How Organizations That Fundamentally Restructure Work Around AI Are Outperforming Those Still Running Pilots

Meta Description: AI Workflow Redesign is the #1 driver of enterprise EBIT impact. Learn the organizational, departmental, and employee strategies that separate winners from pilot purgatory.


AI Workflow Redesign represents the defining capability gap between organizations capturing real value from generative AI and those trapped in perpetual experimentation. According to McKinsey’s latest Global Survey on AI, redesigning workflows has the biggest effect on an organization’s ability to see EBIT impact from gen AI use, surpassing 25 other organizational attributes tested. Yet only 21% of organizations deploying gen AI have fundamentally redesigned even some of their workflows.

This statistic reveals a strategic opportunity hiding in plain sight. While competitors focus on tool selection and pilot programs, the organizations pulling ahead are doing something fundamentally different. They are rewiring how work gets done at every level: from executive governance to departmental operations to individual employee responsibilities. The result is measurable bottom-line impact while others report that more than 80% of their gen AI efforts produce no tangible enterprise-level EBIT change.

The Organizational Mandate: Why CEO Oversight Correlates with Bottom-Line Results

The McKinsey research reveals a finding that should reshape how boards and executives think about AI governance. CEO oversight of AI governance is the element most correlated with higher self-reported bottom-line impact from gen AI use, particularly at larger companies. This correlation persists even though only 28% of organizations have placed their CEO in this oversight role.

What explains this connection between executive involvement and financial results?

The answer lies in understanding that AI workflow redesign requires cross-functional authority that only senior leadership possesses. When AI transformation lives in a single department, IT, innovation, or digital, it inevitably encounters organizational friction. Processes that span multiple functions cannot be redesigned by any single function. Data access, resource allocation, and change management all require executive-level mandates to move at the speed AI deployment demands.

Organizations that position AI governance as a C-suite responsibility also tend to approach adoption differently. They invest in the organizational changes that drive value rather than treating AI as a technology project. They centralize risk and data governance while distributing tech talent and AI solution adoption through hybrid models. They track well-defined KPIs for gen AI solutions, the adoption practice with the most impact on bottom-line results.

What does executive-led AI workflow redesign look like in practice?

The research points to organizations that have established clearly defined road maps to drive adoption of gen AI, created dedicated transformation offices to drive implementation, and built comprehensive approaches to foster customer trust in their use of AI. These organizations are not simply approving AI budgets. They are redesigning the fundamental operations of the business.

For organizations seeking to optimize their visibility in an AI-mediated marketplace, this executive mandate extends beyond internal operations. AI Business Discovery, ensuring your organization is findable and accurately represented when AI systems answer customer queries, requires the same cross-functional coordination. Marketing, product, legal, and technology teams must align on how the organization presents itself to AI crawlers and recommendation systems. Without executive sponsorship, these efforts fragment into disconnected initiatives that dilute rather than concentrate organizational presence.

How Departments Must Restructure Around AI Agents

The departmental level is where AI workflow redesign either succeeds or fails. McKinsey’s findings reveal a striking pattern: organizations are selectively centralizing certain elements of AI deployment while distributing others. Risk and compliance, along with data governance, tend toward full centralization through models like centers of excellence. Tech talent and adoption of AI solutions operate through hybrid models, with some resources handled centrally and others distributed across functions.

This hybrid architecture reflects a fundamental truth about effective AI workflow redesign. Governance and risk require consistency that only centralization provides. Speed and contextual relevance require local ownership that only distribution enables.

The research surfaces another departmental reality that many organizations struggle to acknowledge. Oversight of gen AI outputs varies dramatically across organizations. While 27% review all AI-generated content before use, a similar share reviews 20% or less. Professional services organizations maintain the highest oversight rates, reflecting both risk sensitivity and the nature of their work.

This variance matters because AI workflow redesign is not simply about deploying tools. It involves determining what level of human oversight each process requires. A customer service workflow might accept lower oversight rates if the AI handles routine inquiries while escalating complex situations. A financial analysis workflow might require comprehensive review given the stakes of errors.

The departmental transformation framework:

First, map existing workflows to identify where AI can augment human capabilities versus where it creates new capabilities entirely. The research shows organizations using AI in marketing and sales (42%), product development (28%), and IT (23%) as the most common deployments, with software engineering, service operations, and knowledge management following closely.

