Designing the Enterprise Beyond Jobs and Headcount
- Aug 10
- 12 min read
Written by Rosie Hewat, Founder & CEO of Rosie’s People
Rosie Hewat is a Board and Executive Advisor and the Founder & CEO of Rosie’s People, a leadership and organisational advisory platform. She works with founders, boards, and senior leaders navigating complexity, scale, and high-stakes decision-making across global and regulated environments.
For decades, organisations have largely been designed around a familiar question, "How many people do we need, and what jobs should they do?" From that question come headcount budgets, organisational charts, job descriptions, reporting lines, salary bands and career ladders.
It is an operating model so embedded in how organisations work that we rarely stop to question whether the job itself is still the right unit around which to design an organisation.
Increasingly, I am not convinced it is. Research from Deloitte offers an interesting indication of how far working reality has already moved beyond traditional job architecture. It found that 63% of work was being performed in teams or projects outside people’s core job descriptions, while 81% of executives said work was increasingly crossing functional boundaries. More than a third said work was increasingly being performed by people outside the organisation who had no defined job within it at all.
Perhaps more significantly, organisations taking a skills-based approach were found to be 57% more likely to be agile and 107% more likely to place talent effectively.
This matters because the way work is being performed has changed considerably faster than many of the structures we use to organise it. People move across projects. Specialist expertise may be required intensely for three months and barely at all for the following six. Organisations increasingly combine permanent employees with contractors, consultants, fractional leaders, external partners and technology.
Yet we continue trying to squeeze this increasingly fluid reality into boxes designed around fixed jobs. Perhaps the better question is no longer simply, “Who do we need to employ?” Perhaps it is, “What capabilities does this organisation need to succeed, and which of those capabilities must it own, develop, retain or access?” That distinction sounds subtle. However, I think it fundamentally changes how we design organisations.
From changing work to redesigning the organisation
This question did not appear in isolation for me. In my earlier article, “Why Many AI Productivity Tools Fall Short of Real Automation, and How to Use AI Responsibly”, I argued that adding artificial intelligence (AI) to an inefficient process does not automatically transform the process. Sometimes we simply perform an inefficient activity faster. That principle extends beyond technology.
Removing management layers does not automatically create agility. Outsourcing a function does not automatically create efficiency. Reducing permanent headcount does not automatically create a flexible workforce. Redesigning a job description does not necessarily redesign the work.
More recently, in “What Happens When AI Removes the Work That Teaches Us How to Work?”, I explored what I called the Apprenticing Gap, the risk that as organisations remove many of the activities through which people traditionally learned their professions, they may inadvertently weaken the pipeline through which future expertise is developed.
That article ended with an important question. If the work changes, what replaces the learning? I think there is now a bigger question to ask:
"If the nature of work, careers and organisational boundaries are all changing, what should replace the organisation designed around them?"
That takes us beyond technology. It takes us into organisational architecture.
A job is not the same as a capability
A job is a collection of responsibilities assigned to an individual. A capability is something the organisation must be able to do well. Those are not the same thing.
An organisation may need the capability to understand its customers, manage regulatory risk, develop products, recruit talent, protect its reputation, manage cash, negotiate complex partnerships or make sound strategic decisions.
Historically, our default response has been to convert those requirements into jobs. We need finance capability, so we hire finance people. We need human resources (HR) capability, so we build an HR department. We need technology capability, so we build an engineering team. There is nothing inherently wrong with that. In many circumstances, it remains exactly the right answer. But capability can now exist in several places.
It can sit with permanent employees. It can be shared across teams. It can be brought in temporarily through fractional or interim expertise. It can be supplied by specialist partners. Elements of it can be performed through technology.
This means workforce planning should perhaps begin one level above the job description. Instead of asking, “Which positions should we fill?” leaders should first ask, “What must this organisation always be capable of doing?”
Only then should we decide where that capability should live. This is where I believe some organisations are at risk of making a serious mistake.
Flexibility can become hollowed out
There is an understandable attraction to flexible organisational models. Why permanently employ every capability you might occasionally need? Why maintain layers of management if teams can make more decisions themselves? Why retain activities internally when technology or external providers can perform them faster or more cheaply?
These are legitimate questions. But there is another question leaders need to ask alongside them. What happens to our organisation’s ability to think when too much of what it knows no longer lives inside the organisation?
I have seen versions of this throughout my career. I remember one organisation where an employee left after what was supposedly a comprehensive handover. There were notes. Processes had been documented. Everyone assumed the knowledge had therefore been transferred. A few weeks later, the organisation was panicking. The person had been responsible for an important platform and, despite having the instructions, nobody really understood how to operate it properly. Interestingly, before leaving, the employee had warned that the role was not as straightforward as people assumed.
They were right. What disappeared when they left was not the instruction manual. It was context. They knew what happened when the normal process did not work. They knew which shortcuts were dangerous. They understood why something had been configured in a particular way. They recognised the strange error they had encountered six months earlier. They knew who to speak to when something unusual happened.
