Think in Roles, Not People: Organizing a Growing Business

Why does so much depend on individual people in a growing company?

In a young company, the organization grows around the people. Whoever is good with numbers eventually does the invoicing, whoever likes the phone takes the inquiries. That works as long as everyone sits in the same room. With each new employee it gets harder, because nobody ever wrote down what actually belongs to Petra’s tasks. All that is known is that Petra does it.

If Petra is absent or resigns, the company first has to find out everything that depended on her. This leads to the guiding question of this analysis: Why should growing companies describe their organization in terms of roles rather than people, and how can they do so without creating bureaucracy?

To answer it, we first clarify the terms role and position. Then we introduce a thinking model, Mintzberg’s coordination mechanisms, which explains why organizing around persons eventually stops working. After that we review the evidence, weigh two perspectives, and finally describe the role map as a tool.

An organization that consists only of names has to be reinvented with every staff change.

What is a role, and how does it differ from a position?

Organizational psychology describes a role as the bundle of expectations that others direct at a person in their position. This definition comes from role theory, which also underlies the RACI matrix, as the guide to clarifying responsibilities with the RACI matrix shows. A position, by contrast, is the smallest organizational unit assigned to one person with an employment contract. A position can contain several roles, and a role can be filled by different people over time.

For collaboration, role clarity matters most. It exists when a person has adequate information about the responsibilities, roles, and tasks needed to do their work. That is how Zettna and colleagues (2025) summarize the common definition. When this clarity is missing, research speaks of role ambiguity.

The difference sounds academic but has a practical consequence. Whoever thinks in people asks: who does that? Whoever thinks in roles first asks: which result is needed, and who expects it? Only then comes the question of who takes on the role.

The thinking model: how companies coordinate according to Mintzberg

In 1979, Henry Mintzberg described how organizations coordinate their work. He distinguished five coordination mechanisms. With mutual adjustment, those involved coordinate directly. With direct supervision, one person tells the others what to do. With standardization, coordination is settled in advance, either through work processes, through outputs, or through the skills of those involved (Mintzberg 1979).

Very small companies use almost only the first two. People call something across the hallway, and where things get stuck, the owner decides. Mintzberg called this form the simple structure, and it remains the typical description of small firms (Meijaard, Brand, and Mosselman 2005). It is fast and cheap. However, it has a built-in limit, because the number of possible coordination paths between two people does not grow at the same pace as headcount but much faster.

When mutual adjustment is no longer enough, coordination lands with the owner. This is the crisis of leadership that Greiner (1972) describes as the first growth crisis and that the essay on the owner’s role in a growing business discusses. Roles are one way out. They do not standardize every move, but the outputs: a role states what is delivered and who expects it. How the person achieves that is left to them.

An analogy from the theater makes this tangible. The role of Hamlet has been fixed for more than 400 years, while its casting changes constantly. Every actor plays it differently, but nobody has to rewrite the play when a new one arrives.

What studies show about role clarity and structure

That unclear roles do harm is well documented. Meta-analyses find a relationship between role ambiguity and weaker performance that is moderate but stable (Tubre and Collins 2000). More recent work therefore asks less whether clarity helps and more how it comes about.

Zettna and colleagues (2025) surveyed 270 employees and 47 of their leaders in two disability care organizations in Australia. Role theory sees coworkers and supervisors as role senders who convey to a person what is expected of them. The findings fit this view: employees who received practical help with tasks from coworkers reported more role clarity. More role clarity went along with less psychological distress, less fatigue from constant change, and higher job satisfaction. The relationship was stronger when the leader had a clear understanding of their own role.

For a small company, the last result is notable. Role clarity starts at the top. An owner who cannot say what they are and are not responsible for can hardly convey clarity to the team. The authors also stress that roles consist not only of the job description but of expectations that keep changing in daily work. The study comes from a single sector and shows associations, not proven causes.

