A breakthrough in matrix communications


By Marc S. Caspe, President

M.S. Caspe Co.

San Mateo, California


Matrix management has been misunderstood and misrepresented for many years. Significant confusion exists because people have historically viewed matrix management as an alternative to “pure” project management and “pure” functional management, an alternative which simply allows both project management and functional management to contribute to the project.

A vivid example of this confusion was brought to my attention recently when a senior executive of a major company declined to join PMI because “PMI was biased towards project management, at the expense of the other two management alternatives, namely functional management and matrix management.” Obviously this executive had no understanding of organizational practice, and he’s not alone.

Confusion reigns! Numerous articles in the Project Management Quarterly have attempted to define matrix management as a simple crossing of project responsibility with functional responsibility. This simplistic approach is represented as a matrix of nodes, each node representing a group of unidisciplinary people who theoretically report to two bosses, one functional and the other line. As this concept of “two bosses” became commonly accepted the matrix concept became destined to failure. Many firms have been hurt by the breakdown of communications that arose when people were confused about both their authority to make decisions and that of their supervisor(s).

In truth the matrix concept does NOT assign each person to two supervisors. Instead it creates two linear hierarchies of report responsibility, one for the project organization (which eminates from the Project Manager) and the other for the corporate support organization (which eminates from the Chief Engineer). This concept is shown in a hierarchial format on Exhibit A and in matrix format on Exhibits B and C for the design organization and the construction organization respectively.**


The breakthrough in communications comes when project people understand their role and their decision-making authority under the matrix concept. To explain, let’s define the delegation of authority that eminates from a Company Officer who signs a design and/or construction management contract with an Owner and thereby carries the total legal responsibility for all project decisions made by his project staff.

Because the responsibility for decisions can not be delegated at any level of management the only thing that can be delegated is the authority to make decisions. How then to reliably delegate authority from the Company Officer to the front-line supervisors who must physically get the work done in design or in construction? How to permit young professionals with two to five years of experience to make value judgements for which a Company Officer is responsible and who has no time available to even be aware that the decision is being made?

The answer lies in delegating authority on a dual basis. This dual decision-making authority eminates to each level of management and calls upon peers (as they are shown above one another in the matrices and adjacent to one another in the hierarchial organization chart to agree on the proper course of action. Whenever peers can agree, they have a delegated authority to proceed without involving higher levels of management — their only obligation being to keep the next level of management informed. When they can’t agree, the decision MUST be passed up the ladder. However, when passing a decision up to the next level of management each subordinate supervisor is advised that he/she MUST commit his/her best judgement by recommending a solution. No supervisor should accept a problem from a subordinate until the subordinate has first recommended a solution.

In this matter, the matrix becomes a decision-making filter. It permits the vast majority of decisions to be made by the front-line supervisors, with review at the next level of management in a very conventional manner. The availability of Technical Specialists (who are identified by name for each specialty involved in the project, as shown on Exhibit D) provides a resource of the best expertise that the company can bring to the project in support of the project team. As these Technical Specialists are frequendy functional department heads, they have responsibility for their specialty on a number of projects in addition to administrative duties and marketing duties.

The matrix concept can therefore be seen as a means both for effectively utilizing people’s time and for allowing decisions to rise up through the hierarchy of the organization to a level at which two peers feel comfortable with the decision. It is this dual decision-making that builds reliability into the delegation of authority, as defined by matrix communications. Because we are dealing with value judgements, the role of the functional managers (i.e., as members of the corporate support organization) and project people must be clearly defined if they are to function as a team. It is this clarification of authority to make decisions that has led to improved productivity for our clients.


A Quality Assurance (QA) program, in both design and construction, is an important component of any management effort. QA is a concept which permits a supervisor to delegate authority while reliably maintaining responsibility for the action of others.

