You cannot design a better future if you don’t understand how the current organization actually works. This chapter walks you through a disciplined diagnosis: mapping structure, spans, and layers; understanding real processes and decisions; listening to the organization; and linking findings to performance and health. The output is a concise diagnostic summary and a set of design hypotheses that give the next phase a sharp focus.
A good diagnosis answers five questions:
- How is the organization structurally configured today?
- How does work and decision-making actually flow?
- How do people experience the organization?
- What is the performance and health of the system?
- What does all this imply for the design problem you need to solve?
This chapter walks through each of those lenses and ends with how to synthesize them into a concise set of design hypotheses.
6.1 Structural Mapping: Org Charts, Spans, Layers, Headcount
The starting point is a clean, accurate map of the current formal structure. Many teams assume they already have this. In reality, HR systems, financial systems, and local spreadsheets often tell slightly different stories. Your first task is to reconcile these into a single “source of truth.”
Focus on four structural elements: org charts, spans of control, layers, and headcount.
Org charts: what is formally on paper
Obtain the latest organization data from HR (or the closest equivalent): units, reporting lines, job titles, grades, and locations. From this, build a series of visual org charts at multiple levels:
- Top-level structure: business units, major functions, regions, and corporate center.
- Level below: sub-units within each business and function.
- Critical teams: where needed, detailed charts for hot spots (e.g., commercial, R&D, operations, risk).
Do not aim to produce a detailed chart for every team at this stage. Aim to answer: “Who reports to whom?” and “How many distinct units do we actually have?”
You will often find: informal “teams” that do not exist in HR; leaders with “matrix” relationships that are not captured; historically created units that no one can fully explain. Note these as potential issues, but do not try to fix them yet. Diagnosis is about seeing clearly before prescribing.
Spans of control: how many direct reports per manager
Next, calculate spans of control for each manager. For each person with at least one direct report, count the number of direct reports, including employees and long-term contractors if they are effectively part of the manager’s team.
Look at spans by level and by unit. You are not trying to hit a universal “magic number.” You are looking for patterns that signal potential problems:
- Very narrow spans (e.g., managers with one to three reports) in stable, routine environments may indicate unnecessary layering or inflated titles.
- Extremely wide spans (e.g., 15–20+ direct reports) in complex roles may suggest overstretch and insufficient coaching or control.
- Large variability in spans at similar levels in similar roles may indicate inconsistent design rather than purposeful choice.
Layers: distance from the top to the front line
Count the number of layers between the CEO (or unit head) and front-line employees. Be consistent in your definition of a “layer” (for example, everyone with managerial responsibility counts as a layer). Do this by major unit and by geography if needed.
Focus on:
- Total number of layers: Is the organization more layered than you would expect for its size and complexity?
- Layer pattern: Are there “micro-layers” where multiple levels are separated by small pay or responsibility differences?
- Role clarity: Are there layers where it is hard to explain what unique value that layer adds?
Excessive layering often shows up as slow decisions, diluted accountability, and higher cost. Too few layers in a large, complex operation can also be a problem, leading to overstretched managers and inconsistent control.
Headcount: where people, cost, and work sit
Finally, map headcount (and if possible cost) across the structural dimensions:
- By unit and sub-unit.
- By function and role in the family.
- By geography and location.
- By managerial vs individual contributor roles.
This gives you a picture of where the organization is “heavy” or “light” relative to its purpose. For example, you may find disproportionate headcount in coordination and internal support roles versus customer-facing or production roles; or you may see large variations in support ratios (e.g., finance, HR, IT, analytics staff per 100 front-liners) across units.
A simple structural diagnostic checklist
As you complete structural mapping, ask:
- Do we have a single, clean view of who reports to whom?
- How many management layers do we have from top to front line in each major unit?
- Where do we see very narrow or very wide spans of control?
- Where are we structurally heavy in coordination or support compared to value-creating roles?
- Which parts of the current structure are hardest to explain in terms of our strategy and value agenda?
