There is a particular kind of tiredness that belongs to the founder who has been saying "I really should step back" for three years. The business is doing fine. Staff are capable. Customers are loyal. And yet the founder is still the person who knows which supplier needs a phone call rather than an email, where the VAT working papers live, and what the password to the bank portal is. Stepping back keeps being postponed, not because anyone forbids it, but because every attempt exposes another thing that only one person knows.
This piece looks at why that happens, why it is usually a design problem rather than a character flaw, and what it takes to make stepping back a real option rather than a recurring intention. It is a hypothesis about a common pattern, not a diagnosis of any particular owner.
Wanting to step back is not the same as wanting to leave
It helps to be precise about what the founder wants. Stepping back rarely means walking away. For some owners it means a full week of leave without a laptop. For others it means working on the business rather than in it, choosing the two or three decisions that genuinely need them and letting the rest happen without them. For others still it is a slower, longer runway to retirement or to a different project.
All of these depend on the business carrying on, correctly, when the founder is not the one holding it together. That is a statement about the business, not about the founder's willingness.
How a business ends up built around one person
Nobody sets out to do this. It happens through a series of sensible shortcuts in the early years.
- Speed beat documentation. When there were three people, the fastest way to do something was to ask the founder. Writing it down felt like a luxury.
- Exceptions accumulated. The founder agreed special terms with a long-standing customer, a different payment day with a supplier, a workaround for a client's awkward system. Each was reasonable. None was recorded anywhere except in the founder's head.
- Access followed the founder. The company bank profile, the SARS eFiling login, the domain registration and the accounting software were all set up by the founder, often on a personal email address, and never handed on.
Mellorca's work on key-person dependency as a systems problem makes the same point from the operational side: the dependency is rarely about talent, and nearly always about where knowledge, access and authority happen to be stored.
What the evidence can and cannot say
There is no South African survey that measures how many owners feel unable to step back, and it would be a mistake to invent one. What can be said is narrower, and drawn from general workplace research.
Microsoft's Work Trend Index found that 68% of survey respondents said they did not have enough uninterrupted focus time, and that in its Microsoft 365 data people spent 57% of their time communicating and 43% creating. If that is true for staff in larger organisations, it is plausible that an owner who is the default answer to every question experiences it more sharply. That is an inference, not a finding.
Gallup's work on burnout reports that only 60% of workers in the American workplace study strongly agreed that they knew what was expected of them, and treats unclear roles as one of the main drivers of exhaustion. When no one besides the founder is sure who decides what, the founder is left to decide everything.
Google's engineering handbook on eliminating toil offers a useful idea from another field: manual, repetitive work grows with a system unless someone deliberately removes it. The same logic applies to a business that needs its founder to touch every routine decision.
An illustrative case
Consider an illustrative example, not a real case: a Durban-based company that distributes industrial cleaning supplies to hotels and factories. It has fourteen staff, an office manager, two sales representatives and a warehouse supervisor. The founder started it fifteen years ago and now wants to take three weeks off in December and, over the next two years, spend more of the working year at the coast with family.
Every year the plan stalls in the same places. Credit limits for the larger customers are not written down; the founder remembers which accounts can run to sixty days. Price exceptions live in the founder's phone messages. The bank requires a second approver for large supplier payments, but the second approver is also the founder's spouse, who is on the same trip. The warehouse supervisor will not release an emergency order to a customer without hearing it from the founder personally.
Nobody here is behaving badly: staff are being careful with decisions they were never equipped to make, and the founder is being responsible with their own money. The result is a business that politely refuses to let its owner go.
The human tax: autonomy and optionality
The first cost is autonomy. A founder who cannot be unreachable is not fully free to choose how to spend a week, a season or a decade. The business has become the one thing they cannot decline to attend to.
The second is optionality. A business that depends entirely on one person cannot be easily sold, handed to a successor, financed against future earnings or even run through an illness. The options that most owners assume they have, such as selling in ten years or promoting a deputy, quietly close. The article on the owner facing illness with no one who knows the systems describes the most severe version of this.
The cost also falls on others: a manager who cannot act without the founder is held back too, and a founder whose identity is fused with every task may find that stepping back involves more than logistics. This piece stays with the practical side.
What could realistically change
The fix is not a handover day. It is a sequence of small moves that make the founder's knowledge available to others, and the founder's authority specific rather than total.
- List what only the founder can do. For one month, note every time someone asks the founder a question or the founder steps in. Group the results into decisions, access, knowledge and relationships.
- Move access to the business. Banking profiles, tax logins, domain names and software accounts should belong to the company, with named backups, rather than to a personal email address. SARS's guidance for small businesses is a reminder of how many registrations, from PAYE to VAT, a business accumulates, and each one has a login somewhere.
- Write down the rules behind the exceptions. Credit limits, discount bands, payment terms and escalation thresholds are decisions with reasons. Once they are stated as rules, a trusted person can apply them.
- Give decisions an explicit ceiling. "You can approve up to R15 000 without asking" is far more usable than "use your judgement". It lets staff act, and it tells the founder exactly what they are still responsible for.
- Rehearse the absence. Take a long weekend with the phone off, on purpose, and note what broke. Treat each break as a task to be fixed, not as evidence that the plan cannot work.
The broader principle is covered in what real delegation requires from systems: delegation fails when the person delegated to lacks the information, the access or the permission to succeed, and it succeeds when the system supplies them. A handoff that cannot depend on memory is the practical method.
The business outcome
Questions stop travelling to one person. The business becomes more resilient to illness, leave and growth, and more valuable to anyone who might one day buy it, lend to it or inherit it.
The human dividend
The dividend is optionality and autonomy: the ability to choose, and to have that choice honoured by the business. Some founders will use it for a month at the coast, others to start something new or to work fewer evenings. The point is to make the choice possible, not to prescribe it.
The trade-offs
- Delegated decisions will sometimes be made differently. A founder who steps back accepts that some choices will not be the ones they would have made.
- Control and visibility fall. Founders often lose their informal view of the business and need a lightweight way, such as a weekly summary, to stay informed without staying involved.
- Not every owner wants to step back. The aim is the freedom to choose, not a mandate to retire.
Why it matters
A business that only works while one person holds it together is not independent of that person, however capable the team. Stepping back becomes possible when knowledge, access and authority are spread deliberately rather than by accident.
Sources and further reading
- https://www.microsoft.com/en-us/worklab/work-trend-index/will-ai-fix-work (last verified 10 October 2026)
- Employee Burnout, Part 1: The 5 Main Causes (last verified 10 October 2026)
- Google SRE - What is Toil in SRE: Understanding Its Impact (last verified 10 October 2026)
- Small Businesses - Taxpayers - South African Revenue Service (last verified 10 October 2026)
Related Mellorca reading
- When the System Decides What You Can Do Today
- The Owner Who Can't Decide Without the Spreadsheet Wizard
- The Difference Between Being Informed and Being in Control