When an organization struggles to improve, the response is often to ask more of its people.
Managers are told to drive accountability. Employees are encouraged to take ownership. Everyone is expected to communicate better, embrace change, and become more productive. New training, technology, and performance targets follow.
But these responses can leave a more fundamental question unanswered:
What if the organization’s way of operating is the constraint?
People can work harder, acquire new skills, and adopt powerful tools while remaining limited by unsuitable models of leadership, learning, decision-making, and organizational design—in other words, how the work gets done.
Escaping the status quo begins with examining those models—not simply demanding better results from the people working within them.
The status quo is a set of assumptions
The status quo is more than familiar routines or resistance to change. It includes the assumptions that shape how an organization functions:
Managers give instructions; employees carry them out.
Training transfers knowledge that people will remember and apply later.
Improving each department’s performance improves the whole business.
Communicating a strategy ensures people understand how to execute it.
Poor performance primarily reflects insufficient effort, skill, or commitment.
New technology improves performance without requiring fundamental changes to how work is organized.
These assumptions deserve examination, not automatic acceptance or rejection.
The issue is not whether a model is old or new. It is whether it fits the organization’s objectives, operating conditions, and needs. A long-established practice may remain entirely appropriate. A fashionable management approach may be unsuitable from the outset.
The relevant question is not “Are we keeping up with the latest thinking?” It is “Are our ways of operating helping us achieve the right outcomes?”
AI makes the underlying questions harder to avoid
AI brings these questions into sharper focus.
It invites organizations to reconsider who—or what—performs an activity, how decisions are supported, where expertise resides, and what managers need to oversee.
But introducing AI does not, by itself, resolve unclear objectives, conflicting priorities, weak processes, or misplaced accountability.
Consider a team asked to use AI to produce reports faster. Before accelerating production, it is worth asking who uses those reports, which decisions they support, what information those decisions require, and whether a report remains the best way to provide it.
Otherwise, the organization may become better at producing something it should have reconsidered.
This is the distinction between applying new technology within the status quo and using technological possibilities to challenge it.
The more consequential opportunity is not merely to perform existing tasks faster. It is to rethink how the organization creates value—and build the capability to operate accordingly.
Treat performance as an organizational design question
When someone makes an error, misses a target, or repeatedly needs help, attention naturally turns to the individual.
But a useful investigation also examines the work:
Was the expected outcome clear?
Were the necessary information and support available?
Did the person have appropriate authority?
Were the instructions current and practical?
Did competing incentives encourage the wrong decision?
Was there a workable way to raise an exception or obtain help?
These questions do not remove personal accountability. They make accountability more meaningful by examining whether the organization has made good performance reasonably achievable.
The same reasoning applies when AI is involved.
If an employee is expected to review AI-generated output, “a human will check it” is not a sufficient operating model. What must they verify? Against which standards? Using what evidence? What happens when they cannot establish whether the output is acceptable?
Assigning responsibility without providing the means to discharge it is not enablement.
“Don’t blame; enable” expresses a broader management discipline: investigate the conditions producing the result, then address the causes rather than repeatedly treating the symptoms.
Shift leadership from directing activity to enabling outcomes
An organization can describe its people as empowered while still expecting them to wait for instructions, seek approval for routine decisions, and avoid challenging established practices.
Changing that relationship requires more than new language.
In an enabling model of leadership, managers are responsible for helping people achieve aligned objectives. That includes clarifying priorities, developing capability, removing barriers, coordinating dependencies, and ensuring that unresolved problems receive attention.
Employees, in turn, contribute more than task completion. They identify problems, exercise judgment within appropriate boundaries, and help improve how work is done.
AI adds another dimension to this relationship. Leaders must consider how work should be distributed among people and AI, what authority each role should have, and which decisions require review or escalation.
The leadership question becomes less “How do I get everyone to follow my instructions?” and more “What must be in place for sound decisions and reliable execution to occur?”
That is not less management. It is a different focus for management.
Make strategy usable in everyday decisions
A strategy is not operational simply because it has been announced.
People need to understand what the organization is trying to achieve, why it matters, and what it means for their own responsibilities. They also need to understand how their choices affect other roles and functions.