You are not yet deciding what to change; you are sharpening your understanding of where the current structure may be misaligned or inefficient.
6.2 Process and Decision Mapping: How Work Really Gets Done
Structures tell you how authority is formally arranged. To see how the organization actually works, you must look at processes and decisions. Most recurring pain in organizations comes not from the existence of silos, but from poorly designed interfaces and unclear decision rights between them.
Process and decision mapping should be selective, not exhaustive. Focus on a handful of value streams and decision areas that are critical for your strategy or where pain is most acute.
Choose the right processes and decisions to map
Start by identifying:
- 3–5 critical value streams (e.g., “develop and launch new products,” “acquire and onboard customers,” “plan and deliver projects,” “resolve customer issues”).
- A short list of critical enterprise decisions (e.g., pricing for key segments, capital allocation, product prioritization, major risk and credit decisions, staffing of flagship projects).
These should link directly to your strategic priorities and the design objectives you defined earlier. If a process or decision does not materially affect your value agenda, it is probably not worth mapping at this stage.
Process mapping: how work flows across units
For each high-priority value stream, run a short series of working sessions involving people who actually perform the work, not just managers. Ask them to map how the process works today:
- Key steps from trigger to outcome.
- Who does what at each step (unit, role, or team).
- Hand-offs between units and functions.
- Approvals, checks, and rework loops.
- Pain points: delays, duplication, unclear ownership, exceptions.
You do not need beautifully formatted process maps. Simple boxes-and-arrows on virtual or physical whiteboards are enough as long as they are accurate. The goal is to see where the current structure creates friction: too many hand-offs, no clear owner, or approvals stuck at the wrong level.
Look for patterns such as:
- Activities bouncing between functions without a clear “process owner.”
- Units involved late in the process, forcing late rework (e.g., risk, legal, or IT brought in at the end).
- Over-reliance on informal workarounds and personal networks to get things done.
- Multiple parallel processes for similar tasks in different units, with no shared standard.
Each such pattern is an organizational design signal: either ownership is in the wrong place, interfaces are poorly defined, or the structure is forcing the process to “zigzag” unnecessarily.
Decision mapping: who really decides what
Next, map decision rights for a handful of critical decisions. For each decision, answer:
- What exactly is the decision?
- Who currently proposes, who recommends, who approves, who executes, and who must be consulted or informed?
- In reality (not on paper), where is the “veto power” and where is escalation typical?
- How long does it usually take to go from issue identification to decision?
- What issues are debated repeatedly in multiple forums?
Again, you do not need exhaustive matrices. Focus on a small number of high-impact decisions and be honest about how they happen today. You will often find:
- Decisions “owned” jointly by multiple functions with no clear primary owner.
- Decisions that sit at a level too high (top team is flooded with operational calls) or too low (front line is making calls that expose the firm to undue risk).
- Duplicative governance bodies all reviewing similar decisions.
- Informal shadow decision makers who are not in the official process but de facto must be convinced.
These insights will later guide your design of decision rights, governance, and management processes. For now, capture them as part of your diagnostic data.
6.3 Voice of the Organization: Interviews, Surveys, Workshops
The structural and process views need to be combined with how people actually experience the organization. This is where interviews, surveys, and workshops come in. The aim is not to collect random complaints, but to systematically understand themes about clarity, collaboration, leadership load, and engagement.
Interviews
Start with a structured interview program covering:
- Top team members and next-level leaders in scope.
- A sample of middle managers, front-line supervisors, and individual contributors.
- Representatives from key geographies, functions, and support units.
Interviews should follow a consistent guide, for example:
- What are the top 3–5 outcomes your team is accountable for?
- Where does the current structure help you deliver those outcomes?
- Where does it get in the way?
- Which decisions are too slow or too often escalated?
- Where do accountabilities feel unclear or overlapping?
- How workable is your span of control and your managers’ span of attention?
- How does collaboration with other units actually work?
