KanBo Perspective

Work Coordination: Meaning, Theory and Practice Inside the Organisation

What is work coordination?

Work coordination is the management of the dependencies between pieces of work, so that people, teams and departments contribute to one outcome instead of each finishing their own task. When someone asks what “coordinate work” or “coordination work” means, this is the core idea: work rarely fails because individuals do it badly. It fails because the links between tasks, owners, deadlines and resources are unclear.

In an organisation, coordination answers five plain questions. Why does this work exist? Who owns it? What does it depend on? What does it need in order to be done? What has changed since yesterday? Where those answers are visible to everyone involved, employees and teams coordinate with little effort. Where they live in inboxes, meetings and memory, coordination becomes a job of its own.

• Coordination is the management of dependencies between activities (Malone & Crowston, 1994).
• The right mechanism depends on the kind of interdependence: standardisation for routine work, plans for sequential work, mutual adjustment for reciprocal work (Thompson, 1967).
• It rests on predictability, accountability and common understanding (Okhuysen & Bechky, 2009).
• It works best inside a clear organisational boundary, where authority, governance and shared context already exist (Coase, 1937; Santos & Eisenhardt, 2005).
• Across the internal boundaries between functions, it needs shared and visible records of work (Carlile, 2002).

What does research say about coordination?

Fifty years of organisational research agree on one point: coordination is not a personality trait or a communication habit. It is a property of how work is structured. Five frameworks cover most of what a practitioner needs.

Framework	Core idea	What it means in practice
Malone & Crowston (1994)	Coordination is managing dependencies between activities. Typical dependencies are shared resources, prerequisites, simultaneity and task–subtask structure.	To improve coordination, find the dependencies first. Meetings come second.
Thompson (1967)	Work is pooled, sequential or reciprocal. Each type needs a different mechanism: standardisation, planning, or mutual adjustment.	One method does not fit all work. Routine work needs rules, staged work needs plans, and intertwined work needs frequent adjustment.
Mintzberg (1979)	Organisations coordinate through mutual adjustment, direct supervision, and standardised processes, outputs and skills.	Growth pushes coordination from conversation towards structure. The risk is that structure replaces conversation entirely.
Okhuysen & Bechky (2009)	Coordination means achieving accountability, predictability and common understanding.	Everyone should know who is responsible, what will happen when, and what the work is for.
Gittell (2002)	Relational coordination: shared goals, shared knowledge and mutual respect, carried by frequent, timely, accurate communication.	Tools and plans are necessary but not sufficient. The quality of the relationships between roles matters.

Two threads run through all five. First, coordination needs something shared: a plan, a record, a common picture of the work. Second, it depends on context: who is inside the arrangement, who has authority, and what counts as agreed. The next sections take both threads further.

How do people, teams and departments coordinate?

Employee, team and department coordination differ mainly in how much context the people involved already share, and that decides which mechanism works. The more context is shared, the more can be handled by a plain agreement. The less is shared, the more the work depends on a visible record.

Between employees

Staff coordination is mostly pooled or sequential work in Thompson’s terms: one person finishes something, another picks it up. The mechanisms that matter are clear ownership, agreed dates and explicit handovers. Okhuysen and Bechky call this accountability and predictability. The typical failure is a hidden dependency: two people each believe the other is waiting.

Within a team

Teams work with reciprocal interdependence. Each person’s output changes what the others should do next, so coordination happens through frequent adjustment. Gittell’s relational coordination fits here: shared goals, shared knowledge and mutual respect, carried by timely and accurate communication. A team that has to reconstruct the state of the work in every meeting spends its adjustment time on status, not on decisions.

Across departments

Departments differ in vocabulary, priorities and timelines. Mintzberg’s standardisation of processes and outputs helps, and so do plans, but neither is enough alone. Carlile (2002) describes why: knowledge held in one function is hard to transfer, translate and transform for another. The practical answer is a shared, visible record of work that every function reads in its own terms. The section on organisational boundaries explains why that record has to live inside the organisation.

Increase engagement step by step

Not every question deserves a meeting. A useful rule is to increase engagement only as far as needed. Check the record first: it is asynchronous and interrupts nobody. If it does not answer the question, message the person who owns the work. Only when that fails, call them. This keeps rich, synchronous channels for the reciprocal work that needs them, and it works only if the record is current and trusted.

Why does the organisational boundary matter?

Coordination works best inside a clear organisational boundary, because that boundary is what supplies authority, governance and shared context. The research does not test this claim directly. It is an inference from several bodies of work, which are set out here so you can judge the reasoning.

The firm as a coordination boundary

Coase (1937) asked why firms exist at all. His answer was that coordinating through authority inside a firm can be cheaper than coordinating through market prices and contracts. Williamson (1975) developed this into the contrast between markets and hierarchies. Inside the boundary, one line of authority can settle who decides, one set of rules applies, and accountability is traceable. Outside it, every coordination step is a negotiation.

A boundary is more than a legal line

Santos and Eisenhardt (2005) show that organisational boundaries can be understood in four ways: as boundaries of efficiency, power, competence and identity. For coordination this matters because each one shapes what people can rely on. Where the organisation controls the resources, holds the capabilities and gives people a shared identity, colleagues can assume common goals and common rules. Those assumptions are what allow coordination to proceed without renegotiation.

Boundaries inside the boundary

A clear outer boundary does not remove the internal ones. Carlile (2002), building on the boundary objects of Star and Griesemer (1989), describes three kinds of knowledge boundary between functions: syntactic, semantic and pragmatic. A shared record of work helps at all three levels. It gives functions a common vocabulary, exposes differences in meaning, and shows the dependencies that force trade-offs between them. A boundary is also not a wall: silos form when the record of work stops at the department edge.