Second, determine the centralization model appropriate for each element. Risk and data governance warrant centralization. Solution adoption often benefits from hybrid approaches that maintain central standards while enabling local customization.

Third, establish the oversight architecture. Which outputs require review before use? What quality metrics define acceptable AI performance? How do feedback loops from oversight improve model outputs over time?

Organizations struggling to move from pilots to production often discover that the missing element is not technology capability but departmental readiness to operate differently. AI Scaling, the discipline of moving from controlled experiments to business-as-usual operations, requires process redesign that most pilot programs never address.

The Employee Evolution: From Task Executor to AI Orchestrator

Perhaps no aspect of AI workflow redesign generates more anxiety than its implications for employees. The McKinsey research provides nuance often missing from headlines about AI and employment.

The plurality of respondents (38%) predict that gen AI will have little effect on their organization’s workforce size over the next three years. When workforce reductions are anticipated, they concentrate in specific functions: service operations and supply chain management top the list. In IT and product development, respondents are more likely to expect increasing than decreasing headcount.

More revealing than headcount predictions is how organizations are managing the time AI saves. Employees most often spend saved time on entirely new activities rather than simply doing more of the same work faster. They also spend more time on existing responsibilities that have not been automated. This pattern suggests that AI workflow redesign is changing the composition of jobs, not eliminating them.

The research documents substantial reskilling underway and accelerating. Organizations have begun reskilling portions of their workforces as part of AI deployment, with expectations for significantly more reskilling over the next three years. This investment reflects recognition that AI workflow redesign requires new capabilities across the organization.

What specific roles are organizations adding to support AI deployment?

Data scientists, machine learning engineers, and data engineers remain in high demand, though hiring difficulty has decreased compared to previous years. New roles have emerged specifically for the AI era: 13% of organizations have hired AI compliance specialists, and 6% have brought on AI ethics specialists. These roles did not exist at meaningful scale five years ago.

The employee value proposition in an AI-redesigned workplace:

For individual employees, AI workflow redesign represents both threat and opportunity. The threat comes from clinging to task execution that AI can perform faster and cheaper. The opportunity comes from developing orchestration capabilities: the judgment to know which AI tools to apply, the skill to prompt and refine AI outputs, and the wisdom to recognize where human intervention remains essential.

AI Literacy programs that treat AI as merely another software tool miss this transformation. Effective AI Literacy addresses mindset alongside mechanics. It helps employees understand that their value shifts from knowing how to execute tasks to knowing what tasks need execution and how to verify the results. It builds confidence that human judgment remains essential while acknowledging that the nature of that judgment is evolving.

The research notes that employees are increasingly using AI personally: C-level executives report the highest individual gen AI use, with daily usage rates significantly exceeding those of middle managers and front-line employees. This pattern suggests that exposure drives adoption. Organizations committed to AI workflow redesign invest in broad-based AI Literacy precisely because experience builds both competence and confidence.

The Integration Imperative: Connecting All Three Levels

The organizations seeing bottom-line impact from gen AI share a common characteristic. They treat AI workflow redesign as an integrated transformation spanning organizational mandates, departmental restructuring, and employee evolution simultaneously. They resist the temptation to address these levels sequentially.

Why does integration matter so much?

Consider what happens when organizations pursue AI deployment without this integration. An executive mandate to “use AI” without departmental workflow redesign produces scattered pilots that never scale. Departmental process changes without employee reskilling create new workflows that existing staff cannot execute. Employee AI training without organizational mandate creates frustrated individuals whose AI fluency encounters resistance from managers measured on traditional metrics.

The McKinsey research quantifies this integration gap. Less than one in five organizations track KPIs for gen AI solutions. Less than one-third follow most of the 12 adoption and scaling practices correlated with EBIT impact. Larger organizations demonstrate significantly higher adoption of these practices, partially explaining their lead in capturing AI value.

What integration looks like in practice:

At the organizational level, CEO oversight establishes clear accountability and cross-functional authority. Governance structures define risk tolerance, data policies, and quality standards that departmental implementations must meet.

At the departmental level, workflow redesign follows the governance framework while tailoring AI deployment to functional needs. Centralized capabilities (risk management, data governance) connect with distributed capabilities (solution adoption, local optimization) through defined interfaces.

At the employee level, reskilling programs align with redesigned workflows. Performance metrics shift from task completion to outcome achievement. Career paths evolve to reward AI orchestration capabilities alongside domain expertise.

