
Gartner, a technology research firm, projected in November 2025 that the worldwide low-code development technologies market will reach $58.2 billion by 2029, a 14.1% CAGR. Low-code and no-code development platforms are software environments that replace much hand-written code with visual design, reusable components and, increasingly, AI-assisted generation.
For teams evaluating these platforms in 2026, speed is already proven. Quality, lifetime cost and security now depend on how the platform is governed.
What Are Low-Code and No-Code Development Platforms?
Low-code platforms minimize hand-coding but let developers add custom code, APIs and integrations. No-code platforms rely on visual configuration alone and serve business users with little programming experience. Many teams use the two terms interchangeably.
The vocabulary is looser than the market. Researchers and vendors draw the line between low-code and no-code in different places, and many "no-code" products allow some code for extensibility. So judge a platform by how far you can extend it, not by its label.
The Benefits of Using Low-Code and No-Code Platforms
These platforms reliably shorten delivery time for bounded business applications. Headline savings still overstate realized value.
The sharpest number comes from Forrester, a technology and market research firm. Its July 2024 Total Economic Impact study of Microsoft Power Apps models a 50% reduction in application-development time for professional developers building internal apps. That figure is a modeled result for a composite organization.
Forrester itself discounts that kind of figure. Its July 2024 Total Economic Impact study of Microsoft Power Platform applies a 50% productivity-recapture rate. Only half of the time saved counts as productive work or actual cost reduction. The other half disappears into the working day.
That matches what happens in practice. A developer who finishes a workflow app early may pick up the next backlog item, or may absorb the slack in meetings and context switching. Finance teams should treat TEI time savings as an upper bound and track what the freed hours actually produce.
TEI studies are vendor-commissioned models of hypothetical organizations, so they cannot stand alone as evidence. But an independent 2026 systematic literature review in the Journal of Systems and Software, "What does current research say about the viability of low-code development?", reaches the same conclusion on speed. It found sufficient evidence that low-code speeds up development.
The same review stops short of a quality claim. It found quality and complexity outcomes mixed and dependent on context. So the honest framing of the benefit is faster delivery and lower initial build cost.
The strongest case is narrow: internal workflow apps and prototypes with stable requirements, where a working version in weeks beats a perfect one in quarters. There, MVP development on a low-code platform can validate an idea before anyone commits to a custom build.
How Low-Code and No-Code Platforms Are Changing the Software Development Process
Faster delivery raises a second question: who does the building, and what happens to the engineering team? Low-code, now joined by prompt-based AI builders, is shifting IT from sole builder of every application to platform owner and reviewer.
Developer behavior already shows it. SlashData, a developer research firm, reported in April 2024, based on its global developer surveys, that the share of developers using low-code or no-code tools in their projects rose from 46% in Q1 2022 to 57% in Q3 2023.
Prompt-based building is also arriving inside platforms that IT already governs. Microsoft now offers a Power Pages plugin for GitHub Copilot CLI and Claude Code for building, securing and deploying Power Pages sites with AI. Power Pages is Microsoft's low-code website builder. The plugin works inside an existing environment with permissions and data policies in place, so IT can see it and restrict it.
AI-assisted building sharpens the change. Prompt-based building ("vibe coding") lowers the barrier to entry. It also raises the volume of code nobody on the team wrote or fully understands. Teams adopting it should invest in reviewing AI-generated code before anything reaches production.
The Role of Low-Code and No-Code Platforms in the Rise of Citizen Developers
Employees outside engineering now build working applications through visual tools and prompts. These citizen developers handle internal workflows and dashboards without waiting in the IT queue.
Controlled evidence on their capability is encouraging. Till Guthardt, Jens Kosiol and Oliver Hohlfeld of the University of Kassel, in a September 2024 study presented at the ACM/IEEE MODELS conference, compared programmers and citizen developers on a prototype no-code builder. They found no significant difference in task correctness or processing time. The authors conclude that training concerns for citizen developers may be overstated.
That result invites a tempting conclusion: if citizen developers match programmers, internal apps no longer need professional developers. The study does not support it. The researchers tested constrained tasks on a prototype builder, and production systems add integrations, load and audit requirements.
A 2026 systematic mapping study in Software: Practice and Experience, a peer-reviewed journal published by Wiley, is clear here. It found that while low-code and no-code tools simplify certain tasks, they do not replace the need for professional developers to handle complex technical challenges, change management and system integration.
