Featured Post

Acquire to Accelerate: Inside Accenture's Strategic Growth Engine

Background: From Andersen Consulting to AI PowerhouseAccenture’s journey began in 1957 as the business and technology consulting division o...

Sunday, October 12, 2025

Build vs. Buy is Dead, Start Orchestrating: CIO's Strategy in the AI World

 

For decades, CIOs have faced a relentless challenge: deliver more applications, faster, with fewer resources. The traditional response has been a constant cycle of building custom solutions, purchasing SaaS platforms, and managing their complex integration. This approach has consumed budgets, stretched teams thin, and consistently left business units frustrated with slow delivery.

Artificial Intelligence is fundamentally changing this dynamic. The traditional application strategy, centred on building and maintaining software, is giving way to a new paradigm: orchestrating outcomes through intelligent agents.

Vibe coding empowers domain experts to create software capabilities using natural language prompts. Business teams can now build what they need, when they need it, without waiting for central IT resources. The CIO's role thus transforms from simply delivering applications to orchestrating outcomes—ensuring trust, governance, and measurable results across an ecosystem of AI-driven agents.

This post explores why CIOs must stop thinking about applications and start orchestrating outcomes. We examine how AI reshapes the technology landscape, what vibe coding means for enterprise architecture, and how forward-thinking CIOs are positioning their organisations for this transformation


1. The CIO’s Evolving Role in the AI Era

The role of the CIO in enterprises has undergone a profound transformation, mirroring the region’s journey from isolated markets to global digital players. Once tasked with managing servers and vendor contracts, CIOs are now strategic visionaries, orchestrating intelligent ecosystems that drive business outcomes. The advent of AI, particularly vibe coding and agentic systems, is accelerating this shift, demanding that CIOs redefine their approach to software strategy in a landscape shaped by legacy systems, regulatory rigour, and talent scarcity.

In the 1990s, CIOs were gatekeepers of monolithic systems, where big enterprises like SAP or Oracle promised to unify operations but often tied enterprises to rigid architectures. A major bank’s multimillion-dollar SAP implementation, for instance, required years of advice from management consulting firms, with costs soaring into the tens of millions. These systems, while robust, were inflexible, requiring extensive customisation to meet the needs of the bank's diverse financial services. A CIO’s focus was primarily on infrastructure: ensuring servers hummed, databases synchronised, and vendor contracts aligned.

Fast forward to the 2000s, and the Software as a Service (SaaS) revolution brought agility to enterprises. Companies embraced Saas to streamline customer interactions, deploying solutions in weeks rather than years, shifting budgets from capital to operational expenditure. Yet, SaaS introduced its own challenges: data silos, limited customisation, and escalating subscription costs as enterprises juggled multiple platforms.

Article content
CIO's Evolution

Today, AI is completely changing the CIO's job. Their role is no longer about building software or buying programs. Instead, the CIO is like a city planner in a booming digital city. They don't build every house or road, but they must design the infrastructure to ensure traffic (data) flows smoothly and all the new buildings (AI tools and vibe-coded non-programmer apps ) connect reliably. This focus on orchestration delivers the best business results and ensures speed and governance for the entire enterprise.

This shift is particularly critical, where enterprises face unique constraints. Legacy systems, such as those powering major airlines' booking platforms, remain critical but costly, with maintenance budgets straining IT departments. Regulatory frameworks like APRA CPS 234 in Australia, which mandates robust cybersecurity for financial institutions, and New Zealand’s Privacy Act 2020, which enforces strict data protection, add layers of complexity. Moreover, ANZ’s talent pool is constrained, Australia’s developer shortage drives median salaries to $120,000, while New Zealand’s smaller market limits access to skilled engineers.

