A boat can drift a surprising distance without anyone noticing. The bow may still be pointed toward the destination, the sails may remain properly trimmed, and the crew may feel that steady progress is being made. Beneath the surface, however, a current is quietly carrying the vessel sideways. Unless someone compares the intended course with the actual track, the difference may not become obvious until the boat is well away from where the crew believed it to be.

Organizations drift in much the same way. A process changes slightly to accommodate an unusual case. A temporary responsibility stays with the person who stepped in during an emergency. A report is modified to satisfy one request, then another. A policy remains formally in effect while daily practice moves around it. None of these changes appears large enough to demand a course correction, but together they can carry the organization far from its original intent.

Drift is difficult to recognize because the organization usually remains busy and productive. Work continues. Reports are delivered. Systems operate. Meetings are held. The vessel is moving, but movement alone does not confirm that it is making good way toward the destination.

Small Exceptions Become Normal Practice

Most organizational drift begins with a reasonable accommodation. A deadline cannot be met through the standard process, so someone creates a shortcut. A system does not support an important exception, so staff track it in a spreadsheet. A required review is delayed because the person responsible is unavailable. The immediate decision may be entirely sensible, especially when the alternative is to stop important work.

The difficulty begins when the exception is not revisited. The shortcut works well enough, so it is used again. New staff learn the workaround from experienced colleagues without knowing why it exists. Over time, the workaround becomes the real process while the written procedure remains unchanged. The organization now has two versions of the same operation: the documented one and the one people actually follow.

This gap may remain invisible for years because experienced staff know how to navigate it. They understand which steps can be skipped, which approvals are symbolic, and which spreadsheet contains the information missing from the formal system. The organization appears stable because the crew continuously compensates for the difference between the chart and the coastline.

Responsibilities Drift Toward the Willing

Work also drifts toward people. A capable employee helps resolve a problem because the assigned owner lacks time, authority, or experience. The intervention succeeds, so the same person is asked again. Eventually, the responsibility belongs to them in practice even though their role, workload, and authority have never been adjusted.

Organizations often describe this as flexibility or teamwork, and sometimes it is. A good crew helps wherever needed. The problem arises when temporary support becomes permanent dependence. Skilled people accumulate duties because they can be trusted to complete them, while the underlying ownership problem remains unresolved. Their competence allows the arrangement to continue long after it has stopped being fair or sustainable.

This is how the best helmsman ends up below completing paperwork while someone less experienced steers the boat. The crew may still function, but talent is no longer aligned with the work that most needs it. When performance begins to decline, leadership may conclude that more people or better technology are needed without first examining how existing skills have been distributed.

Measures Drift Away from Their Purpose

Measures are created to answer questions, but the questions change. A metric originally intended to highlight an operational risk becomes part of a monthly report. A target is established under one set of conditions and remains in place after those conditions have changed. A project status indicator is simplified for senior management until it no longer reflects the concerns that matter to the project team.

The measure continues to be produced because it is familiar, expected, and embedded in the reporting cycle. People may no longer remember the decision it was designed to support. They know only that the number is required each month. The organization gradually shifts from using the measure to understand performance toward managing the process required to produce the measure.

The same pattern can affect dashboards, risk scores, service levels, and key performance indicators. Each begins as a representation of something important. Over time, the representation can become detached from its original purpose while retaining the authority of an established instrument. The compass still gives a precise reading, but no one has checked whether it is still properly calibrated.

Technology Preserves Yesterday’s Decisions

Systems give organizational decisions a long life. A field added for a temporary reporting need may remain in an application for decades. A workflow designed around an old approval structure continues after the organization has been reorganized. Interfaces preserve categories that no longer reflect how the business thinks about its work. Each technical component carries assumptions from the time it was created.

As systems become more integrated, those assumptions spread. A classification developed in one application appears in reports, data warehouses, dashboards, and planning models. Changing it becomes difficult because no one can easily see every place where it is used. The organization may recognize that the definition no longer fits, but the effort required to correct it feels greater than the inconvenience of continuing.