What this means for a coordination tool

KanBo is built to operate inside the organisation, not beside it. It is a corporate tool tied to the organisation’s strategy and systems, not a standalone SaaS application that a team rents a corner of. That design choice follows from the research above, and shows up in several concrete ways:

• Governed access. Permissions run on six levels, cascade from Workspace to Space, and can be granted to groups, which follow the organisation’s own structure. Directory synchronisation keeps group membership aligned with the source of truth.
• Inside the organisation’s environment. KanBo can be deployed on the organisation’s own infrastructure, so the boundary of the work record matches the boundary of the organisation.
• Connected to existing systems. Document sources link Spaces to the repositories where documents already live, and Outlook and Teams sit alongside the work instead of competing with it.
• Strategy and execution on one structure. Workspaces reflect organisational structure, and parent and child relations connect day-to-day cards to the initiatives above them.
• Controlled crossing. Visitor access and shared Spaces let outside partners see what they need without joining the internal record.

When the boundary is clear and secure, people spend less effort asking who may see what and who decides. That effort goes into the work itself.

How can you improve work coordination across departments?

Improve coordination by finding the dependencies first, making ownership visible second, and matching the mechanism to the kind of work third. The six questions below turn the research into a diagnostic that a team lead, analyst or researcher can run without special tools.

Question	Framework behind it	A sign it is broken
Can anyone see who is responsible for each piece of work?	Accountability (Okhuysen & Bechky, 2009)	People ask in chat who owns something.
Are dependencies between departments written down, naming both sides?	Managing dependencies (Malone & Crowston, 1994)	Surprises appear at handover.
Is each kind of work coordinated by the right mechanism?	Interdependence types (Thompson, 1967)	Routine work is discussed in meetings, while intertwined work is run by fixed rules.
Does each function read the same record in its own terms?	Knowledge boundaries (Carlile, 2002)	Every department rebuilds the status of the same project.
Is shared capacity visible?	Shared-resource dependencies (Malone & Crowston, 1994)	The same people are committed to competing priorities.
Does the record of work live inside the organisation’s governed boundary?	Authority and boundaries (Coase, 1937; Santos & Eisenhardt, 2005)	Work is tracked in personal files or outside tools that others cannot see or govern.

The first two questions are the cheapest to answer, because writing down an owner and a dependency needs no new process. The later questions matter as the number of departments and projects grows.

Work coordination in KanBo today

KanBo turns the record of work into one governed structure inside the organisation, and each idea from the research has a concrete counterpart in daily work. The core of cards, relations, statuses and views has been stable for years. Resource Management, automation and MCP now add capacity, rules and AI agents to the same records.

Research idea	How it appears in KanBo
Accountability	Every card has a responsible person and co-workers. Comments and an activity stream keep the history of what changed and who decided.
Managing dependencies	Parent/child and predecessor/successor relations work within a Space and across Spaces. KanBo detects scheduling conflicts automatically, such as a child card ending after its parent or a predecessor that is not complete, and grades them by severity. Blockers are named, owned and carry a reason, and their impact can be traced across Spaces.
Predictability	Statuses are defined per Space, and each has a type (Information, Not started, In progress, Completed, Cancelled), so teams keep their own vocabulary while progress stays comparable. Kanban, List, Table, Calendar, Timeline, Gantt, Mind Map, Workload, Forecast Chart and Time Chart views show the same cards from different angles.
Common understanding	A card holds notes, checklists and documents, linked to the repositories where the documents already live. Each person chooses how much notification signal they receive, and MySpace collects a person’s cards from any Space in one view.
Shared resources	Resource Management works on the same cards. A planned role records the capability, amount and dates needed before anyone is named. Matching ranks candidates on skills, availability and preference, and explains its ranking. Allocations can be proposed, accepted or rejected when a Space uses approval. Equipment and crews are resources as well as people.
The right mechanism for routine work	Rules that a Space owner configures (KanBo Robot) run standardised steps automatically. Incoming email addresses turn forwarded requests into cards.
People and AI on one record	Through MCP, an AI agent works on the same governed data as people. The agent prepares and proposes from the live record, and a human applies judgement and decides.

These capabilities operate inside the organisational boundary described above. Because cards, capacity and AI assistance draw on one record, the plan, the resource picture and the assistant do not each carry their own version of the work.

References

• Carlile, P. R. (2002). A pragmatic view of knowledge and boundaries: Boundary objects in new product development. Organization Science, 13(4), 442–455.
• Coase, R. H. (1937). The nature of the firm. Economica, 4(16), 386–405.
• Gittell, J. H. (2002). Coordinating mechanisms in care provider groups: Relational coordination as a mediator and input uncertainty as a moderator of performance effects. Management Science, 48(11), 1408–1426.
• Malone, T. W., & Crowston, K. (1994). The interdisciplinary study of coordination. ACM Computing Surveys, 26(1), 87–119.
• Mintzberg, H. (1979). The Structuring of Organizations. Prentice-Hall.
• Okhuysen, G. A., & Bechky, B. A. (2009). Coordination in organizations: An integrative perspective. Academy of Management Annals, 3(1), 463–502.
• Santos, F. M., & Eisenhardt, K. M. (2005). Organizational boundaries and theories of organization. Organization Science, 16(5), 491–508.
• Star, S. L., & Griesemer, J. R. (1989). Institutional ecology, “translations” and boundary objects. Social Studies of Science, 19(3), 387–420.
• Thompson, J. D. (1967). Organizations in Action. McGraw-Hill.
• Williamson, O. E. (1975). Markets and Hierarchies: Analysis and Antitrust Implications. Free Press.