Welcome to this edition of The Digital Enterprise, where we explore how companies can turn digital ambition into measurable outcomes.
This week, we explore why 78% of AI initiatives fail. Guess what, it’s not because the tech doesn’t work. It’s because organizations are running 21st-century tools with 20th-century operating models.
The Bottom Line Up Front
AI success isn’t about technology, it’s about how your organization operates. Our research shows that companies who evolve their operating models see 3x higher ROI and 60% faster value realization from tech initiatives. Without this evolution, most AI projects collapse under the weight of outdated processes and silos.
The Operating Model Crisis
Despite unprecedented investment in AI and technology initiatives—$1.8 trillion globally in 2024—78% failed to deliver expected business outcomes. The culprit isn’t the technology itself; it’s the failure to evolve operating models to support new technological capabilities.
Organizations investing heavily in AI without corresponding operating model evolution face similar challenges.
According to Gartner, at least 30% of generative AI projects will be abandoned after proof of concept by the end of 2025, due to poor data quality, inadequate risk controls, escalating costs or unclear business value.
The AI Was Ready. The Business Wasn’t.
In 2024, McDonald’s ended its three-year, $50+ million AI drive-thru partnership with IBM after implementing voice-powered ordering systems across 100+ locations. The AI technology was sophisticated, it could process complex orders, handle multiple customizations, and theoretically reduce labor costs while improving speed.
Yet the initiative failed spectacularly.
Social media filled with videos of frustrated customers as the AI added 260 chicken nuggets to orders, confused customer voices between lanes, and recommended bizarre combinations like ice cream with bacon.
After working with IBM for three years to leverage AI to take drive-thru orders, McDonald’s called the whole thing off in June 2024.
The problem wasn’t the AI technology, it was the operating model.
McDonald’s tried to overlay advanced AI onto traditional drive-thru processes without evolving how teams managed exceptions, handled customer escalations, or integrated AI decisions with human oversight. The result: expensive AI tools operating within analog business processes.
This McDonald’s example illustrates the fundamental challenge: AI and technology implementations fail when organizations focus on system deployment rather than operating model transformation.
Despite McDonald’s claiming 85% accuracy rates, the AI couldn’t handle real-world complexities like multiple cars, accent variations, or complex customizations because the operating model wasn’t designed to support AI-human collaboration.
Understanding the Operating Model Evolution Framework
The TDEOS™ (The Digital Enterprise Operating System) framework identifies four critical dimensions where operating models must evolve to support AI and technology initiative success:
Dimension 1: Structure and Governance Evolution
Traditional State: Hierarchical, function-based organization with IT as a separate delivery function.
Digital State: Cross-functional, domain-based teams with shared accountability for business outcomes.
Evolution Markers:
- Decision-making authority moves closer to customer impact
- Technology and business leaders co-create solutions
- Governance focuses on value realization, not project completion
- Resource allocation follows value streams, not departmental budgets
Dimension 2: Process and Technology Integration
Traditional State: Business processes defined separately from technology capabilities, leading to complex workarounds and AI systems that can’t integrate effectively.
Digital State: Process and AI/technology co-designed for optimal user experience and business efficiency.
Evolution Markers:
- End-to-end process ownership transcends departmental boundaries
- AI and technology capabilities shape process design, not just automate existing workflows
- Real-time data integration eliminates handoffs and delays
User experience drives process simplification and AI interaction design
Dimension 3: People and Capabilities Transformation
Traditional State: Skills development focused on technical training within existing roles, without considering AI augmentation.
Digital State: Continuous capability building aligned with evolving business needs and AI/technology potential.
Evolution Markers:
- AI fluency and technology literacy become core competencies across all roles
- Learning and development tied directly to technology initiative outcomes
- Career paths reflect hybrid business-technology-AI expertise
- Performance metrics include AI adoption, technology utilization, and value creation
Dimension 4: Performance and Value Management
Traditional State: Success measured by technical metrics (system uptime, user adoption, project completion) rather than AI/technology value delivery.