Structure as a whole matters too. Meijaard, Brand, and Mosselman (2005) examined a stratified sample of 1,411 Dutch small firms. They found nine typical structural forms and concluded that organizational structure is related to sales growth, profitability, and innovativeness. So small firms are not all organized equally simply, and how they are organized makes a difference.

Interchangeable cogs or stronger people? Two perspectives

There is a serious objection to thinking in roles. It says: the advantage of a small company is its people. Customers come because of Petra, not because of a role called invoicing. Whoever puts everything into roles makes people interchangeable and drives away precisely those who do more than required. This view has a true core. The essay on the owner’s role also describes the owner as a genuine advantage of small companies.

The second view says: roles protect people. Whoever knows what they are responsible for does not have to guess, is not blamed for gaps nobody assigned, and can take a vacation without the business grinding to a halt. The findings on role clarity support this reading.

I consider the second view the more robust one, with one condition taken from the first. A role describes what is expected, not how it is to be done. Precisely this leaves room for what makes a person special. It only becomes dangerous when roles turn into detailed activity lists. Then bureaucracy really does arise, and clarity gets lost in the detail.

The role map: a tool on one page

The tool of this analysis is the role map. It lists all recurring roles of the company on one page and records five items for each: the purpose, the results, the decisions the role may make on its own, the holder, and the deputy. For the decisions, the decision map from the guide on setting decision rights helps; for the deputy, the coverage matrix from the guide on organizing deputy coverage.

The table shows two things. Petra and Jonas each hold two roles, and the owner holds one and deputizes for another. And each role has exactly one holder. This follows the same rule as the RACI matrix, which requires exactly one A per activity: two holders of a role are none.

1. Collect results

   Write down which recurring results the company needs, from the quote to the paid invoice. Do not ask who does something, but what must exist at the end.

2. Bundle into roles

   Combine results that belong together into one role. A small company usually manages with 10 to 20 roles.

3. Five items per role

   Record purpose, results, own decisions, holder, and deputy. One sentence per field is enough.

4. Discuss with the holders

   Go through the map with every person who holds a role. Gaps and overlaps become visible in this conversation.

5. Use it with every change

   When someone joins, leaves, or is promoted, the role map is adjusted first, the position afterward.

What follows beyond the single company, and where the limits are

As shown above, thinking in roles is not a question of size alone but of coordination. A company with eight people who sit together every day does not need a role map. A company with twenty people at two sites does, even if it still feels small. A more general rule follows: the less the people involved see each other in daily work, the more has to be described in advance.

Roles have a second effect that goes beyond the organization. They make visible where responsibility lies, and that is part of transparency. A customer who knows who is responsible for their invoice does not have to call the owner. An employee who knows what they may decide alone does not have to wait. How roles and departments come together in one picture is covered in the analysis Does a small business need an organizational chart?.

The limits must be stated clearly. The studies cited do not come from small German companies, and they show associations rather than causes. Mintzberg’s mechanisms are a description, not a construction manual. And no role map replaces the conversation about whether the expectations of a role match what the person can deliver.

Conclusion: why roles come before people

The question was
growing companies should describe their organization in terms of roles rather than people, and how to do so without bureaucracy. The answer lies in coordination. Small companies coordinate by calling across the room and through the owner. With every new employee, coordination paths grow faster than the team until the owner becomes the bottleneck. Roles settle part of this coordination in advance by recording which result who expects.

Bureaucracy only arises when roles prescribe the how instead of the what. A one-page role map with purpose, results, decisions, holder, and deputy is enough. The sober observation is: people make a company good; roles make sure it stays good when they change.

The next step

1. Write down five results on which money or customers depend in your company.
2. Check for each result whether a role with exactly one holder is responsible for it or only a habit.
3. Record purpose, results, own decisions, holder, and deputy for these five roles in a table.
4. Discuss the table with the role holders and ask what is missing.
5. Write down your own roles too: what are you responsible for, and what explicitly no longer?