The most useful question I ask founders is not about strategy. It is whether they could go away for two weeks with their phone off.
The answers are revealing. Almost nobody says yes. Most say something like "in theory", or "I could, but there's a lot on at the moment", or my favourite, "I did last year and it was fine, though I did check in a few times". That last one is the answer of somebody who did not go away.
The follow-up question is better. What specifically would go wrong? And here the answers stop being about workload and start being about judgement. Pricing on a non-standard job. Whether to push back on a client who is being difficult. Whether that candidate is right. Whether we can promise the delivery date.
Notice what none of these are. None of them are tasks. Everybody in the business can perform tasks. What they cannot do is decide, because the founder has never handed that over, and in most cases has never noticed that this is the thing they are holding.
01Why does hiring more people make this worse?
Because the constraint is decisions per week, and every new person produces more of them.
Founders under pressure reason like this: I am overloaded, so I need more capacity, so I will hire. It is a completely sensible response to a capacity problem, and it is precisely the wrong response to a decision problem. If every non-routine judgement in the business routes through one person, adding people behind that person lengthens the queue in front of them.
You can watch this happen. The business goes from fifteen to twenty-five, the founder's diary goes from busy to unmanageable, and the observable symptom is that everything takes longer. Not because anybody is slower. Because more things are waiting.
The tell is the phrase "I just need to check with him" appearing in more conversations than it did a year ago. That sentence is the bottleneck, spoken aloud.
02What is the difference between delegating a task and delegating a decision?
A task is something with a defined outcome that you can hand to someone competent. A decision is a judgement made under uncertainty, where the founder's version and everyone else's version differ, and nobody has written down why.
Most delegation stops at the first and never reaches the second. The founder hands over quoting, but reviews every quote. Hands over recruitment, but sits in on every final interview. Hands over the client relationship, but takes the difficult calls. In each case the person now does the work and the founder still makes the call, which means the founder's diary has not improved and the other person has learned nothing about judgement.
Worse, it teaches the organisation something corrosive: that decisions are not really theirs. People stop forming views. They bring you options rather than recommendations, because a recommendation might be wrong and an option cannot be. Within a year you have a team of capable people who have been trained not to think, and you conclude, wrongly, that you cannot find good people.
You will know the transfer has worked when you find out about the decision afterwards, and you disagree with about one in five, and you say nothing.
03How do you actually transfer a decision?
By writing down the judgement, which sounds impossible until you do it, because most of it is not intuition at all. It is a set of rules you have never articulated.
Start with one decision. Pick a real one that happens weekly, not the hardest one you make. Pricing a non-standard job is a good first candidate for most businesses.
Write the last ten times you made it. Not in theory. The actual ten, with the actual context, and what you decided, and why. This takes about two hours and it is the entire exercise. What comes out is a set of factors that turn out to be surprisingly consistent: you go higher when the client is disorganised, you go lower when the work will teach you something, you walk away when they have already been to two other suppliers.
Include the ones you got wrong. These are more instructive than the successes and they do something else useful: they make it safe for the next person to get one wrong.
Define the boundary. Where can this person decide alone, where must they consult, and where must it come back to you? Boundaries by value are easiest to start with, though boundaries by type are usually better. "Anything under £40k, decide. Anything for a new logo in the public sector, talk to me first."
Agree the review, and make it retrospective. You look at a sample afterwards, weekly at first, then monthly. Not approval. Review. The distinction is everything, because approval keeps the bottleneck and review dissolves it.
04What about the dip in quality?
There will be one. Plan for it rather than pretending it will not happen, because the founders who fail at this are usually the ones who were surprised by it.
For about three months the decisions will be worse than yours. Some will be worse in ways that cost real money. This is not a sign that you picked the wrong person or that the business is not ready. It is the cost of the transfer, and it is a smaller cost than the alternative, which is remaining the constraint for another five years.
What helps is deciding in advance what an acceptable dip looks like. If the person is making the call and you would have made a different one in one case out of five, that is a working handover. One in two means the standard was not written clearly enough, so go back to the worked examples. Five out of five means you have a hiring problem, not a delegation problem, and you should address that directly rather than by taking the decision back.
After the dip, something better happens. Decisions start being made by the person closest to the work, faster, with information you never had. That is not a consolation. That is the actual benefit, and it is why the businesses that do this outgrow the ones that do not.
05What should you keep?
Not everything should go. Founders sometimes overcorrect and hand over things that genuinely need them, which is its own kind of failure.
Keep the decisions that set direction: what business we are in, what we will not do, who we are for. Keep the ones about people at the level below you. Keep the ones that are genuinely irreversible, because reversibility is the best guide there is to what needs a founder.
Hand over everything that is frequent, reversible and bounded. That is most of what fills your week.
There is a version of this founders find hard to hear, which is that holding on is often not about risk at all. It is about the decisions being the enjoyable part. Pricing the interesting job, having the difficult client conversation, picking the person. If that is what is happening, own it, and be deliberate about which few you keep for that reason rather than dressing them up as too important to hand over.
06The two week test
At some point, run it properly. Two weeks, phone off, no checking in, one named person holding everything.
Tell the team a month beforehand, so the anxiety surfaces while you are still there to work through it. Write down, before you go, the three things you expect to go wrong. Then compare that list with what actually happened when you come back.
Almost every founder who does this comes back to a business that coped better than they expected, and to one or two genuine failures that are extremely informative about where the remaining dependencies sit. That is the point. The holiday is a diagnostic, and it is a considerably cheaper one than discovering the same information during a sale process, when a buyer is asking what happens if you are not there and the honest answer is that nobody knows.