Cutting roles creates budget, not value. How will you redesign work to put people at the centre of your AI future?
As AI adoption accelerates, 80% of organizations are cutting roles, yet only 20% are seeing meaningful revenue impact. The difference comes down to how work gets done: organizations that redesign work are up to 2.5× more likely to outperform and create new value.
Key takeaways
- Cutting roles based on AI creates budget headroom, but rarely drives meaningful revenue impact. Only a small share of organizations see real returns.
- Organizations that redesign work for human–AI collaboration outperform on ROI, achieving up to 2× higher success rates and stronger financial outcomes.
- Future performance depends on building new capabilities, with leaders reallocating talent towards higher value tasks that drive growth and differentiation.
How should organizations treat their workforce as they seek to build AI-enabled workflows? Over the last few years, many prominent companies have followed a familiar playbook: invest in AI tools, reduce headcount, and expect ROI to quickly follow.
Large enterprises across nearly every industry are deploying AI solutions, and workforce reductions are quickly following. In fact, over 80% of companies deploying autonomous business capabilities are also reducing their workforce.1
But what if that approach has been wrong all along? The data and lived experience of leaders are beginning to tell the story about what happens when you equate cost reduction with value creation.
Deloitte research shows no consistent evidence that AI-driven workforce reductions lead to improved financial performance. On the contrary, while 84% of organizations are increasing AI investment, only 20% report meaningful revenue impact.2
So why isn’t AI alone translating into value?
Most organizations stopped at automation. They deployed AI tools and reduced headcount, but they didn’t redesign the work itself. As a result, early efficiency gains haven’t translated into sustained performance or meaningful revenue impact.
Predictably, reducing your workforce creates gaps. Key and distinctly human elements like judgment, context, and creativity remain central to outcomes. When roles are removed without redefining how the work gets done, those capabilities disappear with them. Organizations then find themselves scrambling to rebuild critical expertise, as missed opportunities and execution challenges begin to surface.
Value comes from redesigning work so that people and AI operate together, with each contributing where they have the greatest impact.
Redesign separates value from efficiency
Organizations that capture value from AI treat it as a work redesign challenge rather than focusing on headcount. They rethink workflows, decision-making, and how human judgment is applied alongside machine capability. Deloitte’s 2026 Global Human Capital Trends show that redesigning work makes organizations twice as likely to exceed AI ROI expectations and nearly 2.5 times more likely to deliver stronger financial outcomes than those focused on efficiency alone.3
Many organizations are arriving at this conclusion the hard way. Early workforce reductions, made on the assumption that AI could replace end-to-end work, have exposed capability gaps that slow execution and limit scale. In response, organizations are rehiring or reintroducing expertise to restore what was lost.
When organizations make permanent talent decisions based on new or temporary technology capabilities and treat early productivity gains as evidence of structural displacement, they miss a crucial step: reinvesting capacity through work redesign.
This creates a striking contrast. You wouldn’t implement a new technology solution without redesigning processes, mapping dependencies, and planning for adoption. Yet human capital decisions are often treated as binary: keep roles or remove them. That approach overlooks the more complex and more valuable task of redesigning how people contribute alongside AI.
Just as you rethink workflows when you introduce new technology, creating value from AI requires the same level of intentional thinking about how people contribute.
Three archetypes shaping the future of AI-enabled work
While AI can automate and accelerate parts of work, outcomes still depend on human judgment, context, and decision-making. The organizations that outperform will be those that deliberately design for this human edge, ensuring these capabilities are embedded in how work gets done.
Our Deloitte research has shown that 93% of AI spend is directed towards technology with the remaining 7% directed towards people and change.4 A work redesign approach increases the proportion of spend on people and change.
The organizations that will be most successful will focus on three critical talent archetypes:
- Operators: those who can deploy, run, and maintain AI in production
- Translators: those who connect AI capability to business outcomes and decisions
- Governors: those who ensure trust, risk management, and decision quality at scale
Redesigning work around these archetypes is part of reimagining the business itself. Leading organizations start with the future they want to deliver for customers, then redesign roles and processes to enable that outcome. This is a departure from the common approach of pushing AI use cases and automation, then trying to drive adoption without a clear path to value.
What to do next: eight no-regret moves
Rather than focusing on reducing headcount, the organizations seeing real return are asking an essential question: “How do we redesign work around the opportunity in front of us, and do we have the capability to run it?”
Here are eight no-regret moves that will help answer that question.
