Your leadership program may be asking managers to change behaviours that the rest of the organization keeps making difficult.
You teach supervisors to take ownership, but their decisions are reversed without explanation. You ask them to raise risks early, but another department takes days to respond. You train managers to follow through, yet the work disappears during handoffs they do not control.
When the new behaviour fails to appear, the usual response is to return the problem to L&D.
Add reinforcement. Provide coaching. Send reminders. Run another workshop. Require managers to submit action plans.
For CEOs, this creates a cycle of investment without movement. The organization funds development, but the surrounding system continues rewarding delay, escalation, silence, or rescue.
For L&D leaders, it creates an impossible promise. You are expected to improve accountability, execution, customer response, or quality even though the behaviour depends on authority, information, workflow, workload, measures, and decisions owned by other functions.
Leadership development can improve what leaders notice, decide, say, and do.
It cannot succeed alone when the work requires several parts of the organization to move together.
Leadership development is a cross-functional intervention.
L&D Could Teach the Move but Could Not Complete It
At Meridian Customer Operations, a fictional composite in Impact-First Leaders, Ana Reyes was preparing a small pilot for six supervisors.
The company had recorded 241 rework cases during the previous year. Ninety-two involved errors that had already appeared in the same team. Supervisors usually corrected the immediate output, but the same failure sometimes returned on another shift.
Ana’s team had found a useful supervisor response:
See the repeat. Find what made the wrong action easy. Correct and assign the next move. Check the next output.
This was a leadership contribution close enough to practise.
Supervisors could learn to distinguish a one-time mistake from a recurring pattern. They could examine the work with the employee, correct the immediate move, and place a barrier with someone who had the authority to remove it.
But the pilot could not end there.
What if the supervisor discovered that an obsolete product code remained active in the system? What if Operations and Quality had issued conflicting instructions? What if the next comparable output appeared during another shift? What if the supervisor escalated the barrier but no receiving function accepted ownership?
L&D could help the supervisor respond well.
L&D could not remove the obsolete code, reconcile the instructions, repair the handoff, or create authority in another department.
The leadership move required a field that allowed it to travel.
The Program Was Only One Part of the Response
Ana had four responsible choices.
She could train where supervisors lacked capability. She could recommend a system repair where employees already knew what to do but the environment blocked correct action. She could combine development and repair where both contributed to the failure. And she could recommend no training when another program would only add activity without touching the cause.
The recurring errors at Meridian did not fit neatly inside one category.
Some supervisors needed practice. They corrected the output but did not investigate why the problem had returned. They needed a better response they could use under pressure.
Other cases were mainly systemic. The right instructions were unavailable, obsolete choices remained visible, or another department owned the required repair.
Many cases were mixed. The supervisor needed to make a stronger move while another function changed the condition that kept defeating it.
That meant the pilot could not belong to L&D alone.
Operations had to provide real cases, protect practice time, and ensure that supervisors were not punished for exposing problems that made the numbers uncomfortable.
Technology needed a service owner who could accept a system issue, explain what would happen next, and respond within an agreed time.
HR and internal coaches could support difficult conversations and help supervisors reflect on actual attempts, but they did not need to deliver another complete coaching course.
Quality had to help distinguish an operator mistake from a conflicting or unusable instruction.
Finance had to help the team see the cost of rework and prevent the program from expanding faster than the evidence.
Each function owned a different move.
No one function owned the entire result.
Stop Treating L&D as the Owner of Application
Organizations often describe application as though it begins after training and belongs to L&D.
The facilitator delivers the program. Participants complete action plans. L&D sends reminders and schedules follow-up sessions. Managers receive guides for reinforcing the learning.
These practices can help, but they do not make L&D the owner of everything that happens next.
Suppose a supervisor has learned to diagnose a recurring error and create a clear handoff. The handoff reaches Technology, but the request enters a general queue with no response time. The supervisor has applied the learning. The work still fails.
