Low-Code/No-Code in 2026: How Citizen Developers Are Reshaping Enterprise Software

The Low-Code/No-Code Market at Scale

The low-code/no-code (LCNC) market has moved from hype to mainstream enterprise infrastructure. Gartner predicted that 70% of new enterprise applications would use low-code or no-code technologies by 2025, up from less than 25% in 2020 — and that milestone has largely materialized. The firm now forecasts that by 2026, low-code will account for 75% of new application development, with developers outside formal IT departments making up at least 80% of the user base for low-code tools. The definition of “application” has expanded: every internal dashboard, approval workflow, inventory tracker, and customer survey that previously lived in spreadsheets is now an “application” built on LCNC platforms. Gartner projects the low-code development technologies market will exceed $30 billion in 2026, growing at a compound annual rate of 14.1% to reach $58.2 billion by 2029.

Microsoft Power Platform is the market’s dominant force. With Power Apps (application builder), Power Automate (workflow automation), Power BI (analytics), and Copilot Studio (AI-powered agent builder), Microsoft has embedded LCNC tooling into its enterprise ecosystem — Microsoft 365 and Azure. Microsoft reported 56 million monthly active Power Platform users in 2025 at the Power Platform Community Conference, up from 48 million in 2024 and 33 million in 2023 — a growth trajectory that dwarfs any standalone competitor. The platform is used by 97% of Fortune 500 companies. The advantage is distribution: any organization with Microsoft 365 licenses already has Power Platform access, lowering the adoption barrier to near zero. Microsoft was named a Leader in both the 2025 Gartner Magic Quadrant and the 2025 Forrester Wave for low-code platforms.

Beyond Microsoft, the LCNC landscape segments into distinct categories. Retool (valued at $3.2 billion after its October 2025 Series D) dominates internal tool building — connecting to databases, APIs, and third-party services to create admin panels, dashboards, and operational tools. Its estimated annual recurring revenue reached $120 million by late 2025, boosted by AI features including AppGen (natural language app generation) and Agents. Bubble leads in no-code web application development, with 4.7 million applications built on the platform as of August 2025, enabling non-developers to build full SaaS products with database, authentication, and payment integration. OutSystems ($9.5 billion valuation, $331.6 million revenue in 2024) and Mendix (acquired by Siemens for $730 million in 2018, $169.8 million revenue in 2024) serve large enterprises with complex application requirements, compliance needs, and legacy system integration. Appsmith and Tooljet offer open-source alternatives to Retool. The market is large enough to support deep specialization.

What Low-Code/No-Code Actually Serves Well

The use cases where LCNC platforms genuinely deliver value are well-established and growing. Internal tools — the category Retool, Appsmith, and Tooljet serve — represent the clearest win. Every company has dozens of internal applications that engineers build reluctantly: customer lookup tools, order management dashboards, content moderation queues, employee onboarding workflows. These applications are important but not differentiating; they do not justify custom frontend engineering. Retool’s model — drag-and-drop UI components connected to SQL queries and API calls — enables a single developer to build in hours what a frontend-backend team would take weeks to produce.

Workflow automation is the second strong category. Tools like Zapier (over 3 million users, 2.2 million businesses, $310 million revenue in 2024), Make (formerly Integromat), and Microsoft Power Automate connect SaaS applications through trigger-action sequences without code. When a new row appears in a Google Sheet, create a Jira ticket. When a customer submits a support form, post to Slack and create a HubSpot contact. When an invoice is received via email, extract data with OCR and update the accounting system. These automations replace manual processes that consume hours of employee time weekly.

n8n, the open-source workflow automation platform, has emerged as one of the most significant players in this space. In October 2025, n8n raised $180 million in Series C funding at a $2.5 billion valuation, led by Accel with participation from Nvidia’s NVentures — validating the demand for self-hosted, developer-friendly automation. The platform serves over 230,000 active users and 3,000 enterprise customers including Vodafone, Delivery Hero, and Microsoft. Its December 2025 release of n8n 2.0 introduced enterprise-grade features: secure-by-default execution with isolated task runners and a publish/save workflow model that separates drafting from pushing changes live — addressing a core governance concern for production automations.

Prototyping and MVPs represent the third validated use case. Founders and product managers who want to test a product concept before committing engineering resources use Bubble, Softr, or Glide to build functional prototypes. Dividend Finance, a renewable energy fintech, partnered with Airdev to launch its MVP on Bubble in just six weeks and has since processed over $1 billion in loans. Comet, a French freelance marketplace, built its initial product on Bubble and was generating over $800,000 in monthly recurring revenue within months, going on to raise EUR 16 million across seed and Series A rounds. The value is not that no-code replaces engineering but that it accelerates the learning cycle: build, test with users, iterate, and only invest in custom engineering when the product concept is validated.

The AI Convergence: LCNC Platforms Get Smarter

The defining development of 2025-2026 is the convergence of AI and low-code/no-code platforms. Every major platform has integrated generative AI capabilities, fundamentally changing how citizen developers build applications.

Retool launched AppGen in April 2025, allowing users to describe an application in natural language and receive a functional implementation — connected to real data sources with working UI components. In May 2025, Retool followed with Agents, enabling AI agents to interact directly with Retool-built applications. These features introduced new monetization units: AI prompting credits and agent hours.