1. Redesign the work before you redesign the workforce
- Start with work, not roles. Identify your most critical workflows and break them into tasks to automate, augment, or elevate.
- Sequence changes to workflows so they can be tested and adopted without disrupting performance or customer experience.
- Use data to understand how those tasks are performed today and how they will change with AI.
- Make explicit decisions about what work no longer exists, where technology can intervene, and what new value should be delivered.
2. Redeploy people before you release them
- Pause AI‑driven exits that are not tied to proven work redesign.
- Create short sprints focused on automation, governance, and AI‑enabled, redesigned roles.
3. Lead differently before you scale differently
- AI changes what leadership must look like. Elevate leadership expectations to reflect an AI-enabled environment where judgment, accountability, and orchestration matter more than oversight.
- Treat AI as a core leadership responsibility, be explicit about where human judgment sits, and role-model how to use AI responsibly.
- Make decision trade-offs, risk, and direction visible, especially when AI is involved.
4. Build a skills system instead of a training program
- Shift from episodic upskilling to a durable capability model.
- Establish three capability tiers:
- AI literacy (everyone)
- Workflow application (role‑specific)
- Advanced (operators, translators, governors)
- Map current talent to future needs. Build a clear build-vs-buy talent strategy. Most workforce data today reflects the current state: what roles exist and what skills people have. But redesign requires a forward-looking view of work itself: what tasks will matter, how workflows will change, and where new value will emerge. Platforms such as Deloitte’s Workforce Analyzer are helping organizations make that shift, moving from static workforce assessment to dynamic modeling of future work and capability needs.
- Anchor this model in forward-looking data on how work and tasks are expected to evolve, not just current capability assessments.
5. Change how you hire and onboard for AI
- Hire for demonstrated AI use in real work.
- Design for Day 1 productivity, not orientation.
- Provide immediate access to AI tools.
- Onboard the human and their AI workflows together.
6. Rewire reward systems to focus on impact
- Incentivize productivity gains, automation delivered, and decisions improved, with clear metrics defined upfront.
- Shift rewards from effort to impact by tying recognition and compensation to measurable business outcomes, not activity levels.
- Redesign performance management to reflect AI-enabled workflows, where individuals are accountable for orchestrating outcomes, not just producing outputs.
7. Activate alumni and flexible talent pools
- Reengage former employees through structured alumni programs and reentry pathways, with a focus on talent that brings AI, data, and automation expertise.
- Use contract and flexible talent models to access specialized skills to scale expertise up or down as technology and business priorities evolve.
- Alumni already understand the business. In addition to reduced hiring risk and faster productivity, they become a strategic source of scarce skills, returning with institutional knowledge and organizational agility.
8. Establish AI governance early
- Define where AI decides versus where it supports human decision-making.
- Clarify accountability for outcomes.
- Put risk, audit, and escalation models in place.
Real examples of redesign driving value
Redirecting effort unlocks 30% productivity increase
A European telecom added AI to customer service without redesigning roles or workflows and saw only a ~5% productivity lift. When the organization redirected ~90% of its rollout effort into redesigning human–AI workflows, including escalation paths, trust thresholds, and training, it achieved a ~30% productivity increase.5
Unlocking $60 million in capacity opportunity
A global technology company partnered with Deloitte to move beyond AI tool deployment and focus on structured work and role redesign. Teams identified 170+ AI-enabled use cases and quantified over US$44 million (approximately CAD$60 million) in capacity opportunity.6 This shift enabled the organization to translate AI investment into measurable business value at scale.
Preserving expertise to accelerate problem solving
A global automobile manufacturer risked losing critical expertise as experienced engineers retired, while still needing high levels of technical judgment in complex operations. It launched a Senior Experts Program, rehiring retired engineers into flexible, project-based roles to solve technical challenges and mentor teams. This preserved institutional knowledge, accelerated problem solving, and enabled structured knowledge transfer.7
What it takes to capture the full value of AI
Leading organizations are redefining how they track value, linking work redesign to business KPIs such as customer experience, speed, and decision quality rather than just cost savings. Simply layering AI onto existing processes will continue to deliver only marginal gains.
Combining work redesign with forward-looking data and modeling through platforms such as Workforce Analyzer, Deloitte helps leaders identify where AI value will emerge, define the capabilities required, and track outcomes as work evolves. The focus moves from reducing effort to creating impact, and from short-term efficiency via workforce reductions to sustained performance from your people.
Reach out to our leaders to get started.
Stephanie Hauck-Rachinsky
Stephen Harrington
