Before You Cut the Workforce, Redesign the Work
Updated: Aug 26
Cost reductions may be necessary. Headcount reductions may be part of the answer. And increasingly, leaders are assuming AI will absorb enough of the remaining work to make a smaller workforce viable.
But that only works if the work itself changes.
When roles are eliminated without rethinking processes, decisions, handoffs, governance, technology, and service expectations, the work does not disappear. It migrates—to managers, to already-stretched teams, or to AI tools that were never designed to carry the operating burden being placed on them.
Before you decide how many people the organization needs, redesign the work the organization actually needs done.
Before You Cut the Workforce, Redesign the Work
Sometimes the mandate is straightforward. Costs have to come down.
For many organizations, labor represents a significant portion of the cost base, so when savings requirements are substantial, workforce reductions may be unavoidable. The issue is not that organizations reduce headcount. The issue is what happens when they reduce the workforce without making corresponding changes to the work.
Positions are eliminated and organizational charts are redrawn, but the underlying operating burden often remains. Reports still need to be produced, approvals still need to happen, customers still expect service, and processes continue to carry the same handoffs, governance requirements, and complexity. The organization may have fewer people, but it has not necessarily created less work.
Increasingly, AI is being introduced into that equation as the presumed answer. If AI makes employees more productive, the reasoning goes, then a smaller workforce should be able to absorb the remaining workload.
That may ultimately be true, but it is not automatic.
AI can enable a more productive operating model, just as process simplification, automation, better decision rights, improved data, and clearer accountability can. What it cannot do is eliminate the need to deliberately redesign the work itself.
When Headcount Changes but the Work Does Not
When a position disappears, its responsibilities do not automatically disappear with it. Some work may legitimately stop, some may be automated or simplified, and some may be reassigned. When those choices are not made intentionally, however, the default is usually redistribution.
Managers begin producing reports previously handled by analysts. Several employees absorb pieces of an eliminated role. A management layer is removed, but its approvals remain embedded in the process. Teams are combined without eliminating duplicate meetings, reporting requirements, controls, or stakeholder expectations.
The organization chart becomes leaner while the operating burden remains largely intact.
Over time, that burden begins to surface elsewhere. Decisions may take longer, capacity becomes constrained, accountability becomes less clear, service levels decline, and employees spend more time compensating for gaps in the operating model. Eventually, leaders may conclude that additional capacity is needed and begin hiring again. That does not necessarily mean the original cost objective was wrong. It may mean the work was never redesigned to support the smaller organization.
That is the distinction between reducing headcount and creating sustainable productivity.
Start With the Work
When leaders receive a mandate to reduce headcount by a specific percentage, the natural starting point is the organization chart. They look for positions that can be removed, teams that can operate with fewer people, or management layers that can be eliminated.
Those questions are necessary, but they should not be the first questions.
A more durable approach starts with the work required to deliver the strategy. Leaders need to understand which activities genuinely create value, which are no longer necessary, where work is duplicated, where processes contain unnecessary complexity, and where the organization is spending capacity maintaining legacy practices that no longer serve the business.
This is also the right time to examine service expectations, decision rights, governance structures, reporting requirements, spans and layers, and the way work moves across functions. Some activities may need to stop. Others may need to be simplified, consolidated, automated, or performed differently.
Once those choices have been made, leaders are in a much stronger position to determine the capabilities and capacity the organization actually needs.
Workforce size should be informed by the operating model, not simply imposed upon it.
AI Changes the Possibilities, Not the Need for Redesign
AI materially expands what is possible in work redesign. It can accelerate analysis, reduce administrative effort, automate portions of workflows, improve access to information, support decision-making, and increase employee capacity.
Those benefits, however, depend heavily on how the surrounding work is designed.
Consider a report that once took several hours to draft. Generative AI may reduce the writing time dramatically, but the employee may still spend hours collecting information from multiple systems, reconciling inconsistent data, obtaining stakeholder input, navigating approvals, and producing different versions for different audiences.
