Organizations have become exceptionally good at measuring productivity. But as automation, artificial intelligence, and performance analytics reshape the workplace, a more important question is emerging: Are we improving the numbers, or improving the business?
By Mike Kehoe | Leading Through the Algorithm™
There is an interesting contradiction developing across modern business. Organizations have access to more performance data, sophisticated analytics, and productivity-enhancing technology than at almost any point in history. Executives can monitor operational performance in near real time, managers can measure activities with remarkable precision, and artificial intelligence is introducing new opportunities to automate processes that once required substantial human effort.
Yet despite these advancements, one of the most important questions in organizational performance remains surprisingly difficult to answer.
Are we actually getting better at what matters?
The distinction may seem straightforward, but it reveals a challenge that has existed in business for decades and is becoming increasingly important as technology accelerates the pace of work. Organizations often become so focused on improving efficiency that they lose sight of whether those improvements are producing meaningful value.
A team completes more assignments. A department reduces processing time. A store increases the number of tasks completed during a shift. A company introduces automation that allows employees to accomplish in minutes what previously required hours.
On paper, these developments represent progress. Productivity has increased, resources are being used more efficiently, and measurable performance indicators suggest that the organization is moving in the right direction.
But what happens when those improvements fail to produce a better customer experience, stronger business outcomes, or a more capable organization?
What happens when the work is completed faster, but the underlying problem remains?
This is what I think of as the efficiency illusion: the belief that improving the speed or volume of activity necessarily means improving organizational performance.
Efficiency is important. No responsible leader should dismiss the value of reducing unnecessary costs, eliminating waste, or improving productivity. But efficiency and effectiveness are not interchangeable concepts.
Efficiency measures how economically resources are used to perform an activity. Effectiveness asks whether that activity achieves the intended result.
An organization can become remarkably efficient at doing things that no longer create meaningful value.
And in an increasingly automated business environment, that possibility deserves far more attention than it receives.
When the Numbers Become the Objective
Modern organizations depend on key performance indicators to establish accountability, evaluate progress, and identify opportunities for improvement. Properly designed, these measurements provide valuable insight into whether a business is achieving its objectives.
The problem begins when the measurement gradually replaces the objective it was originally intended to represent.
In a 2019 Harvard Business Review article titled Don't Let Metrics Undermine Your Business, Michael Harris and Bill Tayler examined how performance indicators can unintentionally redirect organizational behavior away from broader strategic priorities. Their analysis highlighted a familiar business problem: when employees and managers become overly focused on measurable targets, they may prioritize improving those numbers rather than advancing the underlying strategy.
This concern is closely related to Goodhart's law, a principle associated with economist Charles Goodhart and commonly summarized as the tendency of a measure to become less useful when it is turned into a target.
The implication for leadership is significant.
If an organization measures success primarily through the number of tasks completed, employees will naturally focus on completing tasks. If managers are evaluated almost exclusively on speed, they will seek ways to accelerate performance. If a department's primary objective is reducing labor hours, decisions may increasingly revolve around labor efficiency.
None of these responses is necessarily inappropriate. The difficulty arises when the organization fails to consider what those activities are supposed to accomplish.
Consider a hypothetical retail operation where store teams are measured heavily on the speed of replenishment activities. Employees may become exceptionally efficient at moving products from storage areas onto shelves. Task-completion rates improve, labor productivity rises, and performance reports indicate success.
Yet if the replenishment process does not prioritize the products customers actually need, or if inventory information is inaccurate, the store may continue experiencing availability problems.
The employees are working efficiently. The measured process is improving. But the customer may see little benefit.
The same principle applies across industries.
A customer service department can reduce average call-handling time while leaving customers dissatisfied. A logistics operation can improve delivery speed while increasing errors. A technology team can accelerate software releases while introducing problems that require additional corrections.
In each example, a narrow measure of efficiency can improve while the broader measure of effectiveness deteriorates.
The lesson is not that organizations should abandon performance metrics. It is that leaders must continually examine whether their measurements remain connected to the outcomes they intend to achieve.
A dashboard can tell us that activity has increased. It cannot, by itself, determine whether that activity was worthwhile.
The Hidden Cost of Organizational Complexity
There is another dimension to the efficiency illusion that receives less attention: organizations sometimes introduce processes intended to improve productivity that ultimately make meaningful work more difficult.
As companies grow, they frequently add procedures, reporting requirements, approval structures, communication channels, and management systems. Each addition may have a legitimate purpose. A new report improves visibility. An additional approval creates accountability. Another meeting helps coordinate responsibilities.
Individually, these changes can appear reasonable. Collectively, however, they can create an increasingly complicated environment in which employees spend more time managing the process than accomplishing its original purpose.
