Home›Blog›Responsibility assignment matrix: who owns each piece of work
FundamentalsBeginner
Published September 6, 2026 · By Vincent KENNEL
Responsibility assignment matrix: who owns each piece of work
Two structures already exist on any contract programme, and neither of them says who answers for what. Crossing them does. What the crossing produces has a name, a single owner, and an object that existed for decades before that name reached it.
In brief
A responsibility assignment matrix, or RAM, crosses the work breakdown structure with the organizational breakdown structure. Control accounts are established where the two intersect, each with a single owner, at the lowest level where technical, schedule and budget responsibility come together. Work elements sit on one axis, so anything nobody owns shows up as an entirely empty column.
On a contract programme, a single work package can go missing in plain sight. An interface job, the qualification of a sub-assembly, sits between the engineering office and a subcontractor. Engineering reads it as the subcontractor's scope. The subcontractor reads it as engineering's. Nothing surfaces until a milestone review asks for progress and finds that neither side holds a budget or a schedule against it. The work was not done badly. It had no owner, and neither the breakdown of the work nor the chart of the organisation shows that on its own, because each is complete on its side.
What a responsibility assignment matrix is
Two structures exist before the matrix does, and they answer two different questions. The work breakdown structure answers what. The organizational breakdown structure, or OBS, answers who. Neither of them answers who answers for what.
The matrix is the crossing of the two, and five referentials converge on that reading. The NDIA Master Definitions List for IPMD Guides (2022) calls it a chart showing the relationship between work breakdown structure elements and the organisation elements responsible for ensuring work is accomplished, and adds that it identifies the control accounts established for the project. ISO 21511:2018 defines it as a documented structure showing the allocation of delegated work responsibilities, created by integrating the organizational breakdown structure and the work breakdown structure. The NASA Earned Value Management Implementation Handbook, revision 5 (April 2026), cross-references the two and states that the resulting matrix is the responsibility assignment matrix. The PMI Practice Standard for Earned Value Management, 2nd edition (2011), describes the same integration in clause 4.4.2 and locates the control account at the point where the two intersect.
The fifth is the one worth reading twice on a contract programme. AACE International Recommended Practice 83R-13 (revision of 1 May 2014) is the only definition in this set that states explicitly that the organisational hierarchy being crossed includes subcontractors. On a programme delivered through a supply chain, that is the case which decides.
Figure 1: where ownership becomes visible
Why the two axes must not coincide
The crossing carries information only because the two structures are built independently, and the referentials say so from both sides. MIL-STD-881F (2022) states in clause 3.1.3 that a work breakdown structure must not be influenced by a contractor's programme organisation, and that the contractor may organise to its own corporate standards while still reporting on a product-oriented breakdown. The NASA Work Breakdown Structure Handbook (June 2025) puts it the same way: the organisational structure should not drive how the work breakdown structure is subdivided. The PMI practice standard carries the counterpart on the other axis, in clause 4.3.1, where the project organisation is documented for itself and may or may not mirror the enterprise.
What follows is a reading rather than a quotation, and it is the reason the matrix exists at all. If the breakdown of the work followed the chart of the organisation, every work element would have exactly one possible occupant, the crossing would collapse onto a diagonal, and it would show nothing. Independence between the two axes is what produces the information.
What sits in a marked cell
A marked intersection is where control accounts are established. The NDIA definitions list describes it as a management point for planning and control: a portion of scope, a single work breakdown structure element, assigned to one responsible organisation element, and the minimum level at which technical, schedule and budget responsibility exist together. The EIA-748-D Intent Guide (2018) says the same from the guideline side, calling it the point where work breakdown structure tasks and organisational responsibility intersect, and MIL-STD-881F puts a person on it by requiring a control account manager to be named there. The PMI practice standard states the uniqueness rule outright: a control account can belong to only one work breakdown structure element and one organisational element.
One nuance travels with all of that, or the picture goes wrong. Several control accounts may sit inside the same responsible organisation element when scope has to be segregated for management control, and the EIA-748-D Intent Guide says so explicitly. One cell, one account is a teaching simplification rather than a rule: what is single is the ownership of each account, not the number of accounts behind a mark. Below the control account, scope is decomposed into work packages, which belong to the work breakdown structure and to its own article.
What it is for in practice
What it reveals
Four sources, forty years apart and none citing the others, state one rule: no gaps and no duplicates. Kerridge and Vervalin, in Engineering and Construction Project Management (Gulf Publishing, 1986), write that integrating the two structures assigns every element of work scope to a single organisation, with no omissions or duplications. The AMA Handbook of Project Management (Dinsmore, 1993) puts it as prevention: naming the responsible organisation for each element is what prevents gaps in responsibility assignments. The NDIA Guide to the Integrated Baseline Review, revision 3 (2019), has the matrix depict the assignment of each control account to a single manager. And John F. McCarthy had written the rule in its plainest form in Matrix Management for Aerospace 2000 (NASA TM-81509, 1980): only one organisational entity has prime responsibility for each task, though support may be required from others.
