Organizational Modelling

Goal

Organizational modelling answers a planning question: who, in an organization, holds which role, whom do they report to and does this arrangement serve the goals being pursued. It produces an artifact, the org chart, and a reading of that artifact: the formal structure it draws, then the informal lines of influence it leaves out.

For the business analyst, the model feeds three pieces of work that fail without it. Stakeholder analysis first: you cannot identify the people a change touches without knowing which units exist and who makes them up. Communication planning next: a message does not travel the same channels depending on whether a decision moves down a reporting line or crosses three departments sideways. Preparing for user acceptance last: a solution founders less on its technical flaws than on the people the rollout disturbs, and those people are not always the ones the org chart places at the top.

The deliverable is twofold. The first is a chart, the org chart, which makes the reporting structure visible. The second is an analysis note that lists the influencers, the people whose real authority over the change does not coincide with their position on the chart. The chart is drawn in an hour from a directory. The note demands the investigative work: no other technique supplies it at this stage of planning.

Usage

When to use it

• Planning a change: knowing who holds which role and who reports to whom before building the communication plan and the stakeholder analysis.
• Reorganization, merger or new operating model: describing the target structure and comparing it to today's, position by position.
• Preparing user acceptance: spotting the influencers and the informal authority before an opponent the chart ignores blocks adoption.
• Framing a project team: recording who took part and what each role covered, so later projects find the right people.

When not to use it

• A stale org chart nobody keeps up to date: modelling a structure that no longer holds produces a fiction, elicit directly through interview and observation how the work actually flows.
• The problem is informal power and the formal structure is stable or beside the point: the model returns little, go straight to stakeholder analysis and influence mapping.
• A small co-located team: the reporting structure is plain to see and the effort returns nothing, describing the roles in two lines is enough.

Description

The elements of the model

An organizational model has four elements. The first is the role. A role carries a set of required skills and knowledge, specific responsibilities, kinds of work that fall to it and defined relationships with the other roles. Modelling a role means describing those four things, not merely writing a job title in a box. The second element is the interface: each unit exchanges with the others in two ways, through communication between the people who hold a role in it and through the work packages it receives from or delivers to another unit. A badly mapped interface is a handover of work with no clear recipient, where process changes get lost.

The third element is the org chart itself, the technique's fundamental diagram. The fourth, and the most important for the analyst, is the set of influencers, which carries the bulk of the work on its own.

The notation: box, solid line, dotted line

The org chart has no recognised standard. No norm fixes its syntax the way the OMG fixes that of BPMN, and this absence makes the few conventions all the more important: they carry the meaning on their own, and a reader who ignores them reads the chart wrong.

A box denotes an organizational unit (a team, a department, a division, a directorate) or a role held by a person. One chart often mixes the two at different levels, a directorate at the top, named roles at the bottom. A line denotes a reporting relationship, that is, a line of accountability and control. The distinction that decides the reading is the form of the line. A solid line marks direct authority, the reporting relationship: the superior sets objectives, arbitrates, appraises. A dotted line marks a transfer of information or a situational authority: the holder must report on a topic or for the duration of a mandate, without the person at the end of the dotted line being their boss. Confusing the two is the technique's first trap. Tying a project lead to a director with a solid line, when the relationship is merely functional, lends them an authority they do not have and skews the communication plan derived from it.

Three structures and their trade-offs

BABOK describes three pre-eminent organizational models. They differ by their grouping principle, and each trades one strength for a known weakness. The theory behind that trade-off is organizational: Henry Mintzberg established the typology of structural configurations and their coordination mechanisms, and Jay Galbraith formalised the matrix structure and lateral organization. The analyst borrows their framework to place an existing or a target structure.

The matrix is a third trade-off. It buys the dual perspective at the price of dual subordination: the employee receives priorities from their functional manager and from their project manager, and when the two diverge, the arbitration escalates or stays in limbo. Galbraith said as much long ago: a matrix works only if the organization invests in the lateral mechanisms that settle these conflicts (integration roles, arbitration rules, decision forums). Failing which it produces paralysis rather than coordination. For the analyst, the consequence is practical: in a matrix structure, the question « who signs off? » has no obvious answer, and underestimating it skews the entire governance plan for the change.

Influencers beyond the chart

The org chart shows only the formal structure. Yet the lines of authority, influence and communication that actually move a decision do not always coincide with it, and they sometimes conflict with what it displays. The executive assistant who screens the director's calendar weighs more on a rollout schedule than three department heads combined. The untitled subject-matter expert everyone consults before deciding is the effective validation point for a business rule. These actors appear nowhere on the chart, and they are the ones who decide acceptance.