Simply put, QA consists of having two experienced engineers evaluate the same decision from two different perspectives. One perspective is that of the line organization, which is concerned with performing the scope of work within both the project budget and the project schedule. The other perspective is that of the corporate support staff, which is responsible for ensuring that the project work is adequate to meet each of the following:

  • All contracted commitments (including operational performance),
  • All legal restraints (i.e., building codes, OSHA, NEPA, etc.),
  • All criteria established for the project, and
  • Good practice (as is common to each trade or discipline involved).

When two competent engineers, one emphasizing quality and the other contractual commitments, can view a particular question and resolve it to their mutual satisfaction, the resulting decision should prove to be satisfactory. This is not to say that the decision will always prove to be optimal or that it will even prove to be correct in every instance. However, the QA procedure will provide each supervisor with a statistical assurance that of the myriad of decisions being made by his/her subordinates, those that gain dual concurrence are less likely to need reversal. In addition, he/she is assured that those decisions which don’t gain dual concurrence will be promptly identified for attention.

In order to reduce the time required to identify disputes, as well as the severity of those disputes, QA staff should be involved in the decision-making process at the start of the production effort. By establishing criteria at the outset, Technical Specialists can guide the production effort by becoming committed to the end product in advance of the expenditure of resources (e.g., man-hours). Milestone reviews, after the production work of the line organization has been completed, will invariably lead to dissension unless criteria are well thought through in advance and remain relatively constant.


The matrix chart for the design team (Exhibit B) shows that in the event of professional disagreements based upon differences in judgement at any level of the design organization, lines of appeal are available to both the Chief Engineer and the Project Manager. Each member of the organizational matrix is shown to have a single accountability, as well as a concurrent responsibility to keep his/her immediate supervisor informed. All parties are responsible for the progress, cost control and quality of the work. However, where line supervisors such as the Design Manager might frequently emphasize schedule and cost control, the Quality Assurance staff such as the Technical Specialists must place emphasis upon quality and operational performance. It is this difference in perspective, as applied by professionals having complementary experience, that is the basis of a Quality Assurance program.

A difference in judgement should not be allowed to delay the progress of the work. When differences do develop, the project should continue as directed by the representative of the line organization, while the staff organization (i.e., the Quality Assurance representative) negotiates through his/her supervisor to see of the next level of management agrees to a reversal of the decision in question.


The Technical Construction Specialists are shown on Exhibit A to be responsible for the Staff Inspectors who represent the heart of the QA program in the field. It is the Inspector’s responsibility to ensure that all work is performed in strict accordance with the vendors’ and constructors’ contracted duties and the performance requirements of the project.

The Technical Construction Specialists report to the Chief of Construction. By issuing independent reports about the performance of the field work, the Inspectors can alert all parties to potential problem areas and — in the event of professional disputes — lines of appeal are established, up through the project organization to the responsible Company Officer, who is legally responsible for the work.

As with the design organization matrix, this appeal procedure should not disrupt the work in the field. Maintaining planned progress, cost control and quality of the work are the responsibility of every member of the project team. However, where the Construction Manager might place emphasis on achieving schedule and budget targets, the Technical Construction Specialists and their corps of Inspectors must place emphasis upon quality and operational performance. This difference in perspective is the basis of a Quality Assurance program in construction.


The project organization shown on Exhibits A, B and C show only a part of the total project team. To completely define the decision-making process it is necessary to add the Owners organization hierarchy to the left-hand side of the matrices, thereby building in the hierarchy of the Owner’s decision-making process. The project is then seen as a coordinated whole, with the Project Manager as the man-in-the-middle and not at the top. Indeed that is the real world!

Many of our Owner clients have found that they do well at managing projects from the Project Manager on down but have a tough time getting balanced and timely decisions out of their own organization. This frequently results in a high “burn-out rate” for qualified project managers. By building this matrix decision-making concept into the Owner’s organization a true team effort can be achieved, eradicating lines of contractual separation.

