The Principle
There is no universal right span. A manager with twelve experienced specialists in a stable operating rhythm may be fine. A manager with six new leads across ambiguous workstreams may be overloaded.
The question is not "how many direct reports is too many?" The better question is "what kind of management work does this group require, and can one person realistically do it?"
Rule of thumb: judge span by management load, not by headcount alone. The span number is the starting point for investigation, not the conclusion.
What Changes The Right Span?
Four factors determine whether a given span is sustainable. When all four point toward "easier management," wide spans work. When two or more point toward "harder management," the span needs to be narrower or the manager needs fewer additional responsibilities.
| Factor | Wider span may work when | Narrower span may be needed when |
|---|---|---|
| Work complexity | The work is repeatable and success criteria are clear. Think: customer support following playbooks, sales reps working named accounts, implementation consultants running a standard onboarding process. | The work is ambiguous, high-risk, or strategically new. Think: a product team exploring a new market, a strategy team with undefined deliverables, a team building the first version of a new platform. |
| Team maturity | People are experienced in their roles, know the domain, and need direction-setting more than day-to-day guidance. | People are new to the company, new to their role, or transitioning into unfamiliar responsibilities. New managers managing for the first time need especially close support. |
| Coordination load | Most work is contained inside the team. Dependencies are few and well-defined. The team can deliver without the manager attending cross-functional meetings. | The manager coordinates across many functions, stakeholders, or geographies. They spend significant time in alignment meetings, resolving cross-team conflicts, or representing the team in planning processes. |
| Manager role scope | The manager mainly sets direction and removes blockers. Someone else handles hiring, operational escalations, and process ownership. | The manager is also recruiter, coach, technical reviewer, escalation path, program owner, and budget holder. Each additional responsibility reduces the attention available for direct reports. |
Function-Specific Benchmarks
"Best practice" span benchmarks without context are misleading, but benchmarks with context are useful starting points. These ranges come from organizational design literature and patterns observed in practice. The ranges are wide because the right answer depends on the four factors above.
| Function | Typical range | What drives the variation |
|---|---|---|
| Software engineering (IC teams) | 5-9 | Lower end for teams doing novel architecture or platform work. Higher end for teams in maintenance or well-defined feature work with strong CI/CD and code review practices. |
| Engineering management (managers of managers) | 4-7 | Lower because managing managers requires more strategic work, hiring involvement, and organizational coordination. Each report is themselves managing a team. |
| Sales (quota-carrying reps) | 6-12 | Higher end when reps are experienced, territory-based, and mostly self-directed. Lower end for enterprise sales with complex deal cycles, or for new reps who need coaching on every deal. |
| Customer support | 10-20 | Support teams with clear playbooks, ticket queues, and SLAs can sustain wide spans. The manager's job shifts from coaching to capacity planning and escalation handling. |
| Customer success | 5-8 | Narrower than support because CSMs manage relationships, renewals, and expansion, which requires strategic coaching. Wider when the book of business is segmented and playbooks are mature. |
| Product management | 4-7 | PMs work across many stakeholders and the coaching is strategic. A Head of Product with 10 PMs reporting directly will struggle to be useful to any of them. |
| Executive leadership (CEO direct reports) | 5-8 | Below 5 often means there's a missing function or a layer that's not earning its keep. Above 8 means the CEO is the integration point for too many functions, creating bottlenecks on cross-functional decisions. |
| Operations / shared services | 8-15 | Stable, process-driven work with clear metrics supports wider spans. Narrower when the operations team is doing project-based or transformation work. |
How to use these: if a manager's span falls within the range for their function and the four load factors look reasonable, there's probably no problem. If the span is outside the range, or inside the range but load factors are unfavorable, investigate further.
The Span-Layers Math
Span and layers are directly linked. For a given organization size, wider average spans produce fewer layers, and narrower average spans produce more. This is arithmetic, not opinion.
| Organization size | Avg. span of 5 | Avg. span of 7 | Avg. span of 10 |
|---|---|---|---|
| 50 people | ~3 layers | ~3 layers | ~2 layers |
| 100 people | ~4 layers | ~3 layers | ~3 layers |
| 200 people | ~4 layers | ~3-4 layers | ~3 layers |
| 500 people | ~5 layers | ~4 layers | ~3 layers |
Why this matters: each layer adds a step between the people doing the work and the people making strategic decisions. Information gets filtered, decisions get delayed, and managers at intermediate layers may not add enough value to justify the overhead.
But removing layers without increasing spans is not possible. If you compress from 5 to 4 layers in a 200-person org, some managers will go from 5 reports to 8-10. You have to be sure they can handle the wider span before you eliminate the layer. See Layers and Organizational Depth for a deeper treatment.
How To Run A Span Audit
A span audit takes 2-4 hours for a 50-200 person organization. It's the fastest diagnostic you can run on an org chart. Here's the process:
Step 1: Pull the span data
For every manager in the org chart, count their direct reports. Sort the list from highest span to lowest. This gives you the raw data in about 10 minutes.
Step 2: Flag outliers
Mark any manager whose span falls outside the typical range for their function (use the benchmark table above). Also flag any manager with a span that looks normal but who is known to be overwhelmed, these are the hidden-load cases you'll investigate in Step 3.
Step 3: Assess load for each flagged manager
For each flagged manager, answer the span review questions below. This turns a number (the span) into a diagnosis (whether the load is sustainable).
Step 4: Build recommendations
For each unsustainable span, there are usually three options: split the team (create two teams with two managers), add a layer (promote or hire a lead beneath the manager), or redistribute reports (move some reports to another existing manager). Each option has tradeoffs. Splitting creates a new coordination boundary. Adding a layer adds management cost. Redistributing may break team cohesion.