Spotting them is the technique's central work, and it rests on two simple methods that BABOK names. The first: over the course of interviews, ask "who can I go to about…?" and note the answers. The same names recur, and they draw the real map of de facto authority, independent of titles. The second: observe who speaks for the group in meetings. The one everyone glances at before an answer, the one whose agreement closes the discussion, holds an authority their job title does not state. These observations are recorded in the analysis note, beside the org chart, because informal influence has no box.

Two traps lie in wait for the analyst. The first is to take the formal chart for the whole truth and build a communication plan that speaks only to the official post-holders: this is the commonest failure, because the decisive opponent was the influencer the chart did not show. The second is to work from a stale org chart without checking it against reality. Organizational models are often out of date, a reorganization outruns them faster than anyone redraws them, and a chart eighteen months old describes a company that no longer exists. The rule is to date the chart and to verify it with two or three people before drawing anything from it.

AI considerations

AI helps with the mechanical part, the one that produces the chart. From an export of the company directory or the HR management system, a tool generates a first org chart in seconds, where drawing it by hand took an hour. A language model then reads a batch of meeting minutes and proposes a first list of who speaks and on which topics, which gives a starting point for the search for influencers. Comparing two structures, the current one and the target, lends itself well to the tool: it aligns positions, flags the roles removed, created or moved and produces the gap table an analyst builds slowly by hand.

The limits are hard and they all bear on the heart of the technique. The informal influence map cannot be derived from any structured data: it is built through interview and observation, it rests on a reading of political nuance and trust, and a model that has never been in the room does not know whom people look at before answering. Handing that reading to a machine produces the official org chart a second time under another name. Input quality governs everything: a stale directory yields a false chart, presented with the misleading confidence of a clean drawing. Finally, an org chart is a set of personal data. Handing it to a cloud service, especially to attach observations about a given person's influence, is a processing operation subject to the Federal Act on Data Protection: the purpose must be bounded, and a note qualifying an employee's political weight has no business travelling further than the project that justifies it.

Examples

What decides how an org chart is read is the form of the line: the solid line carries hierarchical authority, the dotted line carries information or situational authority. The example shows the IT unit of a mid-sized company replacing its ERP, structured functionally. The post-holders stay generic: the notation alone is at stake.

The executive board sits above the IT department, which heads three functional units: Development, Operations and infrastructure, Support and service centre. These links are solid, they are hierarchical. A fifth actor changes the reading: the Finance business lead reports hierarchically to the Finance director, outside the IT branch, but holds on the replacement project a situational authority over everything touching the accounting scope. They are therefore tied to the project by a dotted line. That dotted line carries the decisive information: the lead is not in the IT department's reporting line, and yet no decision on the finance module is made without them. A communication plan speaking only to the solid boxes would forget them, and the rollout of the accounting module would founder on the actor the solid line does not show.

Auditing an existing org chart

The notation also serves as a reading grid for an org chart handed to you ready-made. Three questions put it to the test. Are the lines differentiated, or is everything drawn the same way? A single-line chart says nothing of situational authorities and hides them behind the look of a clean hierarchy. Does the chart carry a date? Without one, nothing guarantees that it describes today's organization rather than one from a past reorganization. Does an influence note come with it, or did it stop at the formal structure? These three gaps are the most common, and each is fixed before the chart serves as an input to anything.

Cost

Tooling

The minimum tooling is a whiteboard and a marker: a first org chart is sketched freehand in a workshop, often the best way to get a group reacting to who reports to whom. For a clean, shareable chart, diagramming tools (Microsoft Visio, Lucidchart, draw.io, Miro) are enough, and their connector styles natively distinguish the solid line from the dotted line, which is the notation the technique depends on.

HR management systems (SAP SuccessFactors, Workday, Abacus on the Swiss side) produce an org chart automatically from position data. It is convenient and it is a trap: the chart they draw is the declared structure, current as of the last import but blind to everything not captured in it, starting with situational authorities and informal influence. For large organizations steering reorganizations, dedicated tools (OrgVue, Ingentis org.manager) model target structures and cost the scenarios. The influencer map, though, remains an analyst's artifact: no tool produces it, because it comes from interview and observation, and it lives in the stakeholder analysis document rather than in the org-charting tool.

Sources

• IIBA, A Guide to the Business Analysis Body of Knowledge (BABOK Guide) v3, §10.32 Organizational Modelling: the purpose, the four elements (roles, interfaces, org charts, influencers), the three models, the solid line / dotted line convention, the strengths and limitations.
• Henry Mintzberg, The Structuring of Organizations (Prentice-Hall, 1979) and Structure in Fives: Designing Effective Organizations (1983): the typology of structural configurations and their coordination mechanisms, behind the functional and market models.
• Jay R. Galbraith, Matrix Organization Designs (Business Horizons, 1971) and Designing Matrix Organizations that Actually Work (Jossey-Bass, 2009): the reference authority on the matrix structure, dual subordination and the lateral mechanisms that make it viable.