Research / Empirical evidence on rollout of organisation-wide low-code and no-code programs

Empirical evidence on rollout of organisation-wide low-code and no-code programs

2026-05-14 · governance-policy workforce-skills tools-infrastructure organisational-design · medium · source → · wiki →
key claims
  1. The strongest accessible empirical studies show that enterprise citizen-development programs succeed when organisations treat them as governed operating models, pairing maker enablement with central standards, technical support, and explicit design choices around security, compliance, and organisational changeBinzer et al. (2024)Viljoen et al. (2024)Binzer et al. (2025)
  2. The available evidence supports faster local solution delivery and broader participation by non-IT staff, but the literature does not yet provide a robust cross-study enterprise effect size for productivity, software quality, backlog reduction, or cost savingsAjimati et al. (2025)Binzer et al. (2024)Digital (2021)
  3. Recurring failure modes across the consulted studies include substandard software quality, tools outside formal IT visibility, technical debt, security and compliance strain, and adoption barriers whose severity depends on the surrounding organisational context rather than on the platform aloneViljoen et al. (2024)Ajimati et al. (2025)Kirikova (2022)Kass et al. (2023)
  4. Successful rollout models repeatedly combine business-user autonomy with structured collaboration mechanisms, including training, ideation intake, coordinated scaling, project execution support, professional catalyst roles, and central review or repository controlsBinzer et al. (2025)Binzer et al. (2024)Digital (2021)
  5. The best adjacent scale-up evidence indicates that pilot success does not generalize automatically, because enterprise rollout introduces coordination, review, and support burdens that require separate scaling phases and more expert involvement than early local experiments doBinzer et al. (2025)Fischer et al. (2022)Digital (2021)
  6. Taken together with prior completed repository items on governed low-code operating models and systems-capability debt, the empirical literature supports reading citizen-development outcomes as a capability-and-governance problem first and a tooling-choice problem secondGithub (n.d.)Github (n.d.)Binzer et al. (2024)Viljoen et al. (2024)

Research Question

What does peer-reviewed and independently verified empirical evidence reveal about the outcomes, success factors, governance models, and failure modes of organisation-wide low-code or no-code (LCNC) programs that let non-Information Technology (IT) business staff build applications, often called citizen development?

Findings

(Populated from section 6 synthesis above.)

Executive Summary

Empirical evidence shows that organisation-wide LCNC programs can improve local delivery speed and broaden participation, but durable rollout depends on treating citizen development as a governed operating model rather than as a simple tooling rollout. Across the strongest accessible studies, the recurring success pattern is bounded maker autonomy supported by central standards, technical experts, collaboration mechanisms, training, review paths, and escalation for higher-complexity work. The recurring failure pattern is socio-technical: studies repeatedly describe substandard software quality, tools outside formal IT visibility, technical debt, security strain, and context-specific adoption barriers when governance and support lag behind growth. Evidence for exact enterprise-scale performance gains remains thin, because the best accessible multi-organisation studies focus much more on governance design and collaboration mechanics than on standardized before-and-after outcome metrics.

Key Findings

  1. The strongest accessible empirical studies show that enterprise citizen-development programs succeed when organisations treat them as governed operating models, pairing maker enablement with central standards, technical support, and explicit design choices around security, compliance, and organisational change.
  2. The available evidence supports faster local solution delivery and broader participation by non-IT staff, but the literature does not yet provide a robust cross-study enterprise effect size for productivity, software quality, backlog reduction, or cost savings.
  3. Recurring failure modes across the consulted studies include substandard software quality, tools outside formal IT visibility, technical debt, security and compliance strain, and adoption barriers whose severity depends on the surrounding organisational context rather than on the platform alone.
  4. Successful rollout models repeatedly combine business-user autonomy with structured collaboration mechanisms, including training, ideation intake, coordinated scaling, project execution support, professional catalyst roles, and central review or repository controls.
  5. The best adjacent scale-up evidence indicates that pilot success does not generalize automatically, because enterprise rollout introduces coordination, review, and support burdens that require separate scaling phases and more expert involvement than early local experiments do.
  6. Taken together with prior completed repository items on governed low-code operating models and systems-capability debt, the empirical literature supports reading citizen-development outcomes as a capability-and-governance problem first and a tooling-choice problem second.

Assumptions

Analysis

The most plausible rival explanation is that LCNC rollout success or failure is driven mainly by platform maturity rather than by governance design. The consulted empirical studies do not fully reject that rival explanation, but they place much more explanatory weight on design choices, technical-expert involvement, collaboration mechanisms, and context-dependent adoption barriers than on any single platform capability. That weighting justifies a conclusion that governance and organisational capability are the dominant levers for rollout quality, while tool choice matters mainly by shaping the complexity boundary and the control surfaces available to the organisation. The weak state of quantitative outcome measurement is also decision-relevant, because it means executives should treat claims of large productivity gains as context-sensitive and should demand local measurement plans rather than assuming enterprise-wide benefits from the literature alone.

Risks, Gaps, and Uncertainties

Open Questions

sources

cites
cites Business-led low-code agent governance: conditions for durable value versus fragmentation in regulated environments
cites Systems capability debt as the root cause of citizen development: empirical evidence and effective governance architectures
related (frontmatter)
related What are the primary failure modes in enterprise Artificial Intelligence (AI) and low-code deployments, and how can governance systems be designed to mitigate them?
related Deployment pipeline as the only enforceable control gate for citizen-developed agents: DevOps literature support, low-code platform hook points, and architectural enforceability
related What maturity model best describes the evolution of governance capabilities for Artificial Intelligence (AI) and low-code in enterprises?
← newerDeclaration of the Independence of Cyberspace
older →Vendor Non-Compliance With or Absence of Implementation Standards

Connected items

Loading…

View full knowledge graph →