The Leadership Quadrants
Every piece of work you hand to somebody involves two separate things you can keep or give away. The outcome, which is what a good result looks like. And the method, which is how they get there. Owners tend to treat these as one decision. They are not.
Split them and you get four ways of leading a piece of work. All four are right somewhere. The trouble in most owner-managed businesses is not that the owner picks the wrong one occasionally. It is that they never pick at all, and default to the same corner for everything from a health and safety incident to a marketing plan.
The two things you can hand over
Run the two dimensions against each other and the four modes fall out. Read each one as a description of a task, not a description of a person.
Four modes. The label goes on the task, not on the person.
Direct. You set the outcome and you specify the method. Correct for a new starter in their first fortnight, for anything with a safety or legal consequence, and for a genuine emergency where there is no time for anyone to work it out. Wrong almost everywhere else, because it produces people who wait.
Delegate. You set the outcome and they choose the method. This is where most work in a healthy business should sit. You define what good looks like, by when, to what standard and with what budget. Then you leave the route alone, including the parts of the route you would have done differently.
Guardrail. They set the outcome and you specify the method. It sounds like an odd corner until you recognise it. A qualified person choosing what to work on within a fixed process. A regulated file that has to be done a particular way, regardless of what the individual is trying to achieve. Brand and tone of voice. Health and safety method statements. Here you are not directing the work, you are fixing the rails it runs on.
Release. They set the outcome and they choose the method. Right for a proven leader running a whole function, where you are agreeing a direction once or twice a year and then reading the numbers. It is the corner every owner says they want and the one that does the most damage when entered early, because from the outside release and abdication look identical.
Why an owner needs the distinction
The commonest complaint from owners is "I have delegated and it came back to me". Almost always, what happened was direct dressed up as delegate. The outcome got handed over in the meeting and the method got taken back the following Tuesday, when the owner saw the work being done in a way they would not have chosen and stepped in.
That is not a small thing. The moment you correct the method after handing over the outcome, you have taught the person that the real specification was in your head all along, and the safest behaviour is to check with you first. You will then experience that checking as a lack of initiative, and the loop closes.
A worked example
The figures are illustrative. A business turning over £1.4m has £96,000 sitting in debtors and an average of 68 debtor days. The owner hands credit control to an office manager who has been in post six months.
The direct version, which is what usually happens: "Ring these four, in this order, use this wording, copy me in, tell me what they say." The office manager makes the calls. The owner reviews the replies, rewrites two of them, and picks up the awkward one personally because it is a good client. Debtor days come down for a month because the owner is on it, then drift back. Nothing has been handed over. A task has been lent out.
The delegate version is a different conversation and takes longer to have. The outcome is stated as a number and a date: debtor days under 45 by the end of the quarter, with no client relationship damaged. The constraints are stated plainly: nobody goes to a solicitor without the owner, and payment plans over £5,000 get signed off. The reporting is stated: one figure on a Monday, and a call if a client goes over 90 days. Then the method is genuinely left alone. If she decides to email before ringing, offer a small settlement discount, or change the invoice wording completely, that is hers.
Some of what she does will not be what the owner would have done. That is the price and it is not optional. What the owner buys for it is a credit control function that keeps working during the fortnight he is not there, and one figure a week instead of an ongoing job.
Notice also which parts stayed in guardrail. Legal action and large payment plans are not method choices, they are rails. Naming them at the start is what makes the rest of the handover safe to do.
Name the quadrant out loud
The single cheapest improvement available here costs nothing: say which mode you are in when you hand something over. "This one is direct, do it exactly this way, and I will explain why afterwards." Or "this is yours, I care about the result and the deadline, and I genuinely do not mind how you get there."
Most of the friction owners have around delegation is not disagreement about the mode. It is that the owner picked one silently and the other person guessed a different one.
How to apply it this week
- List the last ten things you handed to someone. Put each in a quadrant based on how you actually behaved afterwards, not on what you intended at the time.
- Find your default. Almost everyone has one corner with seven or eight items in it. That corner is your leadership style whether or not you would have chosen it.
- Pick one task to move from direct to delegate. Choose something that matters but is not existential. Write the outcome as a number, a date and a standard before you speak to anybody.
- Write the rails down before you hand it over. Three or four things that are genuinely non-negotiable, in writing. If your list runs past six, you are not setting guardrails, you are specifying a method.
- Agree the reporting, then stop looking. One number, one frequency, one exception that triggers a call. Checking in between those points is method interference and it undoes the handover.
- Say the quadrant out loud, every time, for a fortnight. It will feel stilted for about three days and then it will feel obvious.
The mistake most owners make
They move down the grid too slowly and then jump too far. Years of direct, followed by a bad month, followed by handing an entire function to somebody with a "you crack on" and no agreed outcome, no rails and no reporting. That is not release, it is abdication, and when it goes wrong it gets used as evidence that delegation does not work here.
The other mistake is quieter. An owner who genuinely wants to delegate keeps ownership of the method for one reason only: the method is the part they enjoy. Be honest about that. If you are holding on to the how because you like doing the how, that is a legitimate choice, but it is a choice about your own job, not a judgement about your team's capability.
The questions to sit with
- Out of 10, how clearly does your team know which mode you are in when you hand them something?
- Which quadrant did you spend most of last week in, and was that the situation's fault or your habit?
- What are you still specifying the method for, and would the outcome genuinely suffer if you stopped?
- Where have you called something delegation while behaving as though it were direction?
This runs right through The Leader Programme. The transition from owner to leader is not a change of personality, it is learning to choose a quadrant on purpose and then stay in it long enough for somebody else to grow.
