Careful Downstream of the Wrong Question

Christina Sitaras

Oct 13

7 min read

Organizations often associate rigour with doing more analysis.

More interviews. More reviews. More stakeholders. More approvals. More checkpoints. More discussion before a decision is made. Sometimes that additional scrutiny is exactly what is needed.

But rigour has another dimension that receives less attention: where in the decision process it occurs.

An organization can be extraordinarily careful and still manage risk poorly if most of its scrutiny is applied downstream of an assumption that should have been tested earlier.

The problem is not necessarily insufficient analysis.

Sometimes we are simply being careful about the wrong thing.

A Recruitment Process That Revealed a Different Problem

Consider a relatively ordinary organizational situation.

A growing company determines that it needs someone to support the integration of parts of its business. The role is conceived as a combination of project execution and broader integration work: coordinating activity, establishing structure, improving how work moves across the organization and helping create a more integrated operating environment.

Recruitment begins. The process is thoughtful. Multiple stakeholders participate. Candidates are assessed not only for capability, but for fit, motivation, seniority and long-term sustainability. Questions emerge about whether the role is sufficiently challenging, whether the candidate understands how much project management it contains, whether expectations are aligned and whether the position makes sense within the organization's current stage of development.

These are reasonable questions.

Eventually, however, a more fundamental question is asked:

What will this person actually do?

Not conceptually. Not eventually. Week by week. What work exists? What activities need to happen? What capabilities do those activities require?

When the hiring manager works through the role at that level, the conclusion changes.

The organization does not actually need the broader role it has been recruiting for.

It needs X, not Y. The recruitment process stops. The position is redesigned.

In one sense, this is good organizational judgment. The organization challenged its own thinking before making the hire. It did not allow the time already invested in recruitment to justify a decision that no longer made sense. Once the underlying need became clearer, it changed course. But there is another question worth asking:

Why did the exercise that ultimately determined the role occur after recruitment had already been underway for weeks?

The analysis was useful.

The sequence was not.

The Risk Was Upstream

By the time the role was reconsidered, considerable effort had gone into managing what we might call candidate-selection risk.

Can this person do the work? Are they too senior? Will they remain interested? Will they be satisfied with the scope? Do they understand the operational component? Will they fit the organization? Will they stay? What do different stakeholders think of them?

Again, these are legitimate questions. But every one of them sits downstream of a more fundamental assumption:

What work actually needs to be done, and therefore what role actually exists?

If that assumption changes, much of the analysis that follows it loses its relevance.

The organization had been rigorous. It had simply concentrated much of that rigour downstream of the wrong question.

Decisions Inherit Their Assumptions

Most organizational decisions can be understood through a relatively simple sequence:

Assumption → Decision → Execution → Consequence

We make assumptions about what is happening or what is needed. Those assumptions inform a decision. The decision activates work. The work produces consequences.

In practice, of course, organizational decision-making is rarely this linear. Execution generates new information. Consequences reveal assumptions we did not know we were making. Decisions are revisited. People adapt.

That is normal.

Organizations should learn through action. The problem arises when an assumption is material to the premise of the decision itself.

If the assumption is wrong, does the decision still make sense? If the answer is yes, the organization may reasonably move forward, learn and adapt. If the answer is no, that assumption deserves greater scrutiny before the system is activated.

In the recruitment example, many things could safely remain uncertain.

Exactly how the successful candidate would approach the work could be learned. Some responsibilities could evolve. Relationships could develop. Processes could be improved after someone arrived.

The organization did not need perfect knowledge before beginning recruitment. But whether the underlying need was primarily X or a broader Y role was different.

That assumption materially changed who should be hired.

It changed the capabilities required.

It changed the likely seniority.

It changed the candidate pool.

It changed how the opportunity should be described.

And ultimately, it changed whether the original role existed at all.

That is not ordinary execution uncertainty. It is premise-level uncertainty.

Not Every Question Belongs Upstream

There is an obvious danger in taking this argument too far. Organizations could respond by trying to eliminate uncertainty before acting.

That would be a mistake.

Some things can only be learned through execution.

A candidate may reveal a capability the organization had not considered. A customer conversation may change how a product problem is understood. A pilot may expose constraints that were invisible during planning. A transformation may uncover dependencies nobody could reasonably have anticipated.

Over-defining a decision before acting can be just as problematic as under-defining it. The objective is therefore not to answer every question earlier. It is to distinguish between different kinds of uncertainty.

Some uncertainty is appropriate to carry into execution because the organization can learn from it without invalidating the decision.

Other uncertainty sits underneath the decision itself.

