The problem nobody writes down
Most academic institutions have an organisation chart, a set of job descriptions and a statute or ordinance that names the bodies which approve things. Very few have a clear, shared answer to a simpler question: for this decision, who does the work, who carries the can, who must be asked and who merely needs to be told?
The answer usually exists only in people’s heads, and different heads hold different answers. A director believes the dean owns the timetable; the dean believes it sits with the programme office; the programme chair assumes the area chairs will supply faculty; the area chairs assume they will be asked. Each belief is reasonable. Together they produce a gap.
That gap is the no man’s land. In many institutions, a surprising number of critical activities fall into it: not because anyone is negligent, but because the activity sits between functions and no document places it squarely with one of them. Course outlines that are never reviewed before the term starts. Electives offered without anyone checking minimum enrolment. Assurance of Learning data collected by one office and never closed into a curriculum change by another. Student grievances that pass between the programme, the examination cell and the dean’s office until the student gives up. Accreditation evidence assembled in a rush in the final month because everyone assumed someone else was keeping it current. Each of these is an activity that everyone touches and no one owns.
Why clarity matters most at the level of director, dean, programme chair and area chair
Role ambiguity is costly anywhere in an institution, but it concentrates at four levels, for structural reasons.
The director or vice chancellor sets direction and holds final accountability to the board and regulators. When the director’s decision rights are unclear, two things happen: decisions escalate upwards that should never reach the top, and decisions that should reach the top are taken quietly below it. Both erode governance.
The dean sits between institutional strategy and academic delivery. The dean’s role is the one most often defined by what it approves rather than what it owns. Without clarity, the dean either becomes a bottleneck, signing everything, or an observer, informed after the fact.
The programme chair owns a student-facing product: a programme with its architecture, calendar, assessment and outcomes. The area chair owns a discipline: its faculty, its course content, its research and its standards. These two roles form a matrix by design. Every course sits in both a programme and an area. That is the institution’s strength, since it connects what students need with what disciplines know, but it is also where most boundary disputes arise. Who decides which faculty member teaches a core course? Who approves a change to a course outline? Who answers for a poor pass rate? When one person holds both a programme chair and an area chair role, as often happens in smaller schools, the question sharpens further: that person may end up recommending to themselves.
These four levels are where accountability is either anchored or lost. Faculty and staff below them take their cues from how clearly these roles are drawn. If the top of the academic structure is ambiguous, the ambiguity cascades.
The consequences are practical, not abstract. Unclear roles show up as delayed timetables, duplicated effort, decisions reopened after they were taken, accreditation findings about “lack of documented process”, and a quiet transfer of institutional risk onto whichever individual is most conscientious. They also make an institution fragile at transitions: when a dean or chair changes, the unwritten understanding leaves with them.
Background: the RACI matrix
The RACI matrix is the most widely used form of the responsibility assignment matrix, a tool that emerged from project and operations management practice in the second half of the twentieth century and is now a standard part of project management guidance. It lists the decisions or activities of an organisation as rows and the roles as columns, and places one of four letters in each cell:
- Responsible (R): the role that does the work.
- Accountable (A): the single role that answers for the outcome and has the authority to sign off. There should be exactly one A per row.
- Consulted (C): roles whose input is sought before the decision, in two-way communication.
- Informed (I): roles told of the outcome, in one-way communication.
Its value lies in the discipline it imposes. Every row must have an owner. No row may have two. Everyone else must be placed as either someone whose view shapes the decision or someone who simply needs to know. Many variants exist; some institutions add a D for the body that formally approves (a Board of Studies or Academic Council, for example), which is useful in academic settings where statutory approval sits apart from managerial accountability.
A RACI drawn up in a workshop, however, is only a statement of intent. The common failure is to treat the matrix as the answer rather than as a hypothesis. A row can look tidy on paper and still fail in practice, because the job descriptions do not actually give the accountable role the authority the matrix assumes, because too many roles are packed into one row, or because two roles each believe the row belongs to them.
Testing the matrix: reliability, complexity and maintenance risk
The AcademicOS diagnostic does not stop at drawing the matrix. It tests each row against three questions, each scored on a four-point scale from 0 to 3.
Reliability asks whether the assignment is actually backed by the institution’s own documents. If the matrix says the programme chair is accountable for programme architecture, do the programme chair’s job description, the statutes and the operating rules give that role the authority to decide? A score of 0 means the assignment is contradicted or unsupported by any written authority; 3 means it is fully and explicitly supported. Low reliability is the early signal of a row that will be contested the first time a hard decision comes along.