L&D cannot solve that problem with another reminder.
Suppose a manager learns to make decisions within agreed boundaries. A senior leader later reverses the decision without explanation and criticizes the manager for taking the risk. The manager learns that waiting is safer.
L&D cannot repair that signal through a reinforcement email.
Suppose supervisors are taught to pause and investigate repeat errors, but their daily scorecard punishes every minute of stopped production. The training says one behaviour matters. The measure says another.
Application does not fail only because participants forget.
Sometimes it fails because the organization refuses to receive the behaviour it requested.
A New Behaviour Creates Work for Someone Else
This is easy to miss during program design.
When leaders begin behaving differently, their new moves often create obligations for other people.
If supervisors raise risks earlier, someone must receive and act on those risks.
If managers delegate more decisions, executives must stop taking those decisions back.
If team leaders give clearer feedback, employees need time and opportunity to make another attempt.
If leaders escalate fewer problems, decision boundaries must become clear enough for them to act responsibly.
Every desired leadership behaviour has a receiving condition.
This is why a program cannot be designed only around the participant. You must also ask what the new behaviour will require from the manager, the workflow, another function, and the executive sponsor.
Otherwise, you teach a move that has nowhere to go.
At Meridian, the supervisor’s improved response created a handoff. That handoff needed a named owner, usable evidence, and an agreed response time. If those conditions were missing, the case could still disappear even though the supervisor performed well.
The intervention therefore had to follow the complete chain:
Did a credible case appear? Did the supervisor recognize it? Did the supervisor make the critical response? Did the correct owner receive a usable handoff? Did that owner respond in time? Was the next comparable output checked? Did the error repeat?
That chain crossed functions because the work crossed functions.
Cross-Functional Does Not Mean Everyone Attends the Workshop
A cross-functional intervention does not require every department to sit through the same training.
That often makes programs larger without making the response stronger.
Technology does not need to attend a coaching workshop simply because a supervisor may escalate a system issue. Finance does not need to learn the supervisor’s four-step play. Operations does not need every HR framework.
Each function should contribute only what the business movement requires.
At Meridian, supervisors needed practice responding to repeat errors. Technology needed a service owner and a response agreement. Operations needed to protect practice time and change the signals around stopping work. HR coaches needed to observe difficult attempts. Quality needed to reconcile conflicting instructions.
The functions did not need one combined curriculum.
They needed coordinated moves serving the same result.
This distinction protects the intervention from becoming another leadership academy where every department adds its preferred subject. The business objective becomes the centre of gravity, and each function contributes only what helps move it.
Cross-functional does not mean everyone does everything.
It means the necessary owners do their part.
Give Every Function a Visible Contribution
A cross-functional intervention becomes vague when leaders merely agree to “support the program.”
Support must be translated into action.
Operations may commit to giving supervisors access to live cases, protecting thirty minutes for practice, and ensuring managers respond constructively when a case exposes broken work.
Technology may commit to assigning a named receiving owner and responding to qualified system barriers within two working days.
HR may prepare internal observers and help leaders handle the human side of correction without taking the operating problem away from them.
L&D may design the practice, make the play portable, support feedback, and hold the evidence chain together.
Finance may help estimate the cost of repetition, clarify what the pilot has earned, and prevent leaders from funding a company-wide rollout before transfer has been demonstrated.
These contributions are different, but they are connected.
This is impact-first work. Leadership development does not become more strategic because L&D uses more business language. It becomes strategic when the organization can see what leaders will contribute, what conditions other functions must create, and what evidence will guide the next decision.
Follow the Work Across the Handoff
During Meridian’s weekly review, the team discovered that supervisors were getting better at recognizing recurring errors. Their handoffs were also improving.
But the final check remained weak.
The obvious response was another practice session on follow-through.
Grace Bautista, one of the strongest supervisors, asked when the next comparable output actually occurred. In several cases, it ran during another shift. The original supervisor was no longer present, and the incoming supervisor had not received responsibility for checking it.