Stanford researchers Robert Sutton and Hayagreeva Rao have explored this problem extensively. In their work on organizational friction, they describe how unnecessary rules, procedures, communications, and organizational obstacles can consume employees' time while undermining creativity and productivity.
Their research also challenges a common assumption: that removing every obstacle automatically makes an organization better.
Some friction is necessary. Quality checks, thoughtful review, ethical safeguards, and deliberate decision-making can prevent costly mistakes. The leadership challenge is distinguishing between procedures that protect value and those that merely consume resources.
That distinction becomes particularly important when organizations pursue efficiency initiatives.
A company may introduce a new reporting system to improve operational visibility, only to discover that employees are spending additional hours entering information already available elsewhere. A department may establish recurring meetings to improve communication while inadvertently reducing the time available for meaningful work.
The organization has added structure in the name of efficiency, but the structure itself has become a source of inefficiency.
Effective leadership requires the willingness to question these arrangements.
Not every process needs to be accelerated. Some need to be simplified. Others need to be redesigned. And some may no longer need to exist.
One of the most valuable questions a leader can ask is not how to make a process faster, but why the process exists in its current form.
What the Store Floor Teaches Us About Real Performance
Throughout my career in retail operations and emerging technology implementation, I've come to appreciate the difference between what performance reports reveal and what operational environments actually experience.
Retail provides an especially useful illustration because its success depends on numerous interconnected activities. Inventory accuracy, shelf availability, labor allocation, merchandising execution, customer service, and operational communication all influence the customer experience.
An improvement in one area does not necessarily translate into improvement across the entire operation.
For example, a retailer might introduce technology that dramatically reduces the time required to identify shelf conditions. From a technical perspective, the improvement could be substantial. Information that previously required considerable manual effort can now be collected and analyzed more efficiently.
But the business value of that information depends on what happens next.
Are the identified conditions translated into actionable priorities? Do store teams have the time, resources, and information necessary to respond? Are the underlying causes addressed, or does the organization simply become faster at identifying the same recurring problems?
These questions illustrate the distinction between technological efficiency and operational effectiveness.
Faster information is valuable when it enables better decisions. Improved visibility matters when it leads to meaningful action. Automation creates lasting value when it strengthens the broader process rather than simply accelerating one isolated component.
A retailer can become exceptionally efficient at identifying operational problems while remaining ineffective at resolving them.
That is not necessarily a failure of the technology. It may reflect a disconnect between the capabilities being introduced and the operating processes responsible for turning those capabilities into results.
The same challenge exists well beyond retail. In healthcare, manufacturing, logistics, financial services, and professional organizations, the value of better information depends on the decisions and actions that follow.
Technology can accelerate activity. Leadership must ensure that the activity advances the intended outcome.
Artificial Intelligence Raises the Stakes
The rapid adoption of artificial intelligence makes this distinction particularly important.
AI systems can summarize information, generate reports, analyze patterns, automate administrative tasks, and support increasingly complex business decisions. These capabilities offer legitimate opportunities to improve productivity and reduce the amount of time employees spend on repetitive work.
But the ability to complete a task more quickly does not automatically establish that the task should be performed in the same way—or even performed at all.
An organization could use AI to produce reports in seconds that previously required hours. That may represent a meaningful productivity gain. But if those reports rarely influence decisions, the technology has primarily made an existing activity faster.
The more important opportunity may be to reconsider what information decision-makers actually need, how that information should be delivered, and which actions it should support.
McKinsey's 2025 global survey on AI highlighted the importance of organizational changes, including workflow redesign, in helping companies translate AI adoption into business value. The findings suggest that introducing new capabilities is only one part of the challenge; organizations must also rethink how work is structured and managed.
That observation has significant implications for executive leadership.
The question should not simply be how much time artificial intelligence can save. It should also be how that time can be redirected toward higher-value activities.
If automation reduces administrative work, can employees devote more attention to customers? Can managers spend more time developing their teams? Can analysts move beyond assembling information and focus on interpreting its implications? Can organizations use improved visibility to prevent recurring problems rather than repeatedly responding to them?
These are questions about effectiveness, not merely efficiency.
There is also a risk that organizations will use the productivity gains associated with AI primarily to increase the volume of work expected from employees.
If a task that once required several hours can now be completed in minutes, the immediate temptation may be to assign more tasks. In some situations, that may be appropriate. But increasing output without reconsidering priorities can simply reproduce the same organizational problems at a faster pace.
Artificial intelligence may allow businesses to do more than ever before.
The strategic challenge is deciding what deserves to be done.
The Leadership Responsibility Behind Meaningful Improvement
This brings us to an important distinction in how organizations define leadership effectiveness.
Managers are often expected to improve productivity, control expenses, meet performance targets, and ensure that operational responsibilities are completed. These are legitimate expectations and essential components of responsible management.