This requires cascading alignment: translating higher-level strategies and policies into coordinated responsibilities and actions at increasingly lower levels within the organization without losing the reasoning behind them.
Alignment also needs a return path. When conditions differ from expectations, priorities conflict, or a directive proves impractical, people must be able to obtain clarification and communicate what they are observing.
For AI initiatives, this means connecting proposed applications to business objectives rather than treating adoption itself as the objective.
“Use more AI” provides little guidance. A defined outcome—supported by clear responsibilities, constraints, and measures—provides a basis for deciding where AI belongs and whether its use is worthwhile.
Build capability into the work
Organizations often treat competence as something acquired separately from work through courses, workshops, and accumulated experience.
Some capabilities do require instruction and practice in advance. But not everything needs to be committed to memory, and completing training is not the same as being able to perform reliably.
Workflow Engineering offers a different emphasis: build the necessary expertise, support, standards, and checks into the workflow wherever practical.
This can include:
Guidance and reference materials available at the point of need.
Clear examples of acceptable output.
Defined decision criteria and escalation routes.
Feedback close enough to the action to guide improvement.
Quality and compliance checks incorporated into execution.
Mechanisms for capturing expertise and keeping instructions current.
AI can form part of this support—for example, by helping people locate relevant guidance, examine information, or prepare an initial analysis. But its role must be designed around the requirements of the work, including how its contributions will be checked and when human expertise is needed.
The objective is not to eliminate judgment. It is to create conditions in which judgment can be exercised well.
Optimize the whole business, not isolated activities
Making an activity faster or cheaper does not necessarily improve the business.
More production can create unwanted inventory. Faster sales can create delivery problems. Reduced handling time can leave customers with unresolved issues.
The same test should apply to AI-enabled improvements.
If AI accelerates content production, what happens to review capacity and quality? If it reduces the effort required to submit requests, what happens to the teams receiving them? If it automates a routine step, how will exceptions be handled?
These are systems questions.
Systems thinking evaluates an improvement by its contribution to the overall outcome, including the effects it creates elsewhere. It challenges the assumption that a collection of local efficiencies will automatically produce a better organization.
That perspective also changes measurement. Time saved on a task may matter, but so may accuracy, rework, customer outcomes, risk, and the capacity to deliver more valuable work.
The measure of success should follow the business objective—not merely the capability of the tool.
Make standardization and innovation complementary
Escaping the status quo does not mean keeping the organization in perpetual disruption.
Reliable execution matters. So does the ability to discover better ways of operating.
Standardization asks how improvements can be sustained. Innovation also asks whether different work—or a fundamentally different approach—would produce a better result.
Organizations need both.
Those responsible for maintaining dependable operations should help make changes workable and sustainable. Those pursuing new possibilities should remain accountable to business needs rather than novelty.
AI provides a useful test of this balance. Rejecting it solely because it disrupts familiar practices can close off valuable possibilities. Introducing it solely because it is available can create change without sufficient purpose—or even make things worse.
The task is to explore deliberately, evaluate consequences, and incorporate useful changes into a reliable operating model.
Escape deliberately
An organization-wide ambition does not require an organization-wide upheaval.
A practical starting point is an important business outcome that is proving difficult to achieve. Examine the work contributing to that outcome, the barriers people encounter, and the assumptions keeping those barriers in place.
Then redesign what needs to change: responsibilities, decision rights, workflows, support, measures, technology—or the relationship among them.
Implement in manageable steps. Evaluate results. Preserve what works. Revise what does not. Build the organization’s ability to repeat the process.
Leadership matters throughout. Leaders cannot credibly ask people to challenge established ways of working while exempting their own assumptions, decisions, and behaviors from examination.
Beyond a faster version of today
The central question is not how to make people work harder—or how to add AI to everything they already do.
It is how to create an organization in which people, expertise, leadership, and technology work together to produce better outcomes.
That requires more than adopting new tools. It requires the willingness to question familiar models and the discipline to replace unsuitable ones with practical alternatives.
Escaping the status quo is not change for its own sake.
It is refusing to let the way the organization operates today define the limits of what it can achieve tomorrow.