You can document a process. It is much harder to document everything somebody has learned from performing it. The objective is not to minimise permanent headcount. It is to become much more intentional about why capability sits where it does.
When knowledge leaves with the person
This is not a new organisational problem. More than three decades ago, academics Wesley Cohen and Daniel Levinthal introduced the concept of absorptive capacity, an organisation’s ability to recognise, assimilate and apply external knowledge depends partly on the knowledge it already possesses internally.
That principle becomes particularly important as organisational boundaries become more fluid. Outsourcing a capability may be commercially sensible. Using fractional expertise may be exactly what a growing organisation needs. Technology may perform parts of a process more efficiently than people. But the organisation still needs enough internal understanding to evaluate what it is receiving.
If nobody internally understands the work well enough to challenge an external provider, recognise an incorrect output, or understand why a system was designed in a particular way, the organisation has not simply become flexible. It has become dependent.
This is the difference between accessing expertise and retaining capability. That distinction matters because a leaner organisation is not necessarily a more knowledgeable one.
Not every critical person sits near the top
Capability-centred thinking also forces us to reconsider how we assess organisational value. I once worked with an organisation that needed to reduce costs. A relatively junior employee was identified as someone the business could afford to lose. I remember disagreeing.
My view was essentially, I don’t think you can afford for this person to leave. The decision was made anyway. Only after the individual had gone did it become apparent how much of the organisation they had quietly been holding together. Not because they occupied an impressive position on the organisational chart.
They simply knew things. People trusted them. They connected information between teams. They knew how work really happened rather than how the process map said it happened. They solved small problems before they became large ones. Their organisational value was considerably greater than their job title suggested.
I have seen the reverse too. Sometimes someone leaves and productivity improves. Team morale changes for the better almost immediately. Decisions become easier. People communicate more freely.
Both situations reveal the same problem. Headcount tells you how many people you employ. It tells you surprisingly little about the capability those people collectively create.
An organisation chart can show me who reports to whom. It cannot necessarily show me where trust lives, where institutional knowledge sits, who connects otherwise disconnected teams, where critical judgment resides, or whose departure would quietly break an important process. Yet those are precisely the things leaders need to understand before redesigning an organisation.
The career ladder is beginning to look different too
Capability-centred thinking also challenges the traditional career ladder. Expertise and people management are different capabilities, yet many organisations still require outstanding specialists to become managers if they want greater status, influence or compensation. That deserves a much deeper conversation of its own. For now, the organisational design question is whether our structures allow critical expertise to deepen without forcing it into an entirely different profession.
This is one of the areas where capability-centred thinking begins to challenge traditional career architecture. If expertise is strategically valuable, organisations need credible ways for it to deepen and progress without assuming that greater organisational value must always mean greater people management responsibility.
That is a question I will return to separately, because how people progress inside a capability-centred organisation may matter just as much as how the organisation itself is structured.
There is a fascinating tension here. Organisations want experience. But experience has to be created somewhere. That takes us directly back to the question I explored in What Happens When AI Removes the Work That Teaches Us How to Work? If organisations narrow the first rung of the ladder while simultaneously increasing their demand for experienced talent, eventually the maths stops working. Capability cannot simply be purchased indefinitely. Somebody has to build it.
Flatter does not automatically mean better
One response to changing work has been to remove management layers. There can be very good reasons for doing so. Excessive hierarchy slows decisions, creates bureaucracy and can move authority too far away from the people closest to the work. But deleting boxes from an organisational chart is not organisational design.
One of the more interesting real-world examples is Bayer’s Dynamic Shared Ownership operating model. At its April 2026 Annual Stockholders’ Meeting, Bayer chief executive officer (CEO) Bill Anderson said the company had roughly halved the number of organisational layers and reduced management positions by two-thirds. The company now works through focused 90-day cycles and is attempting to move its organisational emphasis from “administering to doing.” What interests me about this is not simply the reduction in managers. It is the attempted redesign of how the organisation operates after those layers disappear.
That distinction matters. Removing managers while leaving the same decision rights, workflows, meetings, responsibilities and approval structures behind simply distributes the bureaucracy among fewer people. The work does not disappear because the box on the organisation chart does. Management itself therefore needs to evolve.
Routine coordination, reporting and administration may become easier to distribute or automate. But setting direction, resolving ambiguity, coaching people, protecting culture, connecting capabilities, managing risk and exercising commercial judgment do not suddenly cease to matter.
The administrative burden of management may shrink. The human and organisational responsibility of management may actually become greater.
Government can change the route into work, and employers must change what people enter into
There is an interesting development happening in the UK as I write this. In July 2026, Prime Minister Andy Burnham announced reforms intended to expand technical education, work experience and employer connections from age 14, with greater parity between academic and technical pathways.
This connects directly to a question I explored in my earlier article, Preparing the Next Generation for Careers in the AI Era and Training for the Right Future- what exactly are we preparing people for when the nature of work itself is changing?