For an engineering leader, the practical consequence is a change in what the team owns. Fewer hours go to building the tenth approval workflow. More go to platform configuration, permission models, data access rules and review standards. Organizations that treat this as a role redesign get speed and keep control.
Those that treat it as headcount relief find the unowned apps later, usually after an incident.
The Future of Low-Code and No-Code Development Platforms
If AI builders are changing who builds, platform vendors are changing what the platform is. Leading low-code platforms are becoming runtimes for AI agents and frequent updates. Buyers must evaluate release cadence, support options and built-in governance as seriously as the visual builder itself.
For market context, Forrester's July 2024 Power Platform study cites Forrester research valuing the low-code market at $13.2 billion in 2023. Set that beside the Gartner 2029 forecast. The two firms define the market differently, so treat the pair as directional.
Two release models now sit side by side:
| Mendix | Microsoft Power Platform | |
|---|---|---|
| Release model | Stable track plus LTS track | Cloud service updated by Microsoft; many features enabled automatically |
| Current reference point | 11.16 stable (Oct 28, 2026); 11.12 LTS (June 30, 2026) | 2026 release wave 1 (April to September 2026) |
| Customer-selected LTS version | Yes | No |
| AI and governance | AI agents, Maia, enterprise governance (11.12 LTS) | Agent security controls, AI-powered governance agents, real-time risk assessment in Copilot Studio |
Mendix, a low-code platform vendor, released 11.12 LTS on June 30, 2026 as a stable, production-ready foundation with AI agent capabilities, Maia improvements, mobile capabilities and enterprise governance. Mendix 11.16, the latest stable release as of October 28, 2026, added Maia planning before it builds, more flexible workflows and smarter pipelines. A buyer picks the stable track for new features or the LTS track for predictability.
Microsoft's 2026 release wave 1 plan for Power Platform covered features released from April through September 2026. Its governance section lists admin controls for agent security, real-time risk assessment in Copilot Studio, AI-powered governance agents that automate tenant monitoring and remediation, and more granular visibility into Copilot credit consumption. Governance now ships as a product feature, which tells you where vendors expect the risk to sit.
A fair objection: fast cloud updates and AI releases cost you control over when your applications change, and that makes production use riskier than before. The risk is real. The same plan notes that some features are enabled for users automatically, and Power Platform has no customer-selected long-term-support version.
The mitigation is practical. Choose an LTS track where one exists, as Mendix 11.12 LTS offers. Where none exists, run automated regression tests on workflows, permissions and connectors before each update reaches production. Treat every vendor release as a change request against your own estate.
Ask each vendor how long each release track is supported and what governance ships in the box: environment separation, audit logging, and controls on what an AI agent may touch. Those answers shape operating cost more than the drag-and-drop canvas does. Teams planning to run agents on these platforms should scope AI agent development skills in-house or with a partner, because an agent acts on your data.
Real-World Examples of Companies Using Low-Code and No-Code Platforms
Forrester's modeled studies provide two documented composite examples. Use them for what they are: composites built from interviews with real customers, with figures you can test against your own pilot.
Forrester's August 2024 Total Economic Impact study of Pipefy, a process-automation platform, models a 50% reduction in process-automation development time and an average 40% time saving on automated tasks for a composite organization. That is a modeled composite built from four customer interviews, not a named customer. The gains sit in process automation with well-understood requirements: approvals and routing logic a platform can template once and reuse. The recapture caveat applies here too.
The Power Apps composite points the same way. Its development-time reduction comes from internal apps built by professional developers, a bounded, repeatable category of work.
Read together, the documented outcomes cluster around standardized internal processes. They say far less about customer-facing systems, heavy integrations or high-volume transaction workloads, which is where the limits begin.
Low-Code and No-Code Is Not a Panacea
The more capable these platforms become, the more the outcome depends on how they are run. Apps that hit complexity, performance or customization limits belong in custom development.
The evidence for speed is firm. The evidence for quality, cost and security is thinner. Writing in IEEE Computer, the magazine of the IEEE Computer Society, in March 2025, Bilal Naqvi and colleagues assessed low-code and no-code platforms against the ISO 25010:2023 software quality model. They found that complexity and customized business needs limit the platforms' effectiveness in high-performance and high-risk environments. Rapid delivery does not establish production performance.