The traditional build versus buy debate—should we build custom software for control, or buy SaaS for speed—is no longer enough. AI introduces a spectrum of new possibilities: vibe coding enables quick, specialised app creation; agentic platforms deliver scalable solutions focused on business results; and orchestration combines all of these into seamless working systems. For example, a Melbourne-based fintech CIO might use vibe coding to create a dashboard that complies with APRA rules in hours, connecting it to internal systems and external APIs. Simultaneously, they could use a platform like Netcore's agentic solution to analyse customer data. This spectrum of tools requires a new mindset: one that focuses on outcomes over features, agility over ownership, and trust over control.

Yet, CIOs face distinct challenges in this transition. Stakeholder alignment is a hurdle, as they must convince boards to invest in AI by demonstrating real returns amidst economic pressures. Legacy modernisation is another key difficulty: integrating these new vibe-coded apps with decades-old systems without causing major disruption is a delicate process. Governance is also critical because uncontrolled app growth risks shadow IT and non-compliance with standards like the APRA or the Privacy Act. The CIO's role is to successfully manage these challenges, using AI to transform enterprises into agile, outcome-driven organisations.

2. The Application Delivery Crisis

2.1 The Traditional CIO Challenge

CIOs operate in a perpetual state of tension. Business units demand new applications to compete, innovate, and operate efficiently. Yet IT departments face:

  • Resource constraints: Limited developers, competing priorities.
  • Technical debt: Legacy systems requiring maintenance and updates.
  • Integration complexity: Dozens of SaaS platforms that don't communicate.
  • Shadow IT: Business units building their own solutions outside IT oversight.
  • Budget pressure: Growing costs for licenses, infrastructure, and talent.

The traditional response—hiring more developers, buying more platforms, and implementing better processes—only delays the inevitable. The list of applications needed by the business grows faster than the IT department can deliver.
Article content
Image Credit: Unsplash

2.2 Why Traditional Strategies Fail

The fundamental problem isn't execution—it's the model itself.

Building applications is expensive and slow:

  • Months of requirements gathering and design.
  • Skilled developers are in short supply.
  • Complex deployment and maintenance.
  • Rigid architectures that resist change.

Buying SaaS platforms creates new problems:

  • Limited customisation for unique workflows.
  • Data fragmentation across multiple systems.
  • Integration nightmares and API management overhead.
  • Subscription costs that compound annually.

The result: CIOs spend 70 to 80% of their budget just maintaining existing systems, leaving very little money or time for innovation. Business units become frustrated and create their own shadow IT solutions, which undermines central governance and security.

The application delivery model is fundamentally broken. AI offers a way out.

3. The AI Inflection Point: From Applications to Outcomes

3.1 What's Actually Changing

AI isn't just another technology trend. It represents a fundamental shift in how organisations create and consume software capabilities.

Three forces are converging:

  • Large Language Models (LLMs) enable natural language interaction with systems.
  • Agentic AI can understand intent, orchestrate workflows, and deliver results.
  • Vibe Coding allows non-technical users to create applications through prompts.

Together, these forces eliminate the traditional bottleneck: the need for developers to translate business requirements into code.

3.2 From Features to Outcomes

The traditional application strategy focuses on features:

  • "We need a dashboard that shows customer engagement."
  • "Build a workflow for approving expenses."

The AI-driven strategy focuses on outcomes:

  • "Increase customer retention by identifying at-risk accounts."
  • "Reduce approval cycle time by 50%."

This shift is profound. CIOs stop being application delivery managers and become outcome orchestrators. The question changes from "how quickly can we build this?" to "what's the most effective way to deliver this outcome?"
Article content
Build vs Buy - Pre and Post AI

3.3 What Vibe Coding Means for CIOs

Vibe coding allows business users to create applications by describing what they want.

  • Instead of: Submitting a ticket to IT and waiting months for development...
  • Now: Tell an AI agent: "Create a dashboard showing customer churn risk, update it daily, and alert the sales team when accounts score above 70." The agent builds it in minutes.

For CIOs, this means:

  • Application creation scales infinitely without adding headcount.
  • Innovation cycles compress from months to hours.
  • IT focuses on governance rather than delivery.

