk to a developer’s desk to a QA desk to deployment.” But with code-facilitating tools such as these, enterprises must also set up governance around which tools are best used for which use cases, Rosen says, noting that many organizations have multiple such tools in operation in house. Organizations with the most experience leveraging low-code and no-code tools have also set up centers of excellence (CoEs) to advise lines of business on which tool to use and when, she says. The CoEs also provide support by coding more complex interactions and integrations that low-code development tools or their users can’t handle, providing reusable components that line-of-business developers can access, and curating them in a marketplace or code repository. One thing that plays into deciding the right tool for the job, and when a professional developer’s help is needed, is the level of interoperability that any given vendor has enabled in their platform, Rosen says. “Do they really just want you to bring all your data and logic to their platform, or are they allowing you to build apps that cross over multiple platforms?” she says. “That’s an important feature that customers can look for.” The cost of keeping up Should CIOs not yet all-in on the trend budget their software development tool spend to keep pace with IDC’s growth forecasts? “It’s not something where they need to make a major investment,” says Rosen. Setting up a CoE and making reusable software components available are affordable steps for most enterprises, she says. “Generally speaking, it’s not expensive to get started,” she adds. “What’s expensive is scale.” Rather than worrying about whether their software spending is keeping pace with that of their competitors, Rosen advises CIOs to ask themselves, “What features are we not offering that we might be able to offer using low code, and that will be impactful for the business?” That approach could lead to cost savings since reusing componentized interfaces may mean a reduced need to hire expensive, expert programmers to build each application from scratch. One key indicator on budgeting would be to weigh the cost per user for low-code platform licenses against the cost of hiring additional staff, Rosen says. For now, the difficulty of finding highly experienced professional developers is tilting that balance in favor of enabling line-of-business staff with low-code tools. Lower down the expertise scale, the decision whether to hire or to reskill existing staff is less clear, she says. At this level, CIOs need to take other advantages of deploying low-code platforms into account: not just develop a new digital business product, but perhaps also empower employees or improve retention. “Once you know what you’re going for, you can look at the platforms with a different perspective,” Rosen says. Related content brandpost Sponsored by Avanade By enabling “ask and expert” capabilities, generative AI like Microsoft Copilot will transform manufacturing By CIO Contributor 29 Feb 2024 4 mins Generative AI Innovation feature Captive centers are back. Is DIY offshoring right for you? Fully-owned global IT service centers picked up steam in 2023, but going the captive route requires clear-eyed consideration of benefits and risks, as well as desired business outcomes. By Stephanie Overby 29 Feb 2024 10 mins Offshoring IT Strategy Outsourcing feature What is a chief data officer? A leader who creates business value from data A chief data officer (CDO) oversees a range of data-related functions to ensure your organization gets the most out of, arguably, its most valuable asset. By Minda Zetlin and Thor Olavsrud 29 Feb 2024 10 mins Staff Management Careers IT Leadership tip The trick to better answers from generative AI Miso.ai co-founders Lucky Gunasekara and Andy Hsieh discuss how going beyond RAG basics to break down question context and assumptions is key for enterprise-grade service. By Sharon Machlis 29 Feb 2024 5 mins Generative AI PODCASTS VIDEOS RESOURCES EVENTS SUBSCRIBE TO OUR NEWSLETTER From our editors straight to your inbox Get started by entering your email address below. Please enter a valid email address Subscribe