Span Review Worksheet
For each manager you're evaluating, answer these six questions. The answers together tell you whether the span is sustainable, not any single question alone.
1. How similar is the work across reports?
Good answer: "All 8 reports are senior account executives running mid-market deals. The work is similar enough that coaching, pipeline reviews, and team meetings serve everyone." Concerning answer: "The 7 reports include a product ops lead, two analysts, a program manager, a UX researcher, a data engineer, and an admin coordinator. Each person needs different coaching, context, and career development."
2. How much coaching does this group need?
Good answer: "Most are tenured. They need 1:1s bi-weekly and an occasional strategic conversation." Concerning answer: "Three of six are in their first year. Two are new to the function. Weekly 1:1s aren't enough and I'm skipping them anyway."
3. How much coordination work does the manager do beyond managing?
Good answer: "Most of my time is spent on my team. I attend one cross-functional weekly." Concerning answer: "I own two planning processes, attend three cross-functional standups, and am the escalation point for any customer issue above Tier 2."
4. Is the manager also hiring?
Good answer: "No open roles right now." Concerning answer: "Three open reqs. I'm screening resumes, running interviews, and writing job descriptions while managing the existing team."
5. Where do decisions actually land?
Good answer: "My team can make most day-to-day decisions. I weigh in on quarterly priorities and escalations." Concerning answer: "Everything comes to me. Scope changes, vendor choices, team conflicts, customer exceptions, all of it."
6. What does the manager's calendar actually look like?
Good answer: "I have protected focus blocks and my 1:1s happen consistently." Concerning answer: "Every day is back-to-back. 1:1s get rescheduled weekly. I catch up on async messages at night."
If three or more answers are concerning, the span is probably unsustainable regardless of the number. If all answers are good, even a wide span may be fine.
Worked Example: Same Span, Different Load
Manager A leads eight senior implementation consultants doing repeatable customer onboarding work. The team has clear playbooks, known customer segments, and stable metrics. All eight do similar work. Five have been in the role for 2+ years. The manager's main job is capacity planning, escalation handling, and quarterly process improvements. She attends one cross-functional meeting weekly.
Worksheet assessment: Similar work (good). Low coaching need (good). Minimal coordination beyond the team (good). No open roles (good). Team makes most daily decisions (good). Calendar has focus time (good). Verdict: the span of 8 works.
Manager B leads seven people across product operations, analytics, launch management, and strategic programs. The work ranges from data pipeline maintenance to executive presentations to cross-functional launch coordination. Three reports are in their first year. The manager also owns the quarterly planning process, is the escalation point for two product teams, and is hiring for two open roles.
Worksheet assessment: Highly varied work (concerning). High coaching need for 3 of 7 (concerning). Heavy coordination load (concerning). Actively hiring (concerning). Decisions centralize through the manager (concerning). Calendar is wall-to-wall (concerning). Verdict: the span of 7 is unsustainable.
Options: (1) Split the team into two groups: "Product Operations" (ops + analytics) and "Programs" (launch management + strategic programs), each with a lead. (2) Move the planning process ownership to a chief of staff or program director. (3) Pause hiring until the current team is stable, then backfill with the explicit goal of reducing the manager's coordination load.
Warning Signals
These patterns on an org chart should trigger a span review. They're not automatically bad, but they're worth investigating.
- The fan: a manager with 12+ reports appearing as a wide fan shape on the chart. Even if the load is manageable, ask whether this person is providing meaningful management or just collecting reports.
- The chain: a manager with only 1-2 direct reports. Unless this is a specialist role (CEO > CTO > VP Eng in a 30-person company), this layer may not be earning its keep. Consider whether the reports could roll up one level.
- The mixed bag: a manager whose reports do fundamentally different kinds of work. Engineering manager with direct reports in engineering, design, QA, and DevOps. This person can't provide expert coaching to any of them.
- The bottleneck: important decisions across several teams all route through the same overloaded leader. This is often visible as multiple teams fanning out from one node, with that person appearing in every cross-functional meeting.
- The phantom layer: team leads or senior ICs who function as managers but don't appear in the chart as managers. Their actual span is hidden, which means their load isn't being evaluated.
Common Mistakes
Using one benchmark for every function
A span of 10 is fine for a customer support manager with an experienced team. It's too high for a product management leader. Different functions require different amounts of coaching, review, and strategic context.
Ignoring hidden management work
A manager with 6 direct reports can be overloaded if they also own planning, hiring, escalations, and cross-functional alignment. The span number only captures one dimension of the load.
Adding layers without decision rights
A new manager layer helps only if it creates real capacity: clearer ownership, faster decisions, or genuine coaching that wasn't happening before. Adding a manager who doesn't own decisions just adds a pass-through.
Optimizing span in isolation
Narrowing spans always adds layers (or costs). Widening spans always increases manager load (or reduces management quality). The question is never "what's the right number?" but "what tradeoff are we willing to make?"
Sources And Further Reading
- Henry Mintzberg, Structure in Fives: Designing Effective Organizations.
- Jay R. Galbraith, Designing Organizations: An Executive Guide to Strategy, Structure, and Process.
- Richard Hackman, Leading Teams: Setting the Stage for Great Performances.
- Matthew Skelton and Manuel Pais, Team Topologies.
Continue Reading
Span and layers are two sides of the same tradeoff. Once you can audit spans, the natural next step is understanding organizational depth.
Next guideLayers And Organizational Depth
How to count layers, when to compress them, and the tradeoff between decision speed and management capacity.