A process has been running the same way for months. Everyone knows the normal path — who submits what, who approves it, how long it takes. Then someone asks a simple question: “Can we make an exception just this once?”

The request may be entirely reasonable. A deadline cannot realistically be met. A volunteer cannot satisfy one procedural requirement. A trusted employee needs a faster answer than the normal process allows. None of this is misconduct. It is ordinary organizational life.

But beneath the simple question sit several governance questions that rarely get asked out loud: Who can approve this? What was the rule protecting? Does this become precedent? Does anyone else need to know? Must it be written down? When does the normal process resume?

Most governance failures do not begin with someone announcing that the rules no longer matter. They begin with a reasonable exception that nobody governed.

Rules Are Written for the Normal Path

Procedures describe recurring, predictable work. They function well when inputs are typical, time is available, responsibilities are clear, and conditions resemble what the designer anticipated. Reality does not stay that cooperative. A customer’s situation falls outside the normal case. A key system goes down at the wrong moment. A qualified person fails to meet one procedural requirement that was never designed with them in mind. No procedure, however carefully built, can anticipate every circumstance it will eventually meet.

A resilient system does not pretend exceptions will never occur. It decides, in advance, how exceptions will be governed when they do. The alternative is not a system with no exceptions — it is a system that governs its exceptions by accident, case by case, under whatever pressure happens to be in the room that day.

The Exception Reveals the Real Authority Structure

When everything follows the standard path, nobody has to ask who holds discretionary authority. The exception exposes it. Who may alter the process? Who owns the risk? Who decides whether urgency justifies bypassing a control? Who is accountable if the decision goes wrong?

Responsibility means carrying the work. Authority means holding the legitimate right to decide. Accountability means answering for the result. These are not interchangeable, and organizations that blur them discover it at the worst possible moment.

The moment nobody knows who may authorize the exception, the organization has discovered a governance gap.

Not Every Exception Is a Problem

Flexibility is not weakness. Emergencies, safety concerns, accessibility needs, proportionality, and unforeseen consequences all create legitimate grounds for deviation. A system that refuses every exception is not more governed — it is simply more rigid, and rigidity has its own costs: good people leave, reasonable requests get denied for no defensible reason, and the organization loses the judgment it needs to handle circumstances its own designers never saw coming.

Governance is not the elimination of judgment — it is the disciplined placement of judgment. The relevant question is never “did someone use discretion?” It is “was the discretion bounded, and by whom?”

The Dangerous Exception Is the Invisible One

The greater risk comes from exceptions that are made but never named: “We always let him do it that way.” “That approval is technically required, but we usually skip it.” “There wasn’t time to document it.” Left unnamed, these accumulate into a shadow operating system — a second, unwritten set of rules running underneath the formal one. Policy loses credibility. Decisions become personality-dependent. New people cannot learn how work actually happens.

A rule that everyone routinely bypasses is no longer governing the organization. The workaround is.

If your organization depends on knowing who can bend which rule and when, the issue may not be the people. The decision structure may need to be designed.

Explore Systems & Structure Design

Exceptions, Precedent, and Policy Are Not the Same Thing

An exception is one case intentionally treated differently. A precedent is a prior decision cited as evidence for how future cases should go. A policy change is a deliberate revision of the standard itself. Organizations blur all three when a one-time accommodation quietly becomes “but you allowed it last time.” If an exception should change the rule, change the rule on purpose. Repeated exceptions should never be left to rewrite the system in secret.

Favoritism Is an Exception Without a Defensible Principle

Two similar cases arise, and they receive different treatment. That may be legitimate — but only if a defensible principle explains the difference, such as authority level, material risk, or documented emergency. Title, friendship, seniority, or fear of confrontation are not defensible principles. The test is simple: could the two decisions sit side by side without embarrassment? This does not require identical treatment in every case — it requires a consistent principle behind whatever treatment is given.

A Governed Exception Path Has Five Elements

Roe & Associates Framework
  1. Trigger — the specific condition that justifies leaving the normal path.
  2. Authority — who is permitted to approve the deviation, stated plainly rather than assumed.
  3. Boundary — how far the exception extends, so one deviation does not quietly remove every related control.
  4. Record — what must be documented: what happened, who approved it, why, and what changed.
  5. Return — how and when the process comes back to the normal standard.
An exception without a return condition easily becomes the new operating model.

Repeated Exceptions Are System Data

One exception may be unusual. Five similar exceptions are evidence. They may indicate that a standard is unrealistic, that authority sits at the wrong level, that capacity is insufficient, or that the policy has simply become obsolete. Do not count exceptions only as failures — study them as operational data. A mature organization lets its exceptions improve the system rather than treating every one as a threat to defend against.

Urgency Is Where Weak Governance Becomes Most Dangerous

Urgency produces predictable language: “We don’t have time.” “I’ll take responsibility.” “Document it later.” Expedited action is sometimes legitimate, but high urgency increases the need for clear authority more often than it removes it. Speed does not eliminate governance — it changes the mechanism: emergency authorization, temporary delegation, and post-action review in place of the normal approval chain. The goal is not to slow the response. It is to preserve accountable decision-making while moving quickly.

Leadership Must Be Governed Too

Standards lose legitimacy when leadership routinely escapes them. This connects directly back to the standards themselves: a clear expectation that bends every time a leader is inconvenienced was never really a standard — it was a suggestion for everyone else. Founders, pastors, board members, and senior volunteers may legitimately hold broader discretionary authority — but broader authority should mean greater accountability, not exemption from it.

Who reviews a leader’s exceptions? What decisions require the board or the owner to be told? What financial threshold triggers a second signature, even for the person who owns the process? Who can say no to the person holding the highest operational authority in the room? These questions matter as much in a five-person nonprofit or a single volunteer club as in a large enterprise. The mechanism does not need to be elaborate — it needs to exist, and it needs to apply upward, not only downward.

The Exception Test

Select one recurring exception in your organization and answer each question in writing.

Practical Diagnostic
  • Purpose — What was the original standard designed to protect or accomplish?
  • Trigger — Why is an exception being requested in this case?
  • Authority — Who legitimately has the right to approve it?
  • Consistency — Would the same decision be made under materially similar conditions for someone else?
  • Risk — What risk increases if the exception is allowed?
  • Documentation — What must be recorded, and where?
  • Return — How does the process return to normal afterward?
  • Learning — If this exception keeps recurring, does the underlying standard need to be redesigned?

If those questions cannot be answered, the problem is larger than the exception. The governance model itself needs clarification.

Closing

“Can we make an exception?” is not an interruption to governance — it is one of the places governance becomes most visible. The ordinary path shows how work should proceed. The exception path shows who has authority, what the organization values, whether standards have a legitimate purpose, and whether the organization is capable of learning.

The strongest system is not the one that refuses every exception. It is the one that can make a necessary exception without losing sight of why the standard existed in the first place.

The rule tells you how the system works when conditions are normal. The exception tells you whether the system is actually governed.