Bubble launched its AI Agent in October 2025, letting non-technical founders describe desired features in plain English and receive functional implementations within minutes. The AI understands specific app context, maintains design systems, and troubleshoots workflows. By November 2025, the AI Agent could create entirely new pages using an app’s established styles. Bubble also launched native mobile app capabilities in public beta, expanding from its web-only origins.

Microsoft is pushing the AI integration furthest. Copilot Studio now enables low-code development of custom AI agents, and the 2026 roadmap shifts Copilot from responding to individual commands toward operating as specialized autonomous agents. Developers and businesses can create, deploy, and sell custom Copilot agents in a dedicated marketplace.

Gartner identifies agentic AI as a key driver accelerating LCNC adoption through 2029, predicting that the typical enterprise will run 4,500 to 6,000 AI-generated apps, workflows, and automations by 2026. The implication is significant: AI is not replacing low-code platforms but supercharging them, making citizen development accessible to an even broader population of business users while simultaneously raising the governance stakes.

Where Low-Code/No-Code Falls Short

The limitations are real and often underestimated by enthusiastic adopters. Performance is the first constraint. LCNC platforms generate abstracted code that is less efficient than hand-written applications. A Bubble application handling 10,000 concurrent users will struggle where a custom Next.js application handles it comfortably. Database query optimization, caching strategies, and connection pooling — performance levers that experienced engineers pull routinely — are often inaccessible or limited in LCNC environments. For high-traffic, performance-sensitive applications, LCNC platforms hit ceilings.

Customization is the second constraint. Every LCNC platform eventually reaches a point where the desired functionality exceeds the platform’s component library or configuration options. The “last 20% problem” is well-documented: building 80% of an application on a LCNC platform is fast, but achieving the remaining 20% of custom behavior — a specific animation, a complex multi-step form with conditional logic, a custom chart type, a non-standard authentication flow — can be more difficult than building the entire application in code. Some platforms (OutSystems, Mendix, Retool) allow injecting custom code, but this hybrid approach requires developers anyway, undermining the “no-code” promise.

Vendor lock-in is the third and often most consequential limitation. An application built on Bubble runs on Bubble’s proprietary infrastructure. If Bubble raises prices, changes terms, or shuts down, migration requires rebuilding from scratch — there is no “export to React” button. This lock-in is acceptable for prototypes and internal tools but creates genuine business risk for production applications serving paying customers. OutSystems and Mendix mitigate this by generating deployable code, but the generated code is rarely clean enough for an engineering team to maintain independently. The open-source alternatives (Appsmith, Tooljet, n8n) avoid vendor lock-in at the hosting level but still create dependency on the platform’s abstraction model.

The Governance Challenge: Shadow IT and Security

The democratization of application building creates a governance headache that IT departments are still learning to manage. When any business user can build an application that connects to production databases, external APIs, and customer data, the risk surface expands dramatically. A marketing analyst building a Power App that queries the customer database might inadvertently expose PII. A sales manager automating a Zapier workflow might create an integration that sends sensitive pricing data to an unsecured webhook. These scenarios are not hypothetical — they are the daily reality of enterprise shadow IT in the LCNC era.

The scale of the problem is growing. Gartner predicts that by 2026, the typical enterprise will run 4,500 to 6,000 AI-generated apps, workflows, and automations — with 66% remaining undiscovered by security and IT teams. Business-user developers are expected to outnumber professional developers 4:1 by 2026, forcing organizations to extend governance programs to citizen developer platforms. The recommended approach — and the one adopted by organizations like Johnson & Johnson, Unilever, and ANZ Bank — is a “center of excellence” (CoE) model: a small team of IT professionals who define guardrails (approved data sources, required authentication methods, security review triggers), provide training, and review citizen-built applications before they access production systems. This balances empowerment with control.

The security dimension extends to the platforms themselves. LCNC platforms hold credentials to customer databases, API keys for third-party services, and often have broad permissions in the enterprise environment. A compromise of a Retool account illustrates the risk: in August 2023, attackers used a multi-vector social engineering campaign — SMS phishing sent to employees, voice phishing with a deepfaked voice impersonating an IT team member, and exploitation of Google Authenticator’s cloud sync feature — to bypass MFA and compromise 27 cloud customer accounts in the crypto industry. Multi-factor authentication, IP allowlisting, audit logging, and regular access reviews are essential — but they are security practices that citizen developers rarely implement without IT department enforcement. The trade-off is clear: LCNC platforms accelerate development at the cost of centralized security control.

Frequently Asked Questions

Sources & Further Reading

• Gartner Forecasts Low-Code/No-Code Platform Market for 2026 — Kissflow
• Microsoft Power Platform Community Conference 2025 — MSDynamicsWorld
• Microsoft Recognized as Leader in 2025 Gartner Magic Quadrant for Enterprise LCAP — Microsoft
• n8n Raises $180M Series C at $2.5B Valuation — n8n Blog
• Retool $3.2B Valuation and Revenue Data — Sacra
• Retool Security Incident: MFA Isn’t MFA — Retool Blog
• Bubble No-Code Platform Review 2026 — Hackceleration
• Citizen Developers Dominate: Development Predictions for 2026 — BetaNews
• Zapier Statistics 2026: Growth Momentum — SQ Magazine
• Dividend Finance Scales No-Code Loan Platform — Airdev