How that rule is read off the sheet is a reading of ours, and no source in the corpus formulates it this way. An empty cell is normal, and most cells are empty. What raises an alarm is an entire column with no mark, meaning a work element that nobody owns, or an entire row with no mark, meaning an organisation mobilised with no scope attributed to it. A column carrying two marks is a defect only when the scope behind it has not been segregated for the reason above.
Responsibility is not execution
AACE RP 83R-13 has the matrix name who is responsible for the work regardless of who actually performs it. The PMI practice standard adds the timing, in clause 4.3.2: at that stage, the individual assigned responsibility for a given element of work may or may not be identified. Clause 4.4.2 carries the sharpest formulation in the corpus. Several organisations may work on the scope inside a control account, but management responsibility and accountability belong to only one of them.
Two consequences follow, and both are practical. The matrix is a framing instrument rather than a staffing list, and it can be built before anybody is named. And it answers the question of who answers for a piece of work, which is not the question of who carries it out.
The mesh is a choice, not a given
How finely to cut the crossing is decided on people, not on structure, and two independent referentials give the same criterion. AACE RP 83R-13, discussing the matrix once budget values are in it, assigns additional managers where an amount is too large for one to handle and consolidates scope under existing ones where it is too small, the stated goal being to balance work scope and budget with the abilities of the managers. The PMI practice standard names both failure directions in clause 4.5.1: too high, and the size of the accounts overwhelms the people who own them; too low, and their number hampers the work and increases stakeholder intervention. Its criterion is the same one, the capability and span of control of the control account manager.
So a matrix that produces unmanageable cells is badly meshed. That is a statement about the matrix, not a verdict on how the project is organised.
What it becomes once the project is running
The matrix is not a start-up deliverable to be filed after kick-off. On the government side of a United States programme it is a numbered report format, Format 8 of the Guide to Cost/Schedule Management (joint services guide, version G, 15 May 1995), and an evaluation checkpoint in its own right: a reviewer is told to select several intersections from it and verify that every account sits at an appropriate organisational level and on a correct contract work breakdown structure element.
It then travels through the life of the contract. The NDIA EVMS Acceptance Guide, revision 3 (2019), lists the dollarized matrix as item 10 of the documents required to have a management system accepted. The NDIA Surveillance Guide, revision 3 (2018), keeps it as a source of evidence for the incorporation of changes. The NASA Integrated Baseline Review Handbook, revision P (February 2026), calls it the primary data artefact for beginning review planning, and has it requested from the supplier early.
Two things are easy to miss here. The matrix exists on both sides of the contract: the 1995 guide has the government programme office maintain its own, on the same breakdown, designating who in the buying organisation answers for monitoring each element for the duration of the contract. And AACE RP 83R-13 notes, in its section on security, that cost and schedule tools use the organisational hierarchy to grant access privileges, a superior position seeing the records of its subordinates, which is why it must hold clearance at least equal to all of them. In defence and space, the mesh of the organisation chart is also a confidentiality decision.
Where it comes from
The object is older than its name, and the two lines took decades to meet.
The crossing was operating by the early 1960s. The organisation status report of the DoD and NASA PERT COST guide showed, for each responsible organisation, the work packages within its responsibility and a breakout of the organisations that would actually perform them. The piece is Supplement No. 1 of March 1963, whose foreword ties those output reports to a base guide of June 1962. The crossing existed as a sorted report, not yet as a sheet.
The rule arrived in 1966, with the cost and schedule planning control specification published by the US Air Force Systems Command: the scheduling, budgeting, work definition and cost accumulation systems must coincide at the cost account level, and a work package is the responsibility of a single organisational unit (Lorette and Roth, 1970). The criteria that followed kept the rule and never named the sheet. DoDI 7000.2 of 10 June 1977 defines the cost account as the work assigned to one responsible organisational element on one contract work breakdown structure element, and the word matrix appears nowhere in it.
The name came from somewhere else entirely. The linear responsibility chart was deposited on 1 December 1953 by a firm of consulting engineers and reported the following year by Alfred G. Larke in Dun's Review and Modern Industry (vol. 64, September 1954): positions along one axis, functions along the other, a symbol at the intersection. Larke's warning has aged well. Despite the apparent simplicity of the charting, he wrote, it still requires expert knowledge to construct.
The graft of one line onto the other is visible sentence by sentence. In May 1980, the Department of Energy implementation guide for the cost and schedule control systems criteria (DOE/CR-0015) wrote that integration of the contract work breakdown structure and the organisational structure at cost account level may be visualised as a matrix, with the functional organisations on one axis and the applicable elements on the other. On 15 May 1995, the joint services guide carried that sentence again with a single clause added: this is called the Responsibility Assignment Matrix. Fifteen years, the same description, and the only thing added is the name. Outside government, Kerridge and Vervalin had already applied the name to the same object in 1986.
