AutoHive Business Consulting applies two screens and one scope boundary before accepting an engagement. The first screen is a size filter: the practice works with South African SMEs turning over between R20 million and R250 million a year. Below roughly R20 million a competent accountant, a signed shareholders' agreement and a clear head deliver more than a structure does, and paying for governance a business cannot yet use is money better kept. Above roughly R250 million the large advisory firms are properly equipped and staffed for the work. The second screen is an intent questionnaire completed before an engagement begins, which establishes what the client intends automation to achieve. The position is stated without hedging: this practice builds for companies that want to scale and grow their headcount, and not for companies looking to cut it. Where the intent is to replace people, the engagement is declined, on the reasoning that a business optimising for fewer humans is optimising for the wrong thing. The bargain runs both ways. Employees are expected to meet automation halfway by upskilling, adapting, adopting the new tools and changing how they work, because automation lifts people who are willing to move with it and leaves behind people who are not. Brynjolfsson and McAfee (2014) argue that digital technologies can complement human labour as well as substitute for it, and that which of the two occurs depends materially on the choices organisations and individuals make. Alongside the two screens sits a filter on the kind of entity. The practice works with companies, trusts and family owned entities, and that is the whole list: it does not take non profit organisations and it does not take body corporates. The reasoning is scope rather than preference. The work here is built around ownership, succession and the transfer of value between generations and between shareholders. A non profit has no owners to succeed and no shares to transfer. A body corporate is governed by its own statutory scheme and serves members whose interests are set by legislation rather than by a shareholders' agreement. Neither is treated as a lesser client, both are simply a different discipline, and a reader in either position is directed to an adviser who specialises in that form, because the governance questions there are genuinely different.
The ethical screen
We turn work down. Here is exactly when, and why.
Any consultancy will tell you it is selective. Very few will publish the rule, because a published rule can be held against you by the next prospect who does not meet it. This page publishes both of ours: the size band we work inside, and the question of intent we ask before anything is designed. If you are going to trust a practice with your structure and your staff, you should be able to read what would make us say no to you.
The first screen
Between R20 million and R250 million.
The band is not snobbery and it is not capacity management. It is the range in which this particular kind of work is the best available use of a client's money, and outside it there are better answers that cost less.
Below R20 million
A good accountant and a clear head beat a structure, every time, at this size. What a business turning over R8 million usually needs is a signed shareholders' agreement, tidy books, one person who knows where the money went and the discipline to look at it monthly. A trust and a holding company add cost, administration and annual returns to a business that has not yet used up the simpler options. We will tell you that for free rather than sell you the alternative.
Above R250 million
The large advisory and audit firms are properly equipped for you: depth of specialist bench, international reach, the capacity to put six people on a question for a month. That is a genuine advantage at that size and it is not one a boutique practice can honestly claim to match. If you are above the band and someone tells you a small firm is better for you in every respect, be careful with the rest of what they say.
In between
This is the band nobody serves properly. Sophisticated enough to need real structure and real automation, too small to be worth a large firm's attention, and quoted accordingly on the rare occasion they ask. It is also the range where a single well-designed structure or one automated process changes the trajectory of the whole business, which is why the practice exists here and not elsewhere.
See the modelThe band is a guide rather than a gate with a guard on it. A business at R15 million with a genuinely complex ownership problem is a fair conversation, and a business at R300 million with one narrow question may be better served here than by a firm that would rather sell it a programme. What we will not do is take on work that a cheaper and simpler answer would have solved.
And the kind of entity
Companies, trusts and family owned entities.
That is the whole list. We do not take non profit organisations and we do not take body corporates. It is a scope boundary rather than a judgement, and it is better read here, at your leisure, than discovered in a meeting you drove across town for.
What we take
Companies, trusts and family owned entities. All three have owners, whether shareholders or beneficiaries, all three eventually have to move value from one generation or one shareholder to the next, and all three live or die on a document that says how. That is the material this practice works in, and the list is short on purpose.
What the foundation coversNot non profit organisations
A non profit has no owners to succeed and no shares to transfer. Succession, the shareholders' agreement, the question of who buys out whom and at what valuation: none of it has a counterpart there. The governance a non profit needs is real, demanding and worth doing well. It is simply a different discipline, and taking it on would mean pretending to a specialism we do not have.
Not body corporates
A body corporate is governed by its own statutory scheme and serves members whose interests are set by legislation rather than by a shareholders' agreement. The questions that decide things there are answered by that scheme, not by the structuring work described across this site. Again, a different discipline, and one that rewards an adviser who does it every week.
Not international structures
Offshore holding companies, cross border groups, entities spanning more than one tax jurisdiction, and arrangements built around an international tax regime. This is a South African practice working under South African law, and local is lekker for a reason: we know this law properly rather than approximately. Structures of that kind need someone who lives in that law daily, and we will say so rather than improvise.