The constraint shifts from development capacity to orchestration capability.
Article content
Software Creation - Pre and Post AI

4. Stop Building: Why the Development Model Is Obsolete

4.1 The Economics No Longer Work

When business users can create applications themselves through vibe coding, the economics of custom development collapse.

Article content
The Economics of Development: Traditional vs AI Driven
CIOs who continue investing heavily in traditional development teams will find themselves outpaced by organisations that embrace orchestration.

4.2 The Talent Problem Is Unsolvable

The global shortage of software developers isn't temporary—it's structural. Vibe coding solves this by democratising creation:

  • Marketing teams build campaign automation tools.
  • Sales operations create pipeline management dashboards.
  • HR professionals design onboarding workflows.

The talent constraint disappears when domain experts become builders.

4.3 Technical Debt Accelerates Faster Than You Can Pay It Down

Every custom application creates technical debt. Traditional development teams spend 60–80% of their time on maintenance rather than innovation.

AI-managed applications reduce technical debt by:

  • Self-healing: Agents automatically detect and resolve issues.
  • Automatic updates: AI adapts to changing requirements.
  • Simplified architecture: Prompt-based changes rather than code refactoring.

The verdict is clear: CIOs must stop investing in traditional application development and start building orchestration capabilities.

5. Start Orchestrating: The New CIO Operating Model

5.1 What Outcome Orchestration Means

Outcome orchestration is fundamentally different from application delivery. Instead of managing development projects, CIOs manage an ecosystem of intelligent agents that deliver measurable business results.

The orchestration model involves:

  • Defining desired outcomes with business stakeholders.
  • Selecting or creating agents to deliver those outcomes.
  • Ensuring trust and governance across all agent activities.
  • Measuring and optimising based on actual results.

CIOs shift from asking "what applications do we need?" to "what outcomes must we deliver, and how do we orchestrate agents to achieve them?"

5.2 The Three Pillars of Orchestration

Successful outcome orchestration rests on three pillars:

Article content
Pillars of Orchestration

5.3 Building Your Orchestration Capability

CIOs should take a phased approach:

Phase 1: Foundation

  • Establish a trust layer platform and a governance framework.
  • Run pilot projects with Vibe coding in 2–3 departments.

Phase 2: Expansion

  • Roll out vibe coding and launch the internal agent marketplace.
  • Implement full outcome analytics.

Phase 3: Optimisation

  • Refine outcome-based pricing models.
  • Enable cooperative agent networks and automated governance.

5.4 Reorganising IT for Orchestration

The shift requires organisational change. The traditional structure evolves:

  • Application Development Teams Trust Engineers and Agent Curators.
  • Project Managers Outcome Architects.
  • Business Analysts Prompt Engineers.

The skills evolve; the mindset transforms.

6. The Trust Layer: Non-Negotiable Foundation

6.1 Why Trust Is the Moat

In a world where anyone can create applications through vibe coding, trust becomes the primary differentiator. Organisations need confidence that agents operate securely, reliably, and in compliance with regulations. Without a robust trust layer, vibe coding creates unacceptable risk.

6.2 Components of Enterprise Trust

A comprehensive trust layer includes:

  • Security: Agent vetting, data access controls, encryption, and threat detection.
  • Compliance: Adherence to GDPR, HIPAA, and industry regulations; automated policy enforcement.
  • Reliability: Service Level Agreements (SLAs), redundancy, and change management.
  • Auditability: Complete activity logs, transparent decision-making, and explainable AI for critical workflows.

CIOs must implement platforms that provide enterprise-grade trust across the entire agent ecosystem. This isn't optional—it's foundational.

7. From Features to Outcomes: The Economic Transformation

7.1 The Outcome-Based Pricing Revolution

Traditional software pricing charges for features (per-user licenses, per-API-call fees), misaligning incentives. Outcome-based pricing aligns incentives perfectly:

  • Pay for results: Cost per conversion, not per email sent.
  • Fee per successful onboarding, not per workflow execution.
  • Charge per accurate forecast, not per analysis run.

This transformation changes procurement fundamentally.