One more word changed, and it explains a good deal of what is read today. The cost account became the control account. The same inherited sentence appears in the DOE Work Breakdown Structure Guide (DOE/MA-0295, 1987) and in MIL-HDBK-881A (2005) with that one word altered and nothing else.
A responsibility assignment matrix is not a RACI chart. The two answer different questions: the matrix crosses two structures and produces control accounts, while a RACI attributes roles to people on activities. A RACI can refine a matrix, and only one of its four roles corresponds to the owner of a control account (AACE International Recommended Practice 83R-13). The overlap has a date. RACI entered the PMBOK Guide at its 3rd edition (2004), presented there as a type of responsibility assignment matrix.
The first pass to make on an existing project
Take the breakdown of the work as it stands, list the organisations actually working on the project, subcontractors included, and cross the two. Then read the columns before the cells. A column with no mark is work nobody has agreed to own, and it surfaces at a milestone review rather than in a status report. A column with two marks is either scope deliberately segregated for control, which is legitimate and should be recorded as such, or a disagreement nobody has named yet. Neither case shows in the breakdown alone or in the organisation chart alone, which is the whole reason the two are crossed.
Stop guessing. See the real impact.
Frequently asked questions
Q.What is the difference between the organizational breakdown structure and the matrix?
The organizational breakdown structure lists the organisations that will be given work; it is not what gives it to them. AACE International Recommended Practice 83R-13 states the distinction directly: the structure defines the elements to which work will be assigned, and the matrix is what assigns it.
Q.What happens when scope or the organisation changes mid-project?
The matrix is updated. Whenever work or budget moves into, out of or within the project, one or more control accounts change, and the PMI Practice Standard for Earned Value Management, 2nd edition (2011), requires that change to be reflected on the matrix.
Q.Is a dollarized matrix a different thing?
No, it is the same matrix with budget values in the cells. The NDIA, the DoD and NASA call that version dollarized; AACE calls it monetized. System acceptance, surveillance and baseline reviews all ask for the valued version rather than the bare one.
References
AACE International - AACE International Recommended Practice 83R-13 - Organizational Breakdown Structure and Responsibility Assignment Matrix - Rev. 1 May 2014
Arthur E. Kerridge, Charles H. Vervalin - Engineering and Construction Project Management - 1986
Department of Energy, Office of the Controller - Cost & Schedule Control Systems Criteria for Contract Performance Measurement - Implementation Guide (DOE/CR-0015) - 1980
DoD - Cost/Schedule Management Guide - Draft, Version G, 15 May 1995
DoD - MIL-HDBK-881A - Work Breakdown Structures for Defense Materiel Items - Revision A, 2005
DoD - DoDI 7000.2 - Performance Measurement for Selected Acquisitions - Edition of 10 June 1977
DoD - MIL-STD-881F - Work Breakdown Structures for Defense Materiel Items - Revision F, 2022
DOE - Work Breakdown Structure Guide (DOE/MA-0295) - Edition of 6 February 1987
Dun's Review and Modern Industry - Alfred G. Larke - Linear Responsibility Chart: New Tool for Executive Control - vol. 64, 1954
ISO - ISO 21511:2018 - Organigrammes des tâches en management de projet et de programme - 1st edition, 2018
NASA - Earned Value Management (EVM) Implementation Handbook - Revision 5, April 2026
NASA - Integrated Baseline Review (IBR) Handbook - Revision P, February 2026
NASA - John F. McCarthy Jr. - NASA TM-81509 - Matrix Management for Aerospace 2000 - 1980
NASA - NASA/SP-20250006071 - Work Breakdown Structure (WBS) Handbook - revision E, 2025
NDIA - Guide to the Integrated Baseline Review (IBR) - Revision 3 - 2019
NDIA - Earned Value Management System Acceptance Guide - Revision 3 - 2019
NDIA - Master Definitions List for IPMD Guides - Revision of 18 April 2022
NDIA - Surveillance Guide - Revision 3 - 2018
NDIA - EIA-748-D Intent Guide - Earned Value Management Systems - Revision D, 2018
Paul C. Dinsmore (direction d’ouvrage) - The AMA Handbook of Project Management - First edition, 1993
PERT Coordinating Group - Supplement No. 1 to DoD and NASA Guide, PERT COST - Output Reports - 1963
PMI - PMI Practice Standard for Earned Value Management - 2nd edition, 2011
PMI - A Guide to the Project Management Body of Knowledge (PMBOK Guide) - Third Edition - 2004
Richard J. Lorette, Berton J. Roth - Cost/Schedule Planning Control Specification - 1970