AI has improved one task within the process. It has not necessarily improved the process itself.
This distinction becomes especially important when organizations begin translating anticipated AI productivity directly into headcount assumptions. A theoretical productivity gain should not be treated as available capacity until leaders understand where that gain will actually occur, what process changes are required to capture it, and what new work may accompany the technology.
AI may eliminate some activities and accelerate others, but it can also introduce new requirements around validation, exception handling, data quality, governance, risk management, and human oversight. In many cases, it augments judgment-intensive work rather than replacing it.
The right question is not simply whether AI can help the remaining workforce do more. Leaders need to ask how the work should be redesigned now that these capabilities exist and what workforce is required once that redesigned model is in place.
Do Not Take the Productivity Before You Create It
There is also a sequencing risk when organizations reduce workforce capacity today based on productivity they expect technology to deliver tomorrow.
The remaining employees may find themselves responsible for maintaining current operations while also learning new tools, supporting implementation, redesigning processes, validating AI-generated outputs, and absorbing responsibilities left behind by eliminated positions. The organization has effectively booked the productivity gain before it has created the conditions required to achieve it.
Business realities will not always allow organizations to redesign every process before making a workforce decision. Cost pressures can be urgent, and leaders sometimes have to act with imperfect information.
Even then, it is important to distinguish among productivity that has already been demonstrated, productivity that is reasonably achievable with identified changes, and productivity that is still largely aspirational. Those categories should not carry the same weight when determining future capacity.
Productivity Is More Than Asking Fewer People to Do More
A reduction in headcount can produce immediate savings. It may also produce a short-term increase in output per employee. Neither necessarily means the organization has created sustainable productivity.
Sustainable productivity comes from reducing the effort required to produce the desired outcome. That might mean eliminating unnecessary work, reducing handoffs, simplifying decision-making, removing low-value approvals, consolidating fragmented activities, automating repeatable tasks, improving the quality and accessibility of data, or changing service expectations to reflect the organization’s new economics.
These changes alter the underlying work equation. They make it possible for a smaller organization to operate effectively without simply transferring more responsibility to those who remain.
That is the difference between making the workforce leaner and making the organization more efficient.
Use the Cost Mandate as a Catalyst
A headcount reduction can create a powerful forcing mechanism. If the organization must operate with fewer people, leaders have an opportunity to challenge assumptions about how work has historically been done.
Processes often cross multiple functions because of legacy organizational structures rather than because the work inherently requires those handoffs. Governance structures accumulate layers of review over time. Reports persist long after the decisions they once supported have changed. Entire roles can evolve around coordinating the friction created by fragmented systems or unclear ownership.
These are precisely the conditions that work redesign should expose.
Technology, including AI, may offer new ways to address some of those problems. But technology should be incorporated into the redesign rather than treated as a substitute for it. Automating an unnecessary activity does not make the activity necessary, and placing AI on top of a cumbersome process does not automatically create a better operating model.
The opportunity is to reconsider the work end to end and then determine how people, process, technology, governance, and organizational structure should come together to deliver it.
Make the Savings Sustainable
A more sustainable approach to workforce reduction begins with the financial requirement but does not stop there. Leaders clarify the savings target, determine what work the strategy requires, identify what can be eliminated or simplified, redesign key processes and decision structures, assess where AI and automation can create real productivity, and then determine the capabilities and capacity required to operate the resulting model.
The savings requirement has not disappeared. The organization may still make difficult workforce decisions.
What changes is the likelihood that those savings will endure.
If an organization intends to operate with fewer people, it must also make deliberate choices about the amount and nature of the work those people will perform. AI can and increasingly will be an important part of that equation, but it should not become shorthand for capacity that has never actually been created.
Before asking how many positions the organization can remove, leaders should first ask how the organization should work after the reduction.
Then they can design the workforce and the technology around that answer.
Comments