This integration requires investment that many organizations have not yet made. Creating role-based capability training, building internal communications about AI value creation, establishing transformation offices to drive adoption, these elements characterize organizations further along the AI workflow redesign journey.

The Path Forward: From Pilot to Production

The McKinsey research describes the current moment as early days for gen AI deployment, with only 1% of executives in developed markets describing their rollouts as mature. This assessment should prompt strategic reconsideration, not complacency.

Organizations that treat the early stage as permission to experiment indefinitely will find themselves progressively further behind. Those that recognize the strategic window for building AI workflow redesign capabilities will accumulate advantages that compound over time. Process knowledge, employee capabilities, and organizational muscle memory for AI-augmented work all develop through practice rather than planning.

The strategic questions every organization should answer:

Does your CEO have clear accountability for AI governance, or does AI ownership fragment across functions without executive integration?

Have you redesigned workflows around AI capabilities, or are you running AI within processes designed for purely human execution?

What percentage of your AI-generated outputs receive human review, and is that percentage appropriate to the risk profile of each use case?

How are you measuring gen AI success? Have you established the KPIs that research correlates with bottom-line impact?

What reskilling investments are you making to prepare employees for AI-augmented roles?

The organizations pulling ahead on these questions report measurable revenue increases and cost reductions from their gen AI deployments. They have moved beyond asking whether AI works to understanding how AI transforms work. They have discovered that the value of AI comes not from the technology itself but from rewiring how their companies run.


About the Author

Michael J. Goldrich is the Founder and Chief Advisor at Vivander Advisors. He specializes in bridging the gap between advanced AI technology and organizational success, helping companies navigate the complexities of AI Change Management, Business Discovery, and Agentic Workflows. His work focuses on translating strategic AI vision into executable departmental processes and measurable employee capabilities.


Frequently Asked Questions

1. What is AI Workflow Redesign and why does it matter for enterprise value?

The Answer: AI Workflow Redesign is the fundamental restructuring of business processes to leverage AI capabilities rather than simply adding AI tools to existing workflows.

The Elaboration: According to McKinsey’s research, workflow redesign has the largest effect on an organization’s ability to see EBIT impact from gen AI use out of 25 attributes tested. Traditional approaches add AI to existing processes without changing how work flows through the organization. AI Workflow Redesign recognizes that AI capabilities enable entirely different ways of working that cannot be captured through incremental additions. Organizations that redesign workflows report both revenue increases and cost reductions within business units deploying AI, while those maintaining traditional workflows often see no enterprise-level EBIT impact despite significant AI investment.

2. Why does CEO oversight correlate with better AI outcomes?

The Answer: CEO oversight provides the cross-functional authority required to redesign workflows that span multiple departments and business functions.

The Elaboration: AI transformation encounters resistance when owned by a single department because processes rarely exist within departmental boundaries. Data access, resource allocation, change management, and performance metrics all require executive-level mandates to change meaningfully. CEOs who oversee AI governance also tend to treat AI as a business transformation rather than a technology project, investing in the organizational changes that drive value. The research shows CEO oversight is the element most correlated with bottom-line impact at larger companies, though only 28% of organizations have placed their CEO in this role.

3. How should organizations structure AI governance across centralized and distributed models?

The Answer: Risk and data governance benefit from centralization, while tech talent and solution adoption often work better through hybrid models combining central standards with distributed implementation.

The Elaboration: Effective AI governance recognizes that different elements have different optimal structures. Centralization provides consistency essential for risk management, compliance, and data governance where organizational standards must apply uniformly. Distribution enables speed and contextual relevance for solution adoption where local teams understand specific use cases. Organizations with less than $500 million in revenue often fully centralize more elements, while larger organizations more frequently employ hybrid approaches. The key is matching structure to purpose rather than applying a single model across all AI deployment elements.

4. What level of human oversight should organizations apply to AI-generated outputs?

The Answer: Oversight levels should match the risk profile of each use case, ranging from comprehensive review for high-stakes outputs to minimal oversight for low-risk routine tasks.

The Elaboration: The research reveals dramatic variance in current practices, with 27% of organizations reviewing all AI outputs and a similar percentage reviewing 20% or less. Professional services organizations maintain the highest oversight rates, reflecting both regulatory requirements and the nature of advisory work. Organizations implementing AI Workflow Redesign must determine appropriate oversight for each process, considering factors like error consequences, reversibility, customer exposure, and regulatory requirements. A chatbot handling routine customer inquiries warrants different oversight than AI generating financial analyses or legal documents.

