In many professional practices and small firms, the people who bring in the work are also the people who keep the books straight, chase the invoices, reconcile the accounts and answer the supplier who has not been paid. Nobody planned it that way. The firm began small, the partner was the only person who knew how everything worked, and each new task simply landed on the nearest capable pair of hands.
This piece looks at a recognisable situation: a partner or owner with real potential to grow the firm, whose time is quietly consumed by administration that someone else could carry. It is a hypothesis, built from general research on workload and from how small businesses tend to operate. It is not a study of any particular firm, and the example below is illustrative.
When the best use of a person's time is not the one they are spending
For some owners, the administrative load is not dramatic. It is a steady, reasonable-looking set of tasks: authorising payments, checking timesheets, approving quotes, preparing the month-end pack. Each one takes a few minutes. Together they can take most of a week. The owner may not even experience it as a problem, since the work is familiar and the business needs it done.
The cost shows up as things that do not happen. The conversation with a prospective client does not take place. The new service line stays at the idea stage. The partner who might have mentored a junior colleague or spent a day thinking about direction spends that day on reconciliations. The firm is stable, but it is not growing in the way its principal could in theory make it grow.
The UK Health and Safety Executive's standard on demands describes the aim that employees have workloads manageable within agreed working hours and jobs that match their skills and abilities. It is written about employee stress, but the underlying idea applies to owners too: a job design where the most valuable skills are used least is a design problem, not a personal failing.
An illustrative example
Take a hypothetical South African accounting and advisory practice with two partners and eight staff. One partner is the rainmaker: clients trust her, and she is good at spotting what a business needs next. She is also the person who approves every supplier payment because the bookkeeper once made an error and trust was lost. She reviews every engagement letter, because the template was never standardised. She handles the practice's own invoicing because "it only takes an hour".
Over a year, those hours accumulate. She can name three advisory opportunities that she did not follow up. She cannot take a proper week away, because nobody else holds the payment approvals. Her partner is a gifted technician who prefers not to take on management. The firm is profitable and respected, and the founder's agency over her own growth is smaller than anyone looking from outside would guess.
This is not a report of a real practice. It illustrates a mechanism: administrative tasks that stay with the person who has the most to lose by doing them.
Why admin tends to stay where it is
Several ordinary reasons keep administrative work with senior people.
- Trust was built on a past mistake. One error leads to a rule that the owner approves everything, and the rule outlives the mistake.
- The process lives in one head. If the owner is the only one who knows how a task is done, delegating means explaining, and explaining feels slower than doing.
- The systems do not support delegation. If approvals, access or reporting only work through one login, the owner holds the keys because the tools demand it. Dependency on one person is discussed in key-person dependency is a systems problem.
- Identity. For some founders, being across every detail is part of how they understand their role. This is explored in when the founder's identity is fused with every task.
The self-determination research summarised at selfdeterminationtheory.org describes autonomy, competence and relatedness as basic psychological needs, and says that contexts supporting them tend to foster better engagement and performance. It does not say that every owner wants to hand off administration, and it would be a mistake to assume so. Some owners find a sense of competence and control in the detail. The argument here is conditional: for owners who want to grow and are blocked by admin, handing it off can restore choice.
What letting go requires from the business
Delegation rarely fails because someone refuses to let go. It fails because the business has not made letting go safe. A few enabling conditions are usually involved.
- Clear decision rules. If a bookkeeper may approve payments under a stated amount, with a record of who approved what, the owner can stop reviewing small items without losing sight of the large ones.
- Documented routines. The steps for month-end, invoicing and engagement letters exist somewhere other than the owner's memory. Guidance on this is in what business processes need documentation first.
- Shared visibility. The owner can still see the numbers that matter, so that giving up the task does not mean giving up awareness.
- A realistic recipient. The person taking the work has the time, training and authority to do it. Otherwise the task simply moves and returns.
The point of what real delegation requires from systems is closely related. The distinction in this piece is the specific consequence for the owner's growth options rather than the mechanics of delegation itself.
The trade-offs of letting go
There is no guarantee that freeing the partner's time leads to growth. Several things can go the other way.
- Money. Taking on administrative help costs salary or fees. In a small firm, that cost is real, and the growth it is meant to enable is uncertain.
- Control and quality. Delegated work may be done differently, and occasionally worse, at first. Some owners will feel the loss of oversight keenly.
- Displaced work. The admin may not vanish. It may land on a junior colleague who was hired for something else, creating the same problem lower down. The HSE demands standard applies to them as well.
- Different choices. A freed-up owner may use the time to grow, to rest, to study or to work fewer hours. All are legitimate. Nobody should read the dividend as an instruction to expand.
It is also possible that, in a given firm, the administrative load is smaller than it feels, or that the owner is the right person to keep a particular control. A careful look might conclude that nothing should change. Even then, the exercise of listing tasks and asking who could carry them tends to show how much of the firm's continuity rests on one person, which is worth knowing whether or not anything is delegated. The cost of an unexpected absence, for example illness, is explored in the owner facing illness with no one who knows the systems.
Interpreting the pattern
The Mellorca reading is that, for some owners, the limit on growth is less about ambition or market than about how much of their working week is claimed by tasks the business could be arranged to carry without them. The honest test is simple: list the recurring tasks you do each month, and for each ask whether it needs your judgement or only your login. The tasks that need only your login are the ones the business can learn to run differently, and the value of doing so is the set of options it gives back to you.
Sources and further reading
- Management Standards - Demands - HSE (last verified 10 October 2026)
- Theory – selfdeterminationtheory.org (last verified 10 October 2026)
- Mental health at work (last verified 10 October 2026)
- Management Standards - Control - HSE (last verified 10 October 2026)
Related Mellorca reading
- Why Standardisation Can Free or Confine
- Why Good Judgement Gets Buried Under Process
- The Quiet Power of Choosing When to Respond