Never Abandon a Sound Objective Because of a Flawed Implementation.

“It didn’t work” sounds like a conclusion. Most of the time, it is only an observation. Before leaders change direction, they need to know what actually failed.

Failure is an outcome, not a diagnosis

Organizations have a tendency to collapse several different questions into one. A new process misses its target. A technology rollout creates frustration. A reorganization does not produce the expected improvement. A leader looks at the result and says, “It didn’t work.”

That statement may be accurate, but it is not particularly useful. What did not work? The objective? The assumptions behind the strategy? The strategy itself? The way it was implemented? The capability of the people executing it? The information they had? The conditions around them? Or did the organization simply decide too early that the result was final?

Those are very different failures, and they require very different responses.

Why leaders confuse implementation failure with objective failure

Changing the objective can feel decisive. It creates a visible response to a disappointing result and gives the organization a new direction. Sometimes that is exactly what leadership requires. But changing direction before diagnosing the failure can destroy a sound objective because the first implementation was weak.

I have seen the opposite problem as well. Leaders become so invested in an objective that every failure is blamed on execution. More resources are added, timelines are extended, and people are changed while the underlying objective or assumptions remain protected from scrutiny.

Neither response is disciplined. Learning requires leaders to be willing to challenge both the implementation and the objective.

A practical hierarchy for diagnosing failure

When an outcome disappoints, I find it more useful to work backward through the system than to jump directly to a new direction.

1. Outcome: What actually happened, and how does it differ from what we expected?
2. Implementation: Was the strategy executed as designed, or did actual practice materially differ?
3. Capability: Did the people closest to the work have the skill, authority, information, resources, and time required?
4. Strategy: If implemented well, was the chosen approach actually capable of achieving the objective?
5. Assumptions: What had to be true for the strategy to work, and which of those assumptions remain valid?
6. Objective: Given what we now know, is the outcome we are pursuing still worth achieving?

The sequence is not mechanical. Complexity rarely gives leaders a perfectly clean causal chain. But the hierarchy forces a better conversation than “it worked” or “it failed.”

When the objective really is wrong

The principle is not an argument for persistence at all costs. There are objectives that should be abandoned. Conditions change. New information appears. Costs become unacceptable. An objective can prove strategically irrelevant, economically irrational, legally or ethically unacceptable, or simply based on assumptions that no longer hold.

The discipline is to identify that explicitly. What evidence changed our view of the objective? What assumption was invalidated? What would have to be true for us to continue pursuing it?

A leader who cannot answer those questions risks replacing judgment with either stubbornness or reaction.

Learning Forward means separating the layers

This distinction is part of what I call Learning Forward in the Human-in-Complexity Framework™. The point of examining failure is not to assign blame or protect a previous decision. It is to improve the next decision.

That requires intellectual separation. A good objective can have a bad strategy. A sound strategy can be implemented poorly. A capable implementation can be undermined by missing information or changed conditions. And sometimes everyone can execute well and still discover that the original assumptions were wrong.

If leaders do not separate those layers, experience does not necessarily create wisdom. It can simply reinforce the wrong lesson.

AI makes this problem more important

AI implementation provides a current example. Organizations frequently talk about whether “AI worked” as though AI were a single intervention. It is not.

A company can choose the wrong use case, select the wrong tool, automate a process that should first have been redesigned, fail to integrate the technology into the actual workflow, neglect training, create poor incentives, remove necessary human judgment, or measure the wrong outcome. Any of those can produce a failed implementation.

The conclusion “AI didn’t work” tells us almost nothing.

The reverse is equally dangerous. Leaders can become committed to an AI initiative because of the investment already made or because the technology is strategically fashionable. A flawed objective does not become sound because the tool is sophisticated.

The Human-Constant AI Implementation Model™ addresses this by beginning with the human work system rather than the technology. The objective is not to deploy AI. The objective is to improve an organizational outcome. AI is one possible source of leverage.

Eight questions before changing direction

1. Is the objective still worth achieving?
2. Are the assumptions behind the strategy still valid?
3. Is the strategy capable of achieving the objective?
4. Was the strategy actually implemented as designed?
5. Did the people executing it have the necessary capability, authority, information, and resources?
6. Did conditions materially change during implementation?
7. Do we have enough time and evidence to judge the outcome?
8. What evidence would cause us to conclude that the objective itself is wrong?

That final question matters. If there is no evidence that could change a leader’s mind, the organization is no longer evaluating the objective. It is defending it.

The leadership discipline

Leaders own outcomes, but ownership does not mean reacting to every poor result by changing direction. It means doing the harder diagnostic work required to understand what produced the result.

Sometimes that diagnosis tells us to change the implementation. Sometimes the capability. Sometimes the strategy. Sometimes the assumptions. And sometimes it tells us the objective itself is no longer sound.

The discipline is knowing the difference.