A wise Project Manager will get the Owner’s organization involved in the project as Technical Specialists to the greatest extent possible. This is particularly important with respect to the Owner’s operations and maintenance people, who must be committed to the criteria of the project from the outset.

During a fast-track operation a very short response time is available to the Project Manager for decision-making. Therefore communications should be free and open between all supervisors on the project regardless of level, discipline or location. Supervisors who cannot delegate authority and who insist that the lines of responsibility shown on an organization chart also serve to prescribe lines of communiction will prove to be a detriment to the decision-making process.

The responsibility of subordinate staff members should be to make decisions on matters which they believe themselves to be secure in their judgement, and to routinely inform their supervisors of these decisions. This responsibility should be specifically delegated by each supervisor, with the understanding that the subordinate staff member must promptly consult his supervisor on those decisions for which he is unsure of his judgement. If a supervisor believes that specific limitations should be imposed on certain types of decisions, these should be defined in advance for each subordinate such as by imposing spending limitations.

A subordinate must always be required to present his/her supervisor with recommended decisions for review rather than presenting questions for resolution. Decision-making provides the supervisor with the subordinate’s best thinking. Questions demand considerably more time from everybody.


Just as the laws of magnetism state that an attracting force is inversely proportional to the square of the distance between two bodies, it can be postulated that communications between two people also vary inversely as the square of the distance between the parties (i.e., C = f (1/D2)).

The question of whether to form multidisciplinay project groups or to work in functional groups whould be resolved on this hypothesis as well as the size and duration of the project. On brief assignments for smaller projects it is best to maintain functional groupings. On larger projects where individuals are dedicated to the project for three months or more it is best to maintain project task groupings.

This three month criterion is based on the increased difficulty experienced when coordinating “between” functions, as opposed to “within” functions. Within functions, personnel have usually been trained for many years to perform their unidisciplinary tasks and frequently have longstanding working relationships with the Functional Group Leader and Technical Specialists in their discipline. Hence they can more readily communicate with one another in spite of the distances placed between them.

However, most unidisciplinary people rarely recognize the many interdisciplinary interfaces across which they must coordinate. In addition, because of the temporary nature of a project group, such communications are frequendy between people who have not related together in the past. Hence, the recommendation to form multidisciplinary project groups for each component task on a major project is based on strengthening communications across the weakest link, that of interdisciplinary coordination.

In either event it is important to recognize that whether a project is formed into unidisciplinary functional groups or multidisciplinary task forces, the matrix concepts defined in this paper remain intact. The only differences between a project group and a functional group is the question of seating arrangement and who gives the day-to-day guidance in the production effort. This can be reflected on the matrices of Exhibits B and C by interchanging the solid line and dotted line shown leading to the Task Leaders in design and the Foremen in construction. No other change is needed!


A final perspective of matrix communications is to emphasize its application to all projects — it is not limited to major projects alone. To exemplify the application of the matrix concept to small projects, Exhibit E is presented as a demonstration of the concepts’ applicability to all multidisciplinary projects.

The decision-making process should be viewed by all levels of management as a series of value judgements that are made by people having less than perfect foresight. The matrix concept defined here will serve to greatly strengthen that foresight. However the decision-making environment should never become a “search for the guilty,” enabling those with 20-20 hindsight to crucify those who had the guts to make a value judgement and to put their judgement on the line.

Management's attitude in this regard is very important. A search for the guilty approch emphasizes the negative incentives (i.e., Do well or you're fired). It is far better to emphasize the positive incentives and develop a self-motivating attitude amongst the project team, an envirnoment where people are encouraged to make decisions in a supportive envirnoment of open communications valid and timely decision-making can only occur where the project peoples' primary concern is the end-product of the team effort the project.







** Position titles and organizational structure are consistent between each Exhibit for purposes of demonstrating the matrix concept. On specific projects these should be changed to suit the requirements of each contract and each organization. For example, on projects where a more formal Quality Assurance Program is required the title “Chief of Construction” can become “QA Manager of Construction.”



Related Content