Encourage interviewees to give concrete examples rather than broad generalizations. Capture their input as themes and illustrative quotes, not as a voting exercise on “favorite structures.”
Surveys
If you have recent organization-health or engagement surveys, mine them for insights relevant to design:
- Clarity of roles and goals.
- Perceptions of speed and agility.
- Collaboration across units.
- Leadership and managerial effectiveness.
- Workload and burnout indicators.
If no such survey exists or it is outdated, you may run a short, targeted pulse survey focused on organizational clarity, decision speed, and collaboration. Keep it short enough that people will respond, and use it to supplement, not replace, qualitative work.
Workshops and focus groups
Finally, use workshops and focus groups to deepen understanding and to validate early diagnostic themes:
- Cross-level workshops with managers to discuss how the structure supports or hinders delivery of strategy.
- Functional or regional workshops to understand local realities and variations.
- Mixed focus groups of individual contributors to hear from those closest to customers and operations.
These sessions should be well-facilitated. The aim is not open-ended venting; it is to test specific hypotheses and gather richer examples. For example: “We’ve heard that decision-making around pricing is slow and unclear. Does that resonate? Can you walk us through a recent example?”
Making the most of the voice-of-organization data
As you collect this input, look for:
- Themes that appear across levels and units, not just in one pocket.
- Points of tension between leadership perceptions and front-line experience.
- Consistency (or inconsistency) between what people say and what your structural and process data show.
Document the themes in plain language. Avoid phrases like “people say the org is bad.” Instead, write: “Many middle managers report spending more than half their time coordinating across functions to resolve issues that lack clear ownership.”
6.4 Performance and Health: Metrics, Benchmarks, Pain Points
Diagnosis is incomplete without a view of performance and organizational health. Structural and process issues matter only to the extent that they affect outcomes. This part of the work connects the “hard” metrics with the “soft” organizational experience.
Performance metrics
Start with a small set of metrics that reflect your value agenda:
- Financial: revenue growth by segment, margins, cost-to-serve, SG&A as a percentage of sales.
- Customer: NPS or equivalent satisfaction measures, retention and churn, complaint volume and resolution times.
- Operational: cycle times (e.g., time to market, order-to-cash, claims processing), error rates, rework levels, project overruns.
- People: regretted attrition, internal fill rates for critical roles, time-to-hire, absenteeism.
The goal is to see where the organization is performing well and where it is struggling, and then ask: “Could structure reasonably be part of why this is happening?”
For example:
- Persistent delays in product launches may link back to unclear ownership between product, marketing, and technology.
- High cost-to-serve in a particular segment may relate to duplicative structures across regions.
- High error rates in operations may reflect fragmented ownership or overstretched supervisors.
Benchmarks
Where feasible, use internal and external benchmarks:
- Compare similar units within your organization (e.g., different regions or business lines) to see where structures and performance diverge.
- Use external benchmarks cautiously to understand whether your spans and layers, support ratios, or governance intensities are broadly in line with peers.
Benchmarks are not targets by themselves. They help you identify outliers and test whether your intuition about “too heavy” or “too thin” is supported by data.
Organizational health
Combine the performance view with an explicit view on organizational health. If you have an organizational health survey, extract design-relevant dimensions:
- Direction and alignment.
- Leadership and management quality.
- Accountability and consequences.
- Coordination and control.
- Innovation and learning.
Overlay these results on your structural map. For example:
- Units with many layers and narrow spans may also show low empowerment and high micromanagement.
- Matrixed teams with dual reporting lines may show low clarity of accountability and high conflict.
- Functions dispersed across multiple business units may show inconsistent leadership quality and weak communities of practice.
This helps you distinguish between areas where structure is likely a significant driver of health issues and areas where the problem may be more behavioral or cultural.
Pain points
Finally, compile a short list of “pain points” that matter most. These are not generic complaints; they are specific symptoms that connect structure, processes, and outcomes, such as:
- “Pricing decisions for strategic accounts regularly take more than four weeks and require approvals from three different committees.”