But leadership requires an additional level of judgment.
It requires understanding how individual activities contribute to the broader organization, recognizing when established measurements no longer reflect meaningful progress, and having the confidence to challenge processes that appear successful on paper but produce limited value in practice.
This is not an argument against accountability. It is an argument for more thoughtful accountability.
A leader who questions whether a performance target is producing the intended outcome is not necessarily resisting measurement. They may be protecting the organization from optimizing the wrong behavior.
Similarly, a manager who invests additional time in employee development, quality assurance, or problem prevention may appear less efficient when evaluated through a narrow productivity measure. Yet those investments can strengthen long-term performance by reducing mistakes, improving judgment, and developing a more capable workforce.
The challenge is that many of these benefits are not immediately visible.
The value of a prevented problem is difficult to measure because the problem never occurred. The benefit of an employee who has developed stronger judgment may emerge gradually. The impact of improved customer trust may not appear in a daily operational report.
These realities make leadership judgment essential.
Numbers provide evidence, but they require interpretation. Metrics can identify trends, but they cannot fully explain every condition influencing performance.
Leaders must understand what the measurements reveal, what they may overlook, and how they relate to the organization's larger objectives.
The most effective organizations are not necessarily those that measure everything. They are those that understand which measurements matter and why.
Redefining What Progress Looks Like
Perhaps the most important change organizations can make is to begin evaluating improvement through the relationship between activity and outcomes.
Instead of focusing exclusively on how quickly a process is completed, leaders can examine whether the process produces better results. Instead of celebrating an increase in output without qualification, they can consider whether that output improves quality, customer experience, employee capability, or financial performance.
This does not require abandoning traditional productivity measurements. It requires placing them within a broader understanding of organizational value.
In retail, that could mean evaluating replenishment productivity alongside product availability and customer experience. In customer service, it could mean considering resolution quality and customer satisfaction alongside handling time. In technology operations, it could mean examining reliability and business impact alongside the speed of deployment.
The objective is to develop a more complete picture of performance.
It also requires organizations to recognize that different measures operate on different timelines. Some improvements produce immediate savings, while others create value through stronger capabilities, fewer recurring problems, or better decisions over time.
A short-term productivity gain should not automatically outweigh a long-term operational consequence.
This is where leadership moves beyond the management of activity and into the stewardship of organizational performance.
It involves asking not only whether teams are working efficiently, but whether their work is advancing the purpose of the organization.
And sometimes, that requires making a decision that appears less efficient in the moment because it creates greater value in the future.
The Question Every Leader Should Be Asking
There is no question that efficiency will remain essential to business competitiveness. Organizations must continue finding ways to reduce waste, improve productivity, and make responsible use of their resources.
Artificial intelligence and emerging technologies will expand those opportunities considerably.
But as businesses become increasingly capable of measuring, automating, and accelerating their operations, leaders must guard against confusing movement with progress.
A faster process is not necessarily a better process. More completed tasks do not automatically create more value. A positive performance report does not always mean the customer, employee, or organization is experiencing a meaningful improvement.
The true purpose of efficiency should be to support effectiveness.
That requires leaders to look beyond isolated performance indicators and consider the broader consequences of the systems they manage. It requires asking whether technology is improving decisions, whether employees are spending their time on work that matters, and whether organizational improvements are producing results that justify the resources invested.
Most importantly, it requires recognizing that progress is not defined solely by how much an organization accomplishes.
It is also defined by what those accomplishments make possible.
As businesses continue embracing artificial intelligence, automation, and increasingly sophisticated performance analytics, the organizations that create lasting value may not be those that simply move the fastest.
They may be those that develop the discipline to determine which activities deserve to be accelerated, which processes need to be reconsidered, and which outcomes are truly worth pursuing.
Because an organization can become remarkably efficient at doing things that no longer create meaningful value.
The real measure of progress is not how much more we are doing. It is whether what we are doing is making a meaningful difference.
And that may be one of the most important leadership questions of the AI era.
About the Author
Mike Kehoe is an author and leadership practitioner whose work explores organizational culture, emerging technology, retail innovation, and the human side of business transformation. Through Leading Through the Algorithm™, he examines how organizations can connect technological advancement with effective leadership, professional development, and meaningful operational results.
Sources and Further Reading
Harris, Michael, and Bill Tayler. Don't Let Metrics Undermine Your Business. Harvard Business Review, 2019.
Sutton, Robert I., and Hayagreeva Rao. Rid Your Organization of Obstacles That Infuriate Everyone. Harvard Business Review, 2024.
McKinsey & Company. The State of AI: How Organizations Are Rewiring to Capture Value. March 2025.
CNA. Goodhart's Law: Recognizing and Mitigating the Manipulation of Measures in Analysis. 2022.