Operating model decisions: end-to-end or functional, make, buy or partner

What an operating model is#

An operating model is the way an institution arranges its people, processes, technology and suppliers so that its strategy can actually be executed. It is a set of decisions, not a diagram: who does which work, where each decision sits, what the institution does itself and what it obtains from others, and how the pieces are held together.

The phrase target operating model is often used for the document that describes the intended arrangement. The document is useful; the decisions inside it are the thing. An institution can have a handsome target operating model and no operating model at all, if the decisions were never actually taken and the organisation continues to run on habit.

What it is for#

The purpose of an operating model is narrow and demanding: to let the strategy run. A strategy that promises speed on a model built for control, or scale on a model built for bespoke service, will not be delivered however good the people are. The COO’s first question of any model is therefore not whether it is efficient or elegant but whether the promises the institution has made can be kept on it.

The first decision: end-to-end or functional#

The first shaping decision is whether work is organised end-to-end — a team owns a product or a customer journey from beginning to end — or by function, where specialists in each discipline serve every product. End-to-end models buy accountability and speed at the cost of duplicated expertise and inconsistent standards. Functional models buy depth and control at the cost of hand-offs, and of a customer whose problem crosses three departments and belongs to none.

Neither is right in general; each is right for a strategy. The cost of choosing wrongly is not paid on the day of the decision but for years afterwards, in the gap between what the institution promised and what its structure can deliver. This is why the choice belongs to the seat and not to a reorganisation exercise.

The second decision: make, buy or partner#

For each capability the institution needs, the second decision is whether to make it, buy it or partner for it. This is usually treated as a procurement question, and that is the error. Sourcing decisions — including where the institution’s systems and data are hosted, and how concentrated it becomes on a small number of providers — are balance-sheet commitments the institution lives with for a decade. They shape cost, control, resilience and the ability to change course, long after the contract that created them has been forgotten.

A make, buy or partner decision taken well states four things openly. The fourth is the one most often missing.

• What the institution keeps, because it is the source of its advantage or its accountability.
• What it buys, because others do it better and the dependence is acceptable.
• What it partners for, because the capability must be shared to exist.
• What the exit would cost, in each case.

An institution can have a handsome target operating model and no operating model at all, if the decisions were never actually taken and the organisation continues to run on habit.

Where digital changes the model, not the process#

Much of what is called digital transformation leaves the operating model untouched: the same work, done by the same people in the same place, faster. That is process improvement, and it is valuable, but it is not a model decision. Digital reaches the model only where it changes who does the work, where the decision sits or what is sourced — where a capability that had to be built can now be rented, where a hand-off disappears because the customer completes the step, or where a function that existed to move information no longer needs to exist. Those are the cases the COO must recognise, because they reopen both of the shaping decisions above.

Where operating models go wrong#

Operating models rarely fail in the design. They fail because the design was never chosen — the institution grew into its shape by accretion; because the sourcing decisions were taken one contract at a time until the institution discovered its concentration by accident; or because the model was set for a strategy that has since moved, and nobody was accountable for noticing. Each of these is invisible in a diagram and obvious in a decision: name the last promise the institution could not keep, and trace it back.

What to do next#

Start with the strategy, stated as the promises it makes, and ask of the current model whether each promise can be kept on it. Then take the two shaping decisions deliberately and in writing — end-to-end or functional, and make, buy or partner for each material capability — with the exit cost stated for every sourcing choice. Digital enters where it changes one of those answers. Working those decisions through real cases, with peers who carry the same seat in other industries, is the first movement of The COO: Operations at Scale.