- “Front-line managers in operations have 25 direct reports on average and spend little time coaching; error rates have risen and engagement is low.”
- “Ownership of the end-to-end customer onboarding process is split across three functions, leading to frequent hand-offs and inconsistent customer experience.”
This list becomes a practical anchor for later testing of design options: “Will this option materially improve these pain points?”
6.5 Diagnostic Summary: Strengths, Issues, and Design Hypotheses
The final step in diagnosis is to synthesize everything into a concise, usable output. A good diagnostic summary does three things:
- Acknowledges the strengths of the current organization.
- Clearly articulates the main structural and organizational issues.
- Proposes a small set of design hypotheses to test in the next phase.
Think of it as a 2–4 page narrative, not a 60-page slide deck.
1. Strengths
Begin by recognizing what works well. This is not just a politeness; it is essential for credibility and for preserving what should not be broken. Examples:
- “Strong accountability and P&L ownership at the business unit level has supported entrepreneurial growth.”
- “Functional communities in R&D and risk have deep expertise and high internal reputation.”
- “The current regional structure has enabled close relationships with local regulators and key customers.”
Be specific. Where possible, tie strengths to performance or health evidence.
2. Issues
Next, articulate the key organizational issues. Group them into a few themes rather than a long list. For each theme, use a simple pattern:
- What is happening.
- Where it is most visible.
- Why it matters (link to strategy, performance, or health).
For example:
- “Fragmented customer ownership. Strategic customers are served across multiple business units with no single accountable owner, leading to inconsistent pricing, overlapping contact points, and missed cross-sell opportunities.”
- “Excessive layering and coordination overhead. There are up to eight layers from the CEO to the front line in core operations, with many managers having three or fewer direct reports. Managers and staff report spending significant time in cross-unit coordination meetings, while cycle times have lengthened.”
- “Unclear decision rights in the matrix. Dual reporting lines between regions and global product leads are not supported by explicit decision-rights definitions. Commercial decisions for global key accounts are frequently escalated, and different units apply different discount policies.”
Limit yourself to three to six such issues. If you cannot prioritize, you are not yet ready to design.
3. Design hypotheses
Finally, translate the issues into design hypotheses—concise statements of potential structural directions to test. These are not yet decisions; they are propositions that will guide the design phase. For example:
- “We should move toward a segment-based go-to-market structure for strategic customers, with clear global account ownership, while retaining regional P&Ls for other segments.”
- “We should reduce management layers in core operations and increase average spans of control for mid-level managers, while investing in better frontline-manager support and training.”
- “We should clarify and codify decision rights between global product and regional sales for pricing, discounting, and product prioritization, possibly supported by a simplified governance structure.”
- “We should consider consolidating dispersed support activities (e.g., analytics, vendor management, PMOs) into shared service or center-of-excellence structures to reduce duplication and standardize practices.”
Each hypothesis should be:
- Directly linked to one or more issues and design objectives.
- Concrete enough to have clear implications for structure, roles, or governance.
- Open enough to allow genuine option development (you are not drawing the final org chart yet).
A simple diagnostic summary template
You can structure your summary along these lines:
- 1. Context and purpose (half a page).
Brief recap of strategy, triggers for redesign, and scope. - 2. Strengths of the current organization (half a page).
Three to five key strengths with supporting evidence. - 3. Key organizational issues (one to two pages).
Three to six themes with concise description and impact. - 4. Priority pain points (half a page).
Short list of specific symptoms that must improve. - 5. Design hypotheses (half to one page).
Four to eight hypotheses that will guide the design phase.
Review this summary with the top team. The goal is not to wordsmith; it is to ensure genuine alignment on the problems you are trying to solve and the directions worth exploring. Once that alignment is in place, you can move into the design phase with confidence that you are not building castles on a weak foundation.
A disciplined diagnosis takes time and effort, but it prevents far larger waste later. It ensures that your redesign is a targeted response to real, understood problems—not a generic reorganization driven by intuition, politics, or fashion.
