An org chart answers important questions: who owns what, who reports to whom, and where formal departmental boundaries sit. It says much less about whom employees approach for advice, who connects two teams, or who keeps unusual work moving.
That is not a defect in the chart. It describes authority, not every daily interaction. Trouble begins when leaders treat the two as the same and try to explain living work through reporting lines alone.
What an org chart shows
A conventional chart records roles, management levels, departments, and formal relationships. It supports accountability, hiring, budgeting, and escalation.
Yet identical boxes can conceal very different operating systems. In one team, people seek help freely across roles. In another, nearly every question passes through the manager. Their charts may look alike while their speed, resilience, and workload do not.
Microsoft Research makes the distinction plainly: the org chart captures the official hierarchy but offers only a simplified view of how people actually collaborate.
What the chart leaves out
At least four flows remain outside the formal model:
- Help. Whom people contact when the playbook fails or the case is unusual.
- Knowledge. Who remembers why a decision was made, understands exceptions, and translates between functions.
- Influence. Whose opinion changes a decision before formal approval begins.
- Coordination. Who carries context across departmental boundaries and keeps work from disappearing between owners.
Missing ties matter too. Departments may share a process on paper while rarely exchanging knowledge. A new hire can sit inside a team on the chart and remain at the edge of its working network.
How informal networks form
People choose working relationships for more than title. They return to colleagues who respond, know the subject, take a question seriously, and help bring work to a conclusion. Repeated exchanges gradually create a network of trust.
Previous projects, physical proximity, language, professional communities, and a history of successful work all shape it. Reporting lines can change in a day after a reorganization; relationships may take months to follow.
In a Microsoft study using aggregated and de-identified email, Teams, call, and meeting data, formal groups sometimes separated into several actual working communities. The difference was not a quality score. Isolation could be harmful or entirely appropriate for the work. Context supplied the meaning.
Five roles hidden in the network
The hidden expert
The job title does not advertise rare expertise, but colleagues repeatedly bring this person difficult cases. The expert speeds up work and can also become a dependency if the knowledge never spreads.
The connector
This person links groups that would otherwise interact little, such as Product and Support or a regional office and headquarters. Losing the tie may matter more than the employee’s formal position suggests.
The informal leader
Others listen without a formal requirement to do so. The person can help a change take hold or strengthen resistance. Visibility and volume alone do not establish that influence.
The isolated person or group
They have few ties to the wider network. This may indicate weak onboarding, an organizational gap, or simply autonomous work. A graph alone cannot diagnose the cause.
The overloaded hub
Questions, approvals, and cross-team contacts converge on one person. Centrality can reflect real value, but a queue around one employee makes the system slow and fragile.
When the informal network becomes a risk
Informal ties are useful. They cover gaps in procedure and let an organization adapt. Risk begins when the workaround becomes the only route that works.
Investigate when:
- a decision regularly stalls without one intermediary;
- an expert answers the same questions instead of transferring knowledge;
- two departments communicate only through their managers;
- old relationships vanished after a reorganization and new ones did not form;
- employees are formally part of a team but receive little help or context.
The purpose is not to eliminate the informal network or rank the “most valuable” people. It is to learn which ties support the work, where no alternative exists, and who needs support.
How to see the real structure
Organizational network analysis (ONA) models people or teams as nodes and a defined type of interaction as a tie. “Defined type” matters: an advice network, an approval network, and a recognition network answer different questions.
Evidence can come from:
- a short network survey, such as “Whom did you approach for professional advice in the past month?”;
- aggregated meeting and message metadata with a lawful basis and clear governance;
- active signals in which employees record help, ideas, or problems themselves;
- interviews and observation of a selected process.
A useful map is always bounded by its question and period. It cannot reveal a person’s character, the quality of all their work, or their “true value” to the company.
Combining the org chart and ONA
- Record the formal model. Include departments, roles, managers, locations, and participation in the process under study.
- Define one type of tie. Do not collapse help, approval, and social contact into a single measure.
- Map a meaningful period. It should cover a normal work cycle rather than only a crisis or holiday.
- Overlay formal attributes. This shows where working communities match departments and where they cross boundaries.
- Validate the hypotheses. Discuss observations with participants and add workload, timing, quality, and process evidence.
- Change the process and observe. Build a second line of expertise, create a direct team-to-team tie, or support an isolated group, then check whether the network changes.
HiveHR complements the formal chart with active signals. Thanks reveal who helped and why; ideas and problems show the subjects around which action forms. The platform can highlight leaders, brokers, and isolation, but these remain prompts for investigation rather than automatic employment decisions.
Frequently asked questions
Does a mismatch mean the org chart is wrong?
No. Formal and informal models describe different aspects of the organization. The question is whether the real network supports the work and whether formal responsibility matches actual load.
Should the manager be the most central person?
Not necessarily. A healthy team may distribute expertise and coordination. A more concerning pattern is one in which every question must pass through the manager and the team has no alternative ties.
Can a network be built from messages and meetings?
Yes, but metadata indicates contact or frequency, not meaning or quality. It requires transparent rules, a lawful basis, data minimization, and careful interpretation.
How often should the two structures be compared?
Revisit them after a reorganization, rapid growth, a leadership change, or a team merger. For a critical process, recurring observation with the same question is more useful than an occasional attempt to measure “every relationship.”
The org chart and the network belong together
One assigns accountability. The other shows how help, knowledge, and coordination actually travel. Comparing the two can expose a risk before it stops the process.
Sources
- Microsoft WorkLab: Patterns Hidden Inside the Org Chart
- Rob Cross: What is Organizational Network Analysis?
- HiveHR: platform guide