We are often briefed on a role that sits in a structure nobody has quite finished designing. The head of a new function, in a reporting line still being argued about, with a remit that will "settle down once someone is in post".
It rarely does. The person arrives, finds the boundaries unclear, and either fights for territory or leaves. We would rather spend two weeks on the design first. Three questions cover most of it.
What does this structure have to deliver in three years?
Not what it does now. If the plan says double the revenue from services, the structure needs a services leader with authority over pricing and delivery, not a business development manager reporting to sales. Write the outcomes first and the boxes second.
Where do the decisions sit?
For each significant decision, who makes it, who must be consulted, and who just needs to know. This is dull work and it is where most structures fail. Two roles that both think they own pricing will produce a conflict inside six months, and the search brief for either role will be wrong.
What does each role need in a person?
Only now write the role. The outcomes it owns, the decisions it makes, the relationships it holds, and from that the experience and capability required. This is the brief. It usually looks different from the job description that would have gone out two weeks earlier.
What this costs and what it saves
For a structure with three or four senior roles this is two to three weeks of work with the CEO and the relevant executives. The saving is one avoided search, which is the usual cost of recruiting someone into a role that was not properly defined, plus the year it takes to notice.
We do not sell organisation design as a standalone discipline. We do it because it is the difference between a search that works and one that does not.