Technology is often blamed for this rigidity, but the system is usually preserving choices the organization made and then stopped examining. The technical debt is real, yet beneath it lies a form of decision debt: old assumptions, unresolved ownership, and compromises that were never reconsidered after the conditions changed.

Artificial Intelligence Can Conceal Drift

Artificial intelligence may help people work across fragmented systems, inconsistent documents, and outdated processes. It can search many sources, translate between terminology, summarize conflicting material, and provide a coherent response where a person once had to assemble the answer manually.

That capability can be enormously useful, but it introduces a subtle risk. An intelligent system may make a drifting organization easier to navigate without helping it return to course. It can explain the differences among five procedures without forcing anyone to decide which one should govern. It can reconcile several definitions well enough to answer a question while allowing those definitions to remain inconsistent. It can reduce the visible burden on the crew while preserving the conditions that created the burden.

A smoother interface does not necessarily mean a simpler or healthier organization. Artificial intelligence can act as an exceptionally capable guide through accumulated complexity, but the organization must still decide whether the complexity belongs aboard. Otherwise, the newest instrument may conceal how far the vessel has moved from the intended track.

Drift Often Looks Like Adaptation

Drift should not be confused with all change. Organizations need to adapt. Conditions evolve, new risks emerge, and experienced people improve processes through practical judgment. A vessel that refuses to adjust for wind and current will not reach its destination simply because it remains loyal to the original heading.

The distinction lies in whether the change is conscious. Adaptation occurs when the organization recognizes that conditions have changed, considers the consequences, and deliberately adjusts the course. Drift occurs when many local decisions alter the course without anyone examining their combined effect.

This is why strong organizations do not merely enforce consistency. They regularly compare intended practice with actual practice and ask whether the difference represents useful learning or unresolved deterioration. Sometimes the written process needs to change because the crew has discovered a better way. Sometimes the workaround needs to end because it is consuming time, creating risk, or depending too heavily on individual memory.

Look at the Track, Not Just the Heading

A navigator compares the course being steered with the track actually made over the ground. The difference reveals the effect of wind, current, and other forces that may not be obvious from the helm. Organizations need similar points of comparison.

What does the policy say, and what do people actually do? Who formally owns the work, and who performs it in practice? What question was the report created to answer, and what purpose does it serve today? Which systems are considered authoritative, and where do staff go when they need a number they trust? These comparisons reveal the movement that ordinary performance reporting may miss.

The goal is not to punish deviation or force every activity back into an old design. It is to make the actual course visible. Once the organization understands where it has moved and why, it can decide whether to return, continue, or establish a new destination.

Course Corrections Should Be Routine

Organizations often wait for a major failure, audit, transformation program, or leadership change before reconsidering how work is organized. By then, the distance between formal design and actual practice may be substantial. Correcting it becomes disruptive because so many people and systems have adapted to the drift.

Smaller and more regular course corrections are less dramatic. Reports can be retired when their purpose disappears. Temporary assignments can be reviewed before they become permanent. Measures can be checked against the questions they were intended to answer. Procedures can be compared with actual practice, and outdated system rules can be documented for eventual removal rather than silently accepted.

Drift is inevitable because organizations are living systems. People respond to conditions, solve immediate problems, and create new ways of working. The danger is not that the vessel moves. It is that no one notices where it is going.

From Bearings to the Pilot Book

These first Bearings have considered the age of natural language answers, the accumulation of organizational weight, the difficulty of answering simple questions, the importance of trust, the limits of representations, and the slow movement of drift. Together, they point toward a practical question: how should an experienced navigator approach familiar conditions without pretending that every voyage is the same?

The Pilot Book turns from broad observations toward practical guidance. It begins with the local knowledge, recurring patterns, and judgment that help a crew recognize the waters it is entering and choose a sensible next move.