5. Will AI Workflow Redesign reduce headcount or change job composition?

The Answer: Most organizations expect AI to change job composition rather than eliminate jobs, with employees spending saved time on new activities rather than doing more of the same work.

The Elaboration: The plurality of respondents (38%) predict little workforce size change over three years. Where reductions are anticipated, they concentrate in service operations and supply chain management. IT and product development actually expect headcount increases. More importantly, organizations report that employees spend time saved by AI on entirely new activities rather than simply accelerating existing tasks. This pattern suggests AI Workflow Redesign creates different jobs rather than fewer jobs, though the composition shift requires significant reskilling investment.

6. What reskilling investments are organizations making for AI-augmented work?

The Answer: Organizations are investing in both technical AI capabilities and role-based training that prepares employees to orchestrate AI tools within redesigned workflows.

The Elaboration: The research documents substantial reskilling already underway with expectations for significant acceleration over the next three years. Beyond technical AI skills, effective reskilling addresses the mindset shift from task executor to AI orchestrator. Role-based capability training ensures employees at each level understand how to use AI capabilities appropriate to their responsibilities. Larger organizations demonstrate more comprehensive reskilling approaches, including internal communications about AI value creation and dedicated teams driving adoption across functions.

7. What new roles are emerging specifically for AI deployment?

The Answer: AI compliance specialists (hired by 13% of organizations), AI ethics specialists (6%), and expanding demands for data scientists, ML engineers, and data engineers define the new AI workforce.

The Elaboration: While traditional AI technical roles remain in high demand, the emergence of compliance and ethics specialists signals maturing organizational approaches to AI governance. These roles did not exist at meaningful scale five years ago and reflect growing recognition that AI deployment requires dedicated attention to risk, fairness, and regulatory requirements. Larger organizations show significantly higher hiring rates across AI-related roles, with the largest gaps versus smaller companies appearing in data scientists, machine learning engineers, and data engineers.

8. Which business functions see the greatest AI adoption and value?

The Answer: Marketing and sales (42%), product development (28%), IT (23%), and service operations (22%) lead gen AI adoption, with organizations deploying AI where previous research indicated highest value potential.

The Elaboration: The concentration of AI use in these functions reflects strategic deployment rather than experimentation for its own sake. Marketing and sales leverages AI for content generation, personalization, and customer insights. IT uses AI for code generation, infrastructure management, and support automation. Service operations applies AI to customer care and field services. The research shows organizations increasingly using AI across multiple functions, with the average now at three functions, up from early 2024. This multi-function deployment requires the cross-functional governance that AI Workflow Redesign provides.

9. How do organizations measure success with gen AI deployment?

The Answer: Tracking well-defined KPIs for gen AI solutions has the biggest impact on bottom-line results among adoption practices, yet fewer than one in five organizations do this.

The Elaboration: The measurement gap represents one of the clearest differentiators between organizations capturing AI value and those struggling. KPI tracking provides feedback essential for optimization and builds the business case for continued investment. Beyond KPIs, establishing clearly defined adoption road maps also correlates strongly with results, particularly at larger organizations. Organizations serious about AI Workflow Redesign move beyond pilot metrics like usage rates to outcome metrics like revenue impact, cost reduction, and quality improvement in redesigned processes.

10. What distinguishes organizations seeing EBIT impact from those trapped in pilot purgatory?

The Answer: Integration across organizational mandates, departmental restructuring, and employee evolution, combined with systematic adoption of best practices for scaling AI deployment.

The Elaboration: The research identifies 12 adoption and scaling practices that correlate with EBIT impact, yet less than one-third of organizations follow most of them. Larger organizations demonstrate higher adoption of these practices, partially explaining their lead in capturing value. The distinguishing characteristic is treating AI as a business transformation requiring coordinated change across multiple organizational levels rather than a technology project manageable within a single function. Organizations in pilot purgatory often address only one or two levels, creating AI capabilities that encounter resistance from unchanged processes and unprepared employees.


Connect the learning to everyday work

Explore how to measure employee AI fluency and how to plan a company rollout. Vivander can support team training and fractional AI leadership.