Digital State: Value realization measured through direct business impact from AI and technology capabilities.
Evolution Markers:
- KPIs directly connect AI and technology capabilities to business outcomes
- Real-time performance dashboards enable rapid course correction for technology initiatives
- Value tracking spans the entire customer and employee journey, including AI touchpoints
- ROI measurement extends beyond implementation to ongoing AI and technology optimization
The Operating Model Implementation Strategy
Operating model evolution can’t happen overnight, but it can’t wait for perfect conditions either. The TDEOS framework provides a staged approach for AI and technology initiatives:
Stage 1: Foundation Building (Months 1-3)
- Governance Structure: Establish cross-functional teams with clear accountability for AI and technology initiative outcomes
- Value Framework: Define success metrics that connect AI and technology capabilities to business impact
- Pilot Selection: Choose high-visibility, manageable scope AI/technology initiatives to demonstrate the new operating model
Stage 2: Capability Development (Months 3-9)
- Process Redesign: Rebuild key processes around AI and technology capabilities rather than legacy workflows
- Skills Integration: Develop hybrid business-technology-AI competencies across the organization
- Technology Enablement: Implement AI and platforms that support the new operating model rather than automate old processes
Stage 3: Scaling and Optimization (Months 9-18)
- Domain Expansion: Extend the evolved operating model across additional business domains and AI use cases
- Continuous Improvement: Establish feedback loops for ongoing optimization of AI and technology initiatives
- Cultural Integration: Embed new ways of working into organizational culture and performance systems
Common Operating Model Evolution Pitfalls
Pitfall 1: Technology-First Thinking Selecting AI solutions or technology platforms before defining the desired operating model leads to expensive customizations and complex workarounds.
Pitfall 2: Incremental Change Management Treating operating model evolution as a series of small changes rather than a systematic transformation limits the impact and sustainability of AI initiatives.
Pitfall 3: Governance Without Authority Creating AI steering committees or technology governance groups without the authority to make cross-functional decisions slows progress and dilutes accountability.
Pitfall 4: Metrics Without Meaning Measuring activity (AI training completed, systems implemented) rather than outcomes (business value created through technology) misses the point of modern initiatives.
Your Operating Model Evolution Assessment
Ask yourself these critical questions:
- Structure: Do your teams have the authority and accountability to deliver end-to-end customer value?
- Process: Are your business processes designed around AI and technology capabilities or constrained by legacy thinking?
- People: Do your employees have the hybrid business-technology-AI skills needed for modern success?
- Performance: Can you draw a direct line from your AI and technology investments to business outcomes?
If you answered “no” to any of these questions, your operating model may be limiting your technology initiative potential.
Taking Action: Your Next Steps
Operating model evolution begins with honest assessment and committed leadership. Consider these immediate actions:
Week 1: Conduct a rapid operating model assessment using the four dimensions framework
Week 2: Identify one high-impact domain where you can pilot the evolved operating model with an AI or technology initiative
Week 3: Define success metrics that connect AI and technology capabilities to business outcomes
Week 4: Establish cross-functional governance with clear accountability for value realization from technology initiatives
Remember: AI and technology implementations without operating model evolution are just expensive system deployments. True transformation occurs when technology enables new ways of creating and delivering value.
🎯 Want to know if your operating model can support AI success?
👉 Take the free TDEOS Operating Model Assessment: https://tdeos.com/contact-us
Connect: Share your operating model evolution challenges and successes, what’s working with your AI and technology initiatives? What obstacles are you facing?
The Digital Enterprise Newsletter is published weekly, delivering insights on enterprise evolution, business strategy, and value realization. Subscribe for strategic perspectives that help you navigate the complexity of technology-driven change.
TDEOS™ – The Digital Enterprise Operating System Bridging the gap between technology implementation and business impact www.tdeos.com