Complexity asks how many roles are live in the row and how much coordination it demands. A row with one responsible and one accountable role is simple. A row where the programme chair, two area chairs, the programme office, the examination cell, the IQAC and the dean all need to be involved is not. A score of 0 means one or two roles; 3 means many roles touching one line. High complexity is not necessarily wrong, since accreditation genuinely involves many people, but it tells the institution where coordination must be designed rather than assumed.
Maintenance risk asks how likely the row is to generate friction or need revision once people are working against it. The drivers are role overlap (two roles both claiming ownership, or a dual-hat holder recommending to themselves), undefined escalation paths, conflict of interest and single points of failure. A score of 0 means clean boundaries; 3 means a row that is likely to become a recurring dispute or a bottleneck. Maintenance risk is distinct from complexity: a row can be large but stable, or small but constantly contested.
The scoring is done row by row, against the full text of the relevant job descriptions, charters and rules, using a structured judgment model that returns a probability-weighted score and a confidence level for each axis. This makes the review consistent across rows, comparable across institutions and repeatable when the matrix is revised. The confidence level matters: a low-confidence score is a flag that the source documents say too little to judge, which is itself a finding.
What the three lenses reveal
Applied together, the three scores sort rows into patterns that a visual check of the matrix misses.
A row with low reliability and high complexity is the classic no man’s land: many roles involved, none of them clearly empowered. Programme architecture and curriculum structure often fall here, because job descriptions for programme and area chairs each claim “ownership” in general language without saying where one ends and the other begins.
A row with two roles marked responsible, or two roles each implicitly claiming accountability, will show high complexity and rising maintenance risk. Accreditation and Assurance of Learning are common examples, where both programme and area chairs describe accreditation as part of their remit.
A row with moderate complexity but high maintenance risk usually hides a sequence that the matrix has compressed. Faculty allocation is typical: the programme defines demand, the area recommends faculty, the programme office checks feasibility and the dean approves exceptions. A matrix that reduces this four-step process to a single R and A will generate friction every semester, and the friction is worst where one person holds both chair roles.
A row with low reliability and low complexity is often simply thin: an activity such as internships or placements that the job descriptions barely mention. These rows are easy to fix, but they are exactly where activities quietly drift into no man’s land because no one felt it was theirs.
In a recent review of a management institute’s programme and area governance matrix, no row out of twelve scored full reliability, which is normal, since job descriptions are written in prose and not row by row. The value lay in the spread: the weakest row on reliability was also among the most complex, the only row with a dual responsible assignment was the most complex of all, and the row with the highest maintenance risk was one that looked unremarkable on the matrix itself. None of this was visible from the matrix alone.
How the methodology creates clarity
The diagnostic produces clarity in four ways.
First, it surfaces the no man’s land explicitly. Activities that appear in no row, or in rows with no credible owner, are listed and assigned. This is often the most valuable single output, because these are the activities that fail silently.
Second, it aligns the matrix with the documents. Where reliability is low, the remedy is usually to amend the job description or charter so that written authority matches assigned accountability, or to move the accountability to the role that already has the authority.
Third, it designs coordination where complexity is unavoidable. High-complexity rows are broken into explicit sequences, with a named hand-off at each step, a clear decision point and a named escalation route to the dean or director.
Fourth, it manages overlap and dual roles deliberately. Where one person holds two chair roles, the diagnostic identifies the rows where that creates self-review and proposes a safeguard, typically a second signatory or referral to the dean for those decisions.
Because the scoring is repeatable, the institution can re-run it after revising the matrix and see precisely which rows improved, which did not, and which changed unexpectedly. Role clarity becomes something that can be maintained, not a one-time exercise that decays as people change.
The AcademicOS approach
The diagnostic runs in five stages.
- Map the decisions. With the director, dean and chairs, list the recurring academic decisions and activities: programme architecture, curriculum and course outlines, faculty allocation and workload, timetable, electives, assessment, moderation and results, student academic and disciplinary matters, internships and placements, and accreditation and Assurance of Learning. Add the activities people say “somehow get done”. These are the candidates for no man’s land.
- Draft or collect the RACI. Where a matrix exists, use it. Where it does not, draft one with the roles concerned, one accountable role per row.
- Gather the source documents. Job descriptions, statutes and ordinances, committee charters and operating rules: the written authority against which each row will be tested.
- Score each row on reliability, complexity and maintenance risk, with confidence levels, and identify the rows and activities that fall into no man’s land.
- Revise and re-test. Agree changes to the matrix and the job descriptions, re-score, and document what moved and why. The final output is a matrix the institution can adopt, backed by documents that support it, with a short record of the risks that remain and how they are managed.
The outcome is simple to state and hard to achieve: for every critical academic activity, everyone in the institution gives the same answer to the question “who owns this?”