Reference
Holding company or operator: the roles a corporate centre can play
Every group of businesses has a centre, and no two centres do the same amount. A centre is right when the way it works fits what its businesses need, and it does that better than any other owner could.
At one end the centre owns the businesses, funds them and reads their results. At the other it writes their plans and runs their shared functions, and it takes their big decisions. Neither end is right.
The mistake is not choosing the wrong role, but not choosing at all. Such a centre drifted into its role one decision at a time. It costs what a hands-on centre costs, yet behaves like a shareholder.
Two jobs, two kinds of management
A business competes in a market for customers. Its job is to win them with an offer that beats the alternatives. Its test is competitive advantage.
A corporate centre competes in a different market: the market for owning businesses. It stands between the investors and the businesses. It has no customers of its own, so it works through the businesses it owns.
Its test is whether each business is worth more in this group than with any other owner. Both kinds of management answer to the same investors. They earn their return in different places.
That difference makes the centre's role a choice. A parent can make a business worth more in several ways, and each asks for a different centre.
Five roles, from hands-off to hands-on
| Role | What the centre decides | What the businesses decide | What it suits |
|---|---|---|---|
| Holding company | Who runs each business, how much capital it gets, whether to keep it | Everything else | Unrelated businesses that are each well led |
| Financial controller | Profit and cash targets, set tight and checked monthly. Capital against a return hurdle | Their own strategies and plans | Mature, steady businesses where the gains come from discipline |
| Strategic controller | Approves each business's strategy after challenging it. Balances long-term goals against this year's numbers | Write their own strategies, within limits the centre sets | Groups with some links between businesses and some that need building |
| Strategic architect | Works on each business's plan with its leaders, sets long-term goals, coordinates across businesses | Execution | A few related businesses in markets that are changing fast |
| Operator | Runs shared functions, sets common processes, takes the major decisions | Selling and serving customers | Businesses so linked they are divisions of one business |
The three middle roles are the classic description of how a parent company manages its businesses. At each end, the holding company does less than a financial controller. It appoints and funds, but does not set operating targets. The operator does more than an architect. It shapes the plan and also runs part of the machine.
Moving along the continuum, the centre uses more of the ways a parent can add value. A holding company appoints leaders and allocates capital, and nothing else. A financial controller adds targets and follow-up. A strategic controller adds challenge to strategy.
An architect and an operator add the linkages. These are pooled buying, shared assets and skills, and coordination between businesses that would otherwise compete or duplicate each other. Each step costs more and adds value only if the businesses need it.
What decides the right role
How related the businesses are. Some businesses share no customers, suppliers, skills or assets. The centre has nothing to link. Every step right of financial control adds cost and nothing else.
Businesses that share a great deal leave money on the table under a hands-off centre. None of them will build the shared capability for the others.
How settled their markets are. A mature business in a stable market gains most from tight targets and capital discipline. A business in a fast-moving market needs a parent that understands the change. That parent must be willing to fund a plan that will not pay back this year. Tight monthly profit targets do not suit that business.
What the centre is good at. Some centres are run by people who built and ran businesses in the group's core industry. They can challenge a strategy credibly.
A centre run by financiers and lawyers can allocate capital well. It will do harm if it starts writing business unit plans. The role has to fit the people the centre has.
Whether it can do this better than another owner. Suppose a specialist owner in the same industry would add more than a hands-on centre does. Then the centre is a cost. For every business, ask whether this role, played by this centre, makes it worth more than the alternatives would.
How a mismatch shows
Mismatches come in three kinds, with familiar symptoms.
The centre is doing too much for the businesses it has. The businesses treat central functions as free, so demand for them grows, and so do the functions. Business leaders say "corporate charges us and we have no idea for what." The centre reviews operating decisions it does not understand. The good business unit managers leave for a company that will let them run something.
The centre is doing too little for the businesses it has. Each business rebuilds the same capability separately: its own systems, its own buying, its own graduate programme. Two businesses in the group bid against each other for the same customer.
The centre does not know which role it is playing. This case is the commonest. The group calls itself decentralised, but the centre approves every hire above a set level. It says it manages a portfolio, but every business gets the same capital rules and the same targets. It says it adds value through linkages, but nobody can name one.
Reading your own centre
Write the role in one sentence, then test it against last year.
List the decisions the centre took for each business in the last twelve months. Include leaders appointed, targets set, strategies sent back, capital granted or refused, functions merged and deals done. Put each decision on the line above.
If they cluster where the one sentence says, the centre knows its role. If they are scattered from holding company to operator, it does not. The businesses are working that out one surprise at a time.
Then ask the businesses which of the centre's decisions they would have made the same way on their own. Whether they like the centre is beside the point. The decisions they would not have made are the centre's contribution. If there are none, the centre is a holding company, whatever it calls itself. It should cost what a holding company costs.
A centre can play different roles for different businesses. It might keep close control of a business being built, and stay hands-off with a mature one. That works when the role for each business is stated and the business knows it. It fails when the role changes with the mood of the monthly review.
What the strategy check assumes
The group questions in the strategy check start by asking how the centre runs the businesses, using the roles on this page.
A holding company or a financial controller then answers questions that fit its model. They ask whether capital goes where it earns its return, and whether the centre costs what its role needs. Other groups answer questions on how the centre gives each business a role and moves capital between them. The rest of the check applies to the group as a whole.
Frequently asked questions
Is a holding company a bad model? No. It is the right model for unrelated businesses that are each well led, and it is the cheapest centre to run. It becomes a bad model when it carries the cost of a hands-on head office. Or when the businesses are related enough to need more than an owner.
Which role is most common? In our experience, the unwritten one. Most groups we have worked with could describe their centre's role in a sentence, and most of the time their decisions put it somewhere else.
Should the role change over time? Yes, when the portfolio changes. A group that sells its unrelated businesses and keeps a related core has fewer reasons to be a holding company. A group that buys into a fast-changing market needs a centre that can engage with it. Or it needs an arrangement that leaves that business alone on purpose.
How big should the centre be? As big as its role, and no bigger. The role comes first, and the size of the centre follows from it. The businesses complain about the centre that grew first and looked for a role afterwards.