A useful question is:

If we are wrong about this, does the decision still make sense?

If it does, move.

Observe.

Learn.

Adjust.

If it does not, the assumption deserves greater attention before significant downstream activity begins.

This is where proportionality matters. Good judgment is not maximal analysis. It is applying analysis proportionate to the uncertainty in front of us.

Reversibility Changes the Calculation

There is another dimension leaders should consider: reversibility.

Many organizational decisions can technically be reversed.

A requisition can be cancelled.

A project can be stopped.

A process can be redesigned.

A priority can be changed.

A structure can be reorganized.

That can create the impression that the original decision was relatively low-risk.

But the reversibility of a decision is not necessarily the same as the reversibility of its consequences.

A role can be withdrawn. The weeks already spent recruiting for it cannot be returned.

A project can be cancelled. The capacity diverted from other work cannot necessarily be reconstructed.

A transformation can change direction. The credibility lost through repeated changes may not immediately return.

A restructuring can be revised. The people who left because of it may already be gone.

The longer a decision operates inside an organizational system, the more other decisions begin forming around it. Resources are allocated. People adjust. Expectations form. Other opportunities are deferred. Plans begin to depend upon the decision continuing to be true.

The original decision may remain reversible while its consequences become progressively less so. That changes where rigour belongs.

When People Enter the System

This becomes particularly important when organizational decisions begin affecting real people.

Recruitment makes the dynamic unusually visible because both sides are making decisions.

Organizations allocate recruiter time, executive attention and budget. Candidates allocate time as well, but also attention, preparation, competing opportunities and expectations.

As a recruitment process deepens, it is reasonable for both sides to update their understanding of what is happening.

A candidate should not interpret interest as an offer. An organization cannot reasonably be responsible for every expectation a candidate develops. But neither is candidate investment entirely independent of the signals the organization creates.

The deeper and more substantive the process becomes, the more consequential an upstream error can become.

This principle extends well beyond recruitment.

Employees reorganize their work around announced priorities. Managers build plans around approved structures. Teams prepare for transformations. Customers adapt to product decisions. Leaders make commitments based on decisions made elsewhere in the organization.

Once people begin acting on an organizational decision, changing the decision does not automatically unwind everything that happened because of it.

That does not mean leaders should avoid changing course. Sometimes changing course is precisely the responsible thing to do.

It means the cost of discovering an upstream error changes once other people are downstream from it.

The Correct Decision Can Still Come Too Late

This is one of the more uncomfortable aspects of organizational judgment.

A leader can ultimately make the correct decision and still have made an earlier error.

Those two observations are not contradictory.

In the recruitment example, stopping the process was arguably the responsible decision once the actual work had been examined. Hiring someone into a role that no longer made sense simply because everyone had invested time in the process would have compounded the original problem. But making the correct decision at the end does not erase the question of why the premise was not sufficiently tested at the beginning.

Likewise, acknowledging the sequencing problem does not require turning anyone involved into a villain.

Capable people can act in good faith. Different leaders can hold reasonable interpretations of the same organizational need. New perspectives can expose assumptions that others missed.

A leader can recognize an error, correct it and take responsibility for it. And the process can still have been badly sequenced.

Organizational problems do not require incompetent people. Sometimes reasonable decisions simply occur in the wrong order.

Rigour Is Also About Sequence

Executives frequently have to act without complete information. The answer cannot be to analyze everything before doing anything.

Nor should every decision receive the same degree of scrutiny. The leadership challenge is more demanding than that.

It requires determining:

What are we assuming?

Which assumptions are material to the decision?

Which uncertainties can safely remain unresolved?

What can we learn through execution?

What happens if we are wrong?

How reversible is the decision?

And how reversible will its consequences be once other parts of the system begin responding to it?

Those questions help determine not simply how much analysis a decision deserves, but when that analysis matters most.

Sometimes speed is appropriate.

Sometimes experimentation is appropriate.

Sometimes the best decision is to begin with incomplete information and allow reality to teach us what planning cannot.

And sometimes the most important work happens before execution begins, because one unanswered question sits underneath everything that follows.

The discipline is knowing the difference.

Organizational rigour is not measured by how many checkpoints a process contains. It is measured, in part, by whether the organization is examining the right uncertainty at the right point in the decision. Because organizations can spend enormous effort managing downstream risk while leaving the premise itself relatively untouched.

They can execute thoughtfully.

They can involve capable people.

They can ask reasonable questions.

They can ultimately correct course.

And they can still discover that they were careful downstream of the wrong question.

A Question for Reflection

Where is your organization applying the most rigour today - and what assumption sits immediately upstream of it?