Security risk comes from how the platform is used. A 2024 governance study presented at ECIS, the European Conference on Information Systems, identifies an increase in shadow IT and data-security breaches as risks of low-code democratization. Business units build apps nobody inventoried, on data nobody classified.
Picture a regional operations manager who builds a purchase-approval app in a few days. It works, so finance adopts it, then procurement. A year later it routes spend for three departments, pulls vendor banking details and has no owner on the IT side. When the platform changes a connector, approvals stall and nobody knows who can fix it.
Cost is harder to pin down. Licensing, training, migration, maintenance and governance costs differ by context, so the same platform can save one team money and cost another more. Vendor lock-in belongs in that calculation. An app built on proprietary components can be expensive to leave, and per-user pricing grows with every new user.
The usual triggers to move off a low-code platform are an app that became business-critical without a named owner, a data model the platform cannot scale, or licensing that grows faster than usage. Teams that plan the handoff early schedule the move on their own terms.
A reasonable objection: tiers, owners and exit planning can slow citizen developers and erode the speed that justified low-code. The ECIS authors address this. Their preliminary findings suggest that strict, centralized governance is not always suitable and that more differentiated approaches are needed. Viljoen, Stelzl, Yang and colleagues, writing in Information Systems Journal, a peer-reviewed information systems research journal, in 2025, go further. They find that too much flexibility can compromise stability while too much standardization can stifle the autonomy and creativity that make these platforms valuable, so governance has to balance both at once.
So match the control to the risk. Personal productivity apps get light controls. Departmental, customer-facing and regulated apps get full review. Most building stays fast, because most apps sit in the lightest tier.
How to Get Started with Low-Code and No-Code Development Platforms
Start small, assign ownership before launch and plan the exit before you need it.
- Define the user problem before choosing a platform. Name who uses the app, what decision it supports and what success looks like.
- Build a narrow proof of concept with real data. Test with production-scale data and peak user counts, since a small pilot can hide problems that appear at scale.
- Assign each app a risk tier and a named owner.
- Verify data export before production. If you cannot extract your data and logic cleanly, you have found your lock-in cost.
- Plan the custom-development handoff early, so an app that outgrows the platform moves on a schedule. A structured custom software development discovery is the right first step for apps heading that way.
Two adjacent concerns deserve attention: modern architecture decisions for platforms that have aged into liabilities, and security for AI-assisted development as prompt-based builders enter your estate.
Frequently Asked Questions
What is the difference between low-code and no-code?
Low-code platforms minimize hand-written code but still let developers add custom code, APIs and integrations. No-code platforms rely entirely on visual configuration and serve business users with little programming experience. The boundary between the two is blurry, so judge platforms by their extensibility rather than their label.
Is low-code a replacement for custom software development?
No. Low-code suits standardized workflows, internal tools and prototypes where requirements are well understood. Custom development suits complex, performance-sensitive, security-critical or differentiating systems. A 2026 systematic mapping study in Software: Practice and Experience found that low-code and no-code tools do not replace the need for professional developers to handle complex technical challenges, change management and system integration.
Are low-code apps secure?
Low-code apps can be secure, but their security depends on governance. An ECIS 2024 governance study identifies shadow IT and data-security breaches as low-code risks. A March 2025 IEEE Computer article found that complexity and customization needs limit these platforms in high-risk environments. Role-based access, audit logging and environment separation are essential controls.
Does AI replace low-code platforms?
No. AI is being built into low-code platforms. Mendix 11.12 LTS added AI agent capabilities and Maia improvements. Microsoft's 2026 release wave 1 plan added AI-powered governance agents and real-time risk assessment in Copilot Studio. The platforms are absorbing AI rather than being displaced by it.
How many developers use low-code tools?
Developer adoption is rising. SlashData's global developer surveys, published in April 2024, found that the share of developers using low-code or no-code tools in projects rose from 46% in Q1 2022 to 57% in Q3 2023. Over the same period, the share using these tools for at least a quarter of their development work grew from 26% to 32%.
Inventory every production app built on low-code and no-code development platforms this quarter. Give each one a named owner, a risk tier and an exit path. Any app that has hit platform limits, carries regulated data or has become differentiating should be scoped for custom development now, while the move is still a decision you schedule.
Azumo is a SOC 2-certified nearshore software development company founded in 2016, with engineers in 20+ Latin American countries working in US time zones. We have served 100+ customers and hold a 4.9 rating on Clutch.





