Organisational structure is largely determined by how many people report to each manager, and the consequences are more mechanical than they appear.
Span of control
The number of direct reports per manager.
Which determines how many layers an organisation needs at a given size.
Narrow spans produce tall organisations; wide spans produce flat ones.
The arithmetic
Doubling the span roughly halves the number of layers required.
Which has direct cost implications.
Management cost as a proportion of payroll follows from this directly.
What determines appropriate span
Task similarity, employee experience, geographic distribution and how much coaching is required.
Which means there is no universal correct number.
Routine similar work supports wider spans than varied complex work.
Layers and decision speed
Each layer adds delay and information loss to decisions.
Which is why flattening is periodically attempted.
Removing layers without changing decision rights produces overloaded managers rather than speed.
Manager as bottleneck
Wide spans reduce the coaching and development each report receives.
Which affects retention and capability building.
Junior staff are affected substantially more than experienced ones.
Player-coach roles
Managers with substantial individual work alongside management.
Which is common in smaller organisations and produces neglected management.
The management portion is generally what gets dropped under pressure.
Promotion into management
Technical performers promoted without management training.
Which is extremely common and produces predictable difficulty.
Parallel technical career tracks address this and are unevenly implemented.
The design question
What decisions need making, at what speed, by whom, and how much development each person needs.
Structure follows from those answers rather than from a target span.
Matrix structures
Reporting to more than one manager.
Which addresses cross-functional coordination and creates conflicting priorities.
Clarity about who decides what is essential and is frequently absent.
Remote and distributed teams
Spans generally need to be narrower where informal contact is reduced.
Which is a real cost of distributed working.
Deliberate one-to-one time substitutes partially.
Manager training
Most managers receive little formal preparation.
Which is consistently identified in employee surveys as affecting experience.
Manager quality is the strongest predictor of team engagement in most research.
Reorganisation costs
Restructuring disrupts relationships and productivity for months.
Which is a genuine cost frequently omitted from the business case.
The design question
What decisions need making, by whom, at what speed, and what development each person needs.
Delegation
Pushing decisions to the level with the information.
Which reduces load and speeds decisions.
Managers who cannot delegate become bottlenecks regardless of span.
Skip-level meetings
Senior managers meeting people two levels down.
Which surfaces information that filters out through layers.
It requires care to avoid undermining intermediate managers.
Team size research
Coordination cost rises faster than team size.
Which is why small teams frequently outperform larger ones on the same task.
Communication paths increase quadratically with team members.
Individual contributor tracks
Senior roles without management responsibility.
Which retains technical expertise that would otherwise be lost to management.
The practical question
Not how many layers, but who decides what and how quickly.
What span of control means
How many people report directly to one manager.
Which determines how many layers an organisation needs.
Wide spans mean fewer layers and less individual attention per report.
What determines the workable span
Task similarity, employee experience and how much coordination is required.
Which means no single number applies across an organisation.
Experienced people doing similar work support much wider spans.
Layers and information
Each layer filters information travelling in both directions.
Which is why senior leaders in tall organisations receive distorted pictures.
Reducing layers is frequently framed as cost saving and is mostly about information.
Manager capacity
Managers with too many reports do not manage; they react.
Which shows up as unaddressed performance problems and unexpected departures.
The one-to-one
Regular individual time between manager and report.
Which is the mechanism through which most management actually happens.
Wide spans compress or eliminate it, which is where the cost appears.
Growth transitions
Points at which structure must change as headcount rises.
Which produce recognisable discomfort around certain size thresholds.
Adding a layer is usually delayed until it is overdue.
Flat organisations
Minimal hierarchy with very wide spans.
Which works at small scale and generally develops informal hierarchy at larger scale.
The hierarchy becomes undocumented rather than absent.
Manager selection
Promoting strong individual contributors into management.
Which is the default and requires different skills entirely.
Training rarely accompanies the promotion.
Reviewing structure
Look at where decisions stall and who has too many reports.
Which identifies the real constraint faster than an org chart review.
Restructuring without that diagnosis mostly moves boxes.
A closing caution
None of this is prescriptive. Businesses differ by sector, by scale and by stage, and practices that work well in one context fail in another for reasons that are not always visible from outside.
What is consistent is that the businesses handling these questions well tend to have written something down, measured it in a defined way, and reviewed it on a schedule rather than when a problem forces the issue.
Where a decision carries legal, tax or employment consequences, professional advice specific to your jurisdiction is worth the cost, and this article is general description rather than advice.