If you run a non profit, or you sit as a trustee of a body corporate, the useful next step is an adviser who specialises in that form. The governance questions in front of you are genuinely different ones, and they deserve somebody who has answered them many times rather than somebody adapting a model built for shareholders. We would rather publish that here than take the meeting and work it out afterwards.
The second screen
The intent questionnaire.
Before an engagement begins, the client completes a structured questionnaire. It is short, it is answered by the people who will actually make the decisions, and its purpose is singular: to establish what this business intends automation to do. Not what it hopes the technology can do. What the people commissioning it intend to happen afterwards.
We ask it in writing and in advance for a simple reason. Intent stated at the start of a project is very different from intent discovered at the end of one, and by the end there is a system built, an invoice paid and a great deal of pressure on everybody to keep going. Asking early is the only version of this question that is worth asking at all.
What the questionnaire asks
What the business wants to be doing in three years and at what size. What the current team spends its hours on, and which of those hours the owners consider wasted. What would happen to the people currently doing the work that is being automated: redeployed to what, retrained into what, on what timeline. What success looks like as a number, and whether that number is revenue, capacity or payroll.
What happens after it
The answer that gets a yes
A business that wants to grow. It is at capacity, its best people are spending half their week on work that should not need a person at all, and it cannot take on more without either breaking or hiring ahead of the revenue. Automate the repetition, move the people to the work that actually needs judgement, and the business grows its headcount rather than shrinking it. That is the client this practice is built for.
How the growth case is built
The answer that gets a no
A business whose stated objective is a smaller payroll. If the intent is to replace people, the answer is no, and we will say it in the first meeting rather than three invoices in. The reason is not sentiment. A business optimising for fewer humans is optimising for the wrong thing: it is managing a cost line instead of building a capability, and it will hit the floor of that strategy quickly and have nothing left to do next.
There is no hedging available here and we are not going to invent any. Both kinds of client exist in this market and both have money. We would simply rather not build the second thing, because the systems described across this site are perfectly capable of being pointed either way, and the only thing that decides which is the intent of the person commissioning them. That decision gets made once, at the start, in writing.
Nor is this a claim that no job ever changes. Roles change constantly under this work: what a person does on a Tuesday afternoon in year two frequently looks nothing like year one. The distinction is between a business restructuring work around people it intends to keep, and a business buying software in order to have fewer of them. Those two are not the same project and they do not feel the same from inside the company.
The other half of the bargain
Employees have to meet this halfway.
Everything above is an obligation we place on the employer. It would be dishonest to publish it without the matching obligation on everybody else, so here it is, stated as directly and with as little contempt as we can manage.
If your employer commits to keeping you and to retraining you, you have to actually do the retraining. Upskill, adapt, adopt the tools, change how you work. Learn the new system rather than waiting for it to be abandoned. Move the parts of your job that a machine now does better and put your hours into the parts that need a person: judgement, relationships, the exceptions, the things that go wrong at four in the afternoon and need somebody who understands the customer.
Automation lifts people who are willing to move with it, and it leaves behind people who are not. That is not a threat and it is not something we are pleased about. It is simply what happens, in every business, every time, and pretending otherwise is not kindness. Telling a person that nothing will change, when it plainly will, takes away the one thing that would have helped them most: enough warning to prepare. The kind version of this conversation is the early and specific version.
Brynjolfsson and McAfee (2014) make the case that digital technologies can complement human labour as well as substitute for it, and that which of the two dominates is not a fixed property of the technology. It depends materially on the choices organisations and individuals make, including whether people build the skills that pair well with machines rather than compete directly against them. We take that as the honest framing for this page: the technology does not decide on its own, and neither does the employer alone.
In practice this means the training portal in bucket four is not decoration on an automation project. It is the mechanism by which the employer keeps their side of the bargain, and the record by which you can see who took it up. Where a client is serious, that portal is busy. Where it is empty six months in, the commitment was never real, and we would rather find that out and say so.
If you have read this far
The rest of the argument.
The value chain model
Governance as the foundation, four buckets on top of it, and why the order they are built in matters more than the tools.
See the model
Governance, the foundation
Structures, constitutions, succession, boards and financial agents. The work that has to exist before automation is worth commissioning.
Read bucket one
People and the lifecycle
Recruitment, onboarding, training and labour law support, and the portal that carries the retraining commitment described above.
Read bucket four
The network
Six divisions, what each is for, when you would need it, and an honest note about what an internal referral is worth.
See the network
One clear next step
Ask for the intent questionnaire.
If what you have read here describes the business you are trying to build, ask for the questionnaire and complete it before we meet. It takes about twenty minutes, it commits you to nothing, and it means the first conversation starts at the real question rather than at the polite one. If your answers put you outside the screen, we will tell you that straight away and suggest where else to look.
Request the intent questionnaireOr read what each engagement involves first.
References
- Brynjolfsson, E. and McAfee, A. 2014. The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies. New York: W.W. Norton.