The failure looked like a leadership problem.
Part of it was a workflow problem.
Meridian added one question to its three-minute shift-risk check:
What open case must this shift verify?
No new workshop was required.
The following week, more next-output checks were completed. The review had exposed where the behaviour disappeared, and the organization repaired that part of the handoff.
This is why proof must travel beyond the learning event.
If L&D had measured only attendance, practice performance, or participant confidence, it might have concluded that supervisors needed more reinforcement. Following the case into the work revealed that another shift needed to receive the move.
The Sponsor Must Own More Than Approval
Executive sponsors are often asked to open the program, explain why it matters, and encourage participation.
That is useful, but sponsorship becomes real when the leader changes the conditions surrounding the behaviour.
An Operations executive may need to adjust a measure that rewards speed while making a quality response unsafe. A CEO may need to clarify which decisions managers can make without approval. A functional head may need to assign an owner to work that currently disappears between teams.
The sponsor’s job is not merely to lend authority to the program.
The sponsor must help make the new behaviour possible.
This is also where senior leaders demonstrate whether the intervention genuinely matters. Employees watch what happens when the first participant uses the new play and creates inconvenience.
Will the supervisor who stops production to investigate a credible risk be supported or blamed for lost output? Will the manager who makes a decision within agreed boundaries be trusted or overruled? Will the employee who raises bad news receive help or become associated with the problem?
Those moments teach more than the workshop.
What CEOs Must Demand
CEOs should stop asking L&D to solve cross-functional business problems through development alone.
When a proposal promises to improve accountability, decision speed, customer response, or quality, ask which functions shape the result.
What must leaders become able to do? What authority and information do they need? Who receives their handoffs? Which measure might punish the behaviour? What other function must act before the response becomes complete?
Then assign ownership.
The CEO does not need to manage every detail. But someone must ensure that departments cannot place the entire burden on the participants and the training team.
A cross-functional business problem needs a cross-functional response.
What L&D Must Lead
L&D should not retreat merely because it does not control the whole system.
Its role becomes more valuable.
L&D can help the organization distinguish capability gaps from system conditions. It can turn strong internal performance into a play other leaders can practise. It can identify what the behaviour requires from managers and receiving functions. It can collect evidence close enough to guide decisions.
Most importantly, L&D can refuse to accept an impossible promise.
A communication workshop should not be blamed for an ownerless queue. A coaching program should not be asked to remove an obsolete system option. An accountability workshop cannot compensate for responsibility without authority.
Practical leadership development programs should begin where the game is stuck, build the moves leaders need, and involve the owners of the conditions that allow those moves to travel.
That is not L&D abandoning training.
It is L&D protecting training from being used as a substitute for leadership elsewhere.
One Business Result, Several Necessary Moves
Meridian’s pilot became stronger when every function stopped treating the problem as someone else’s program.
L&D designed and supported the supervisor practice. Operations changed the field around that practice. Technology accepted system barriers. Quality repaired conflicting instructions. HR coaches supported difficult attempts. Finance helped the company decide what the evidence had earned.
The result did not belong to one department.
Each function owned a necessary move.
That is what makes leadership development cross-functional.
The purpose is not to involve more people for the sake of inclusion. It is to ensure that the behaviour taught in the room can enter the work, cross the handoff, and contribute to the result the business needs.
Train where capability is missing. Repair where the system blocks correct action. Combine the two when both must change.
But do not keep sending people back to training when the next move belongs to someone else.
About Jef Menguin
Jef Menguin is a leadership development consultant, professional speaker, and founder of Strategic Learning Consultants. He helps CEOs, HR leaders, and L&D teams connect leadership development to business goals and turn strategic priorities into daily leadership behavior.
Explore his work in leadership training, learn about his motivational speaking programs, or connect with him on LinkedIn.