Article content
Source: Gen AI The New Reality - 2023

7.2 Measuring What Matters

Outcome orchestration requires new measurement frameworks. Traditional IT metrics (uptime, response time) become secondary. Outcome metrics take precedence:

  • Revenue impact and cost reduction.
  • Risk mitigation and compliance adherence.
  • Process efficiency gains and time to value.

CIOs must build analytics capabilities that connect agent activities to business results, providing clear ROI visibility to executives.

8. Strategic Implications for the Modern CIO

8.1 Rethinking the IT Operating Model

The shift requires fundamental changes to how IT operates:

  • From Project-Based to Continuous: Replace fixed roadmaps with dynamic optimisation and continuous monitoring.
  • From Centralised to Federated: Empower business units with creation capabilities while IT provides the platform, governance, and oversight.
  • From IT as Deliverer to IT as Enabler: Focus the CIO's value proposition on platform capabilities, trust infrastructure, and enabling outcome achievement across the organisation.

8.2 Budget Reallocation

The orchestration model requires significant budget shifts. Over 2–3 years, CIOs should aim to shift 40–60% of application budgets from traditional development and SaaS subscriptions to orchestration platforms and outcome-based services.

Conclusion: The CIO's Orchestration Mandate

The application delivery model is obsolete. CIOs who continue investing primarily in traditional development and SaaS procurement will find their organisations outpaced by competitors who embrace orchestration.

The transformation is already underway. Forward-thinking CIOs are:

  • Stopping large-scale custom development projects.
  • Investing in orchestration platforms and trust layers.
  • Empowering business users to create through vibe coding.
  • Measuring success by business outcomes, not IT metrics.

The organisations that master this transition will unlock unprecedented innovation velocity, dramatically reduce IT costs, and position themselves as AI-forward leaders in their industries.

The future belongs to the orchestrators. Get started today by exploring how orchestration empowers your organisation to stop building and start delivering outcomes at scale.

Note on the Evolution of Enterprise Software

The enterprise software landscape has evolved dramatically, shaped by technology, economics, and business needs. Understanding this journey illuminates why AI and vibe coding are transformative for enterprises.

Article content
The Evolution of Enterprise Software

1. The Monolithic Era (1970s–1990s): Early enterprise software was monolithic—integrated platforms like SAP and Oracle that managed finance, supply chains, and more. Barclays’ $50 million SAP implementation in the 1990s exemplifies this era. Implementation required extensive consultancy, long timelines, and high costs.

Characteristics:

  • Centralised databases with proprietary silos.
  • Heavy reliance on IT teams.
  • Long development cycles.
  • Vendor lock-in with high switching costs.

These systems offered stability but were rigid.

2. The SaaS Revolution (2000s–2010s): The 2000s brought Software-as-a-Service (SaaS), democratising access to tools like Salesforce and Workday.

Benefits:

  • Rapid deployment.
  • Subscription-based costs.
  • Vendor-managed updates.
  • Reduced infrastructure burden.

Challenges:

  • Limited customisation.
  • Data silos across tools.
  • Integration complexity.
  • Rising subscription costs.

3. APIs and Composability (2010s–2020s): SaaS’s limitations led to API-first platforms and microservices. UK fintechs like Monzo used AWS and Stripe APIs to build modular systems. Tools like Zapier enabled data flows, and Azure provided scalable infrastructure.

Trends:

  • Flexibility over completeness.
  • Developer empowerment via APIs.
  • Ecosystem thinking.
  • Platforms as products.

However, building custom solutions remained complex, requiring developers and regulatory governance.

4. The Build vs. Buy Debate

The build vs. buy decision hinges on:

  • Build: Bespoke software for control and differentiation.
  • Buy: COTS for speed and support.

Factors:

  • Time to market.
  • Budget constraints.
  • Internal capabilities.
  • Strategic importance.

Summary: From monoliths to APIs, software has become more accessible but complex. AI offers a way to overcome these constraints.


Related Posts:

No comments: