# Benchmarking organizational structures: which comparisons hold up

Author: Marin Niehues, Niehues Consulting · Published: 2025-12-11 · Updated: 2026-11-16 · Organizational Design
Source: https://www.niehuesconsulting.com/blog/benchmarking-organisational-structures/

Benchmarking organizational structures compares metrics such as span of control and number of layers with reference values in order to justify a structural decision. Benchmarks hold up when they come from comparable value creation, share a uniform definition and are tested against delivery results.

A common mistake is comparing against industry-wide averages and deriving target values for spans of control and layers directly from them. The result looks objective and remains unusable for the decision, because the values mix different tasks and different levels of maturity. This article describes a test framework for benchmarks, a worked example and an approach that links the comparison to delivery results.

**A structural comparison is robust when peer group and definition are settled and the value is tested against delivery results. Internal comparison provides the most robust statement about impact, external comparison provides the corridor.**

## Why structural comparisons often mislead

An organizational structure is the answer to a specific task. The sociologist Joan Woodward showed this in 1965 in a study of around 100 manufacturing firms in the English county of Essex: the span of control at the first management level was widest in mass production and markedly narrower in unit production as well as in process production, while the number of layers increased with technical complexity. Averaging metrics across such differences produces a value that describes no real company.

Differing definitions add to this. Does a team lead with specialist tasks count as a manager? Do external staff belong to the span? How are matrix relationships recorded? Even small differences in counting change the picture markedly. The two core metrics are covered in detail in [Spans and layers](/blog/spans-and-layers-management-levels/) and [Span of control: how many direct reports a manager can lead effectively](/blog/span-of-control-direct-reports-per-manager/).

A benchmark also shows only how others are organized. Whether these organizations deliver well remains open. A value describing the average may therefore reflect a structure that, on average, delivers poorly.

## What types of benchmarking are there?

Benchmarking distinguishes four types: internal comparison, comparison with competitors, and functional as well as generic benchmarking. The classification goes back to Robert Camp, who helped develop the method at Xerox and described it in a standard work in 1989.

- Internal benchmarking compares divisions, sites or subsidiaries of the same organization. Definitions and systems are identical, and the data are fully accessible.
- Competitive benchmarking compares with direct competitors. Its relevance for one's own market is high, but access to data is narrow.
- Functional benchmarking compares a function, such as procurement or customer service, with companies in other industries that perform this function in a similar way.
- Generic benchmarking compares fundamental processes such as order handling or invoice checking across industries, independent of the function.

For structural questions, a clear sequence applies. Internal comparison comes first because it shows impact. Functional benchmarking follows for cross-functional areas such as finance or HR, whose tasks are similar across industries. Comparisons with competitors serve as a plausibility check.

## Four test criteria for benchmarks

Before using any benchmark, the organization answers four questions. They can be worked through in a few minutes per metric and separate usable values from decorative ones.

**Test framework for benchmarks**
- Peer group: Do the comparison companies come from similar value creation, with similar size and similar maturity?
- Definition: Are the counting method and scope of the metric documented and identical to one's own?
- Decision relevance: Which concrete decision changes if the value is higher or lower?
- Evidence of impact: Is there a demonstrated link between the metric and delivery results?

If a value fails one of the questions, it remains a pointer for analysis and does not become a target. This also applies to values from renowned studies. What counts is applicability to one's own decision.

## Which metrics can be meaningfully compared?

Three groups of metrics can be meaningfully compared, provided every structural metric is linked to a performance metric:

- Structural metrics such as span of control and number of layers as well as the share of management and support functions in the workforce: externally comparable only with a clean definition and peer group, internally very well across divisions and over time
- Performance metrics such as lead time and schedule reliability: they show the impact of the structure and can be compared between divisions and over time
- Cost metrics such as personnel costs per unit of output: meaningful only together with a performance metric

Internal comparison provides the most robust statement about impact, and external comparison with a properly formed peer group provides the corridor within which one's own values should lie. Divisions of the same organization share definitions and systems, so differences between comparable divisions show which structural variants deliver better. External values provide hypotheses that the organization tests against its own data. The article [Reducing overhead without loss of performance](/blog/reduce-overhead-without-losing-performance/) describes how to reduce the share of management functions without weakening delivery.

## Where do benchmarks come from?

Benchmarks come from three sources. The first are the benchmark databases of strategy and HR consultancies, the second are association and industry surveys, the third is the company's own personnel data. The databases of consultancies and HR service providers contain most values on spans and overhead shares and usually become accessible within an engagement or through participation in the survey. Association surveys cover individual industries or functions. The company's own personnel data are the only source that guarantees full comparability of definitions.

## Example: what does a structural benchmark look like in practice?

A structural benchmark links a structural metric with a delivery metric across comparable units. The following example shows the process for an internal comparison using invented but typical values.

**Worked example: customer service at four sites**
A company runs customer service at four sites with the same task and the same systems. Only line managers are counted; team leads who spend more than half their time on specialist work count as staff.

Site A: 120 employees, 10 managers, span 12, three layers, first-contact resolution 78 percent.

Site B: 110 employees, 16 managers, span just under 7, four layers, first-contact resolution 71 percent.

Site C: 90 employees, 7 managers, span just under 13, three layers, first-contact resolution 76 percent.

Site D: 100 employees, 5 managers, span 20, two layers, first-contact resolution 69 percent.

Finding: the best delivery values are at the sites with a span of 12 to 13 and three layers. Site B has an additional layer without better delivery, site D a span at which leadership for onboarding and quality is lacking. The target corridor for this task is a span of about 10 to 14.

Question for the executive team: which decision follows for B and D, and against which delivery metric will it be reviewed after six months?

**Exhibit 1 (Schematic): In the example, the sites with a span of 12 to 13 and three layers deliver best**

| Label | Value |
| --- | --- |
| Site A, span 12, three layers | 78% |
| Site C, span just under 13, three layers | 76% |
| Site B, span just under 7, four layers | 71% |
| Site D, span 20, two layers | 69% |

Schematic illustration, not measured data.
Source: Worked example in the article
Note: Worked example from the article, first-contact resolution by site
Graphic: [SVG](https://www.niehuesconsulting.com/blog/benchmarking-organisational-structures/grafik-1.svg) · [CSV](https://www.niehuesconsulting.com/blog/benchmarking-organisational-structures/grafik-1.csv) · [JSON](https://www.niehuesconsulting.com/blog/benchmarking-organisational-structures/grafik-1.json) · Niehues Consulting, [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/deed.en)

The example shows the core of the method. The corridor emerges from the organization's own delivery and applies to exactly this task. An external value would at best have confirmed that 12 to 13 lies within the usual range.

## From comparison to decision: a four-step approach

In the first step, the question is defined. A decision such as merging divisions or changing the number of layers determines which metrics are needed at all. A benchmark without a decision produces reports and no change.

In the second step, the peer group is formed. Internally these are units with similar tasks, externally selected references with similar value creation. In the third step, the metrics are normalized: a uniform counting method, adjusted for the respective task mix. Only then is a comparison permissible.

In the fourth step, impact is tested. The structural metric is linked to delivery metrics such as lead time and schedule reliability. If divisions with a narrower span show better delivery performance, that supports a target value. If no link appears, the target is dropped. The result is a target corridor per division, justified by task and delivery results. The article [Organizational diagnosis: how an assessment works](/blog/organisational-diagnostic-how-an-assessment-works/) describes the underlying assessment, and the article [characteristics and metrics of a high-performing organization](/blog/high-performing-organisation/) sets out the overall framework.

## Reading and applying benchmarks correctly

A benchmark describes a range. The dispersion within the peer group matters more for interpretation than the mean, because it shows how much design latitude exists for the task in question. If values lie close together, this points to a stable logic of the task. If they vary widely, the quality of execution is decisive, and comparison with the best units of one's own organization yields more than any external average.