The new policy direction potentially addresses part of that challenge. But only part. Government can redesign the route into work. Employers must redesign what people enter into.
The World Economic Forum’s Future of Jobs Report 2025 found that employers expect 39% of workers’ existing skill sets to be transformed or become outdated by 2030, while 63% already identify skills gaps as a major barrier to transformation.
Yet 81% of employers still expect work experience to remain their most common way of assessing candidates. That exposes an important contradiction. We want new skills, new pathways and a more adaptable workforce, while continuing to rely heavily on previous experience as proof that somebody can perform.
If employers want different capability entering the workforce, they cannot leave their own job architecture, hiring assumptions and progression models untouched.
From headcount planning to capability architecture
Too much workforce restructuring begins with subtraction. We need to remove 10% of cost. Which roles can go? Which layer can disappear? Which activity can be outsourced? Which vacancy can remain unfilled? Etc.
So perhaps the workforce planning conversation needs to change. Instead of beginning with, “How many employees do we need?” Boards and executive teams might begin with, “What must this organisation always know how to do?”
Then ask:
Where does that capability currently live?
Who holds the knowledge?
Who else is learning it?
What happens if those people leave?
Which capabilities are critical to our competitive advantage?
Which capabilities could safely be bought, borrowed or flexed?
Which capabilities have we outsourced so extensively that we can no longer properly evaluate them?
Where are we confusing management seniority with expertise?
Where are we carrying jobs that no longer reflect how the work actually happens?
Where are we saving money today by quietly creating a capability deficit for tomorrow?
These are different questions from traditional workforce planning. They require leaders to look beyond the organisation chart and understand the organisation as a living system of knowledge, judgment, relationships, expertise and execution.
That, to me, is what capability-centered organisational design really means. It is not the end of jobs. It is not the end of permanent employment. It is not the end of management. It certainly is not an argument for replacing people with technology. It is an argument for recognising that none of those things, individually, represents organisational capability.
Conclusion: The question boards should be asking
There is a temptation whenever a new workforce model emerges to turn it into another trend. Flatten the organisation. Become skills first. Hire fractionally. Automate. Build project teams. But organisational design should not be a collection of fashionable interventions. It should answer a much more fundamental question.
What capabilities must this organisation preserve, develop and continuously renew in order to remain effective when the people, technology and environment around it change?
Answering that requires leaders to look beyond how many people they employ. They need to understand where knowledge lives, how expertise develops, which relationships hold the organisation together and which capabilities would be difficult to recreate if they disappeared.
It also requires resisting the assumption that less automatically means better. A smaller workforce is not necessarily a more capable one. A flatter organisation is not automatically more agile. Accessing external expertise does not remove the need for internal understanding.
As I explored in The Apprenticing Gap, organisations cannot think only about the capabilities they need today. Someone still has to become the expert of tomorrow.
The organisations best prepared for the next decade may therefore not be those that predict exactly how work will change. Few of us can. They may be those that understand, with unusual clarity, what they must always know how to do, and have deliberately designed themselves so that they never forget.
Perhaps that is the real shift from headcount planning to capability architecture. The question is no longer simply, “How many people do we need?” It is, “What must our organisation always be capable of doing, where should that capability live, and are we designing the organisation so that it survives change?”
Read more from Rosie Hewat
Rosie Hewat, Founder & CEO of Rosie’s People
Rosie Hewat is a Board and Executive Advisor and Founder & CEO of Rosie’s People, a leadership and organisational advisory platform. A former Group Chief People Officer and Non-Executive Director, she has supported leadership teams and boards operating in high-growth and regulated environments. Rosie is also a trustee and an Executive Contributor to Brainz Magazine, where she writes on leadership, governance, power, and organisational risk.
References and further reading:
Deloitte: Skills-based organizations - research into work beyond job descriptions and skills-based organisational models.
World Economic Forum: Future of Jobs Report 2025 - employer expectations around changing skills, workforce strategy and transformation.
Indeed Hiring Lab: The Labor Market Is Tilting Toward Seniority - July 2026 evidence on the divergence between entry-level and senior hiring.
Bayer: 2026 Annual Stockholders' Meeting address - current evidence on Bayer’s organisational delayering and Dynamic Shared Ownership.
UK Government: Technical education and employment pathway reforms - the July 2026 Burnham Government announcement.
Cohen, W. M. & Levinthal, D. A. (1990), Absorptive Capacity: A New Perspective on Learning and Innovation, Administrative Science Quarterly, 35(1), 128–152.
Why Many AI Productivity Tools Fall Short of Real Automation, and How to Use AI Responsibly - earlier article in this series.
Preparing the Next Generation for Careers in the AI Era and Training for the Right Future - earlier article examining education and preparation for changing work.
The AI Workforce Shift and Who Benefits, Who Pays, and What Happens Next - the wider workforce and economic implications explored earlier in the series.
What Happens When AI Removes the Work That Teaches Us How to Work? - the previous article and the foundation for the “Apprenticing Gap” argument.