A deviation from the reference value is initially a question and not yet a finding. A wide span may point to lean management or to overloaded managers who have little time for their teams. Delivery metrics and team feedback show which interpretation applies.

The value is also observed over time. A single reference date captures one-off effects such as vacancies or a recently completed reorganisation. A trend over several quarters separates lasting patterns from snapshots. Finally, every consequence is formulated as a hypothesis and trialled in a limited area before being rolled out widely.

## How Niehues Consulting supports organizations with benchmarking

[Niehues Consulting](/) links structural comparison with the delivery perspective. Marin Niehues works as a management consultant and builds organizations that deliver, in projects ranging from VC-funded scale-ups to DAX 40 corporations, with assignments in 20+ countries across six continents.

The approach follows the Niehues Delivery Model with the phases Understand, Build and Implement. For a structural comparison, two review questions matter most: What is the baseline, and how do we measure success? Which roles, decision paths and interfaces does operation require? In Understand, the question and data basis are established: peer groups with uniform definitions, linked to lead time and schedule reliability. This produces a decision paper with target corridors per division. In Build and Implement, Niehues Consulting translates the decision into roles and decision paths and steers implementation as delivery manager or product owner until the new structure delivers reliably.

**Benchmarks at a glance**
Criterion | External industry comparison | Internal comparison | With Niehues Consulting
Comparability | Low for industry averages, medium for a properly formed peer group | High with identical definitions | Peer group and definitions set before analysis
Statement on impact | Low | Medium to high | Linked to lead time and schedule reliability
Value for the decision | Orientation within a corridor | Concrete pointers per division | Target corridors with rationale
Implementation | Open | Open | Roles and decision paths, steered through to results

**Key points for decision-makers**
A benchmark is an aid to decision-making. It works when peer group and definition are clarified and structure is measured against delivery results. The first step is an internal comparison of similar divisions with a uniform counting method. The [free initial consultation](/#kontakt) offers a conversation on this.

**Sources**
- Woodward: Industrial Organization: Theory and Practice, Oxford University Press, 1965
- Camp: Benchmarking: The Search for Industry Best Practices that Lead to Superior Performance, ASQC Quality Press, 1989

## Frequently asked questions

### What is organizational structure benchmarking?

Organizational structure benchmarking compares metrics such as span of control and number of layers with reference values, internal or external. The aim is a well-founded decision on structure and management.

### What are the types of benchmarking?

There are four types of benchmarking: internal comparison, comparison with competitors, and functional as well as generic benchmarking. For structural questions, internal comparison provides the most robust statement, while functional benchmarking suits cross-functional areas such as finance or HR.

### Which benchmarks are meaningful for organizational structures?

Meaningful values come from a comparable peer group with a documented definition. In addition, every value needs a clear link to a decision and demonstrated impact on delivery results. Internal comparisons between similar divisions meet these criteria most often.

### What does a structural benchmark look like in practice?

A structural benchmark places span and layers of comparable units next to a delivery metric such as lead time or resolution rate. The units with the best delivery values define the target corridor. Deviating units receive a decision with a review date.

### Where do benchmarks for span of control and layers come from?

Sources are the benchmark databases of strategy and HR consultancies as well as association surveys, plus the company's own personnel data. External values are measured against the four test criteria before use. Only one's own data guarantee full comparability of definitions.

### What are the advantages and disadvantages of benchmarking?

The advantage of benchmarking is an objective reference point that makes deviations visible and provides hypotheses for design. The disadvantage lies in confusing average and target: a benchmark shows how others are organized and leaves open how well they deliver. Linking it to delivery metrics resolves this problem.

### How often should an organization benchmark its structure?

An organization benchmarks its structure before every major structural decision. A few core metrics also run in management reporting, and the target corridors are reviewed annually.
