When Should You Delegate Work to Your Team?
Once someone can do a task at roughly seventy percent of your standard, it's time, and that usually comes sooner than feels comfortable. Hand over the decision with the task in a short written note, and keep only the few calls that are hard to undo.
If I could give my younger self one instruction for working with a team, it would be to delegate faster than seemed reasonable, and then to keep going past the point where it started to feel like negligence. I don't say that because delegation is a fashionable virtue. I say it because the cost of waiting is paid mostly by people other than you, which makes it very easy to miss.
The standard advice is to delegate once you've found the right person, or once the business can afford the help, or once the task is documented. All three sound careful. In practice they are three separate reasons to wait, and an owner who is good at the work can always find a fourth.
Why do expert owners wait too long to delegate?
The owners most prone to this are the ones who arrived as the expert. If you built the company on your own skill, or you stepped into a leadership role as the person who had done that exact job before, you know what good looks like. You also know it will take you forty minutes to do the thing and a week to work with someone on it, and the forty minutes wins every time.
Each choice is rational. Added up over two years, you've become the most expensive person in the building doing some of the least leveraged work, and you've told everyone around you, without a word, that the interesting work stays with you.
I did this in a way that looked like generosity. A team member would be drafting a quote or a schedule or a reply to a client, I'd see it was slightly off, and I'd fix it, because that was faster and the client would never know. I told myself I was protecting them. What I was teaching them was that their version was a draft for me. They checked their own work a little less carefully each month, because there was a safety net, and the safety net was me.
Most of the growth people get from a job comes from owning an outcome, including the part where it goes a bit wrong and they have to deal with it. If I remove the slightly wrong outcome every time, I remove the growth along with it. Nobody puts that on a spreadsheet, and I now think it's a bigger cost than the hours.
Should you wait for the right person before delegating?
I used to say I'd delegate when I found the right person. What I eventually noticed is that you mostly find out who the right person is by delegating. Nobody shows up with a label. Someone who looks average on tasks you give them with a leash may be excellent when the leash is gone, and the reverse happens too, which is useful to learn early and cheaply.
So I use a blunt rule for deciding what to hand over. The first time I do a task, I do it myself, because I'm still working out how it goes. The second time, I write down what I did while I do it, three or four lines. The third time it goes to someone else. It's crude, and it ends the argument about whether a task is "worth" delegating, an argument owners tend to win against themselves.
How good does someone need to be before you delegate?
The second rule is about quality. If someone can do the task at roughly seventy percent of how I'd do it, it's theirs. Not theirs with my review on top. Theirs.
People push back on this, so here is the reasoning. The gap between seventy and a hundred is mostly made of preferences, a few details of taste, and a small number of genuine standards. The genuine standards you can name and teach in an afternoon. The preferences you'll find, within a few months, that you can live with, and one or two will turn out to be better than yours. Waiting for ninety-five percent means waiting for the person to turn into you, which won't happen and wouldn't help if it did.
There is a real price. Early work comes back worse than you'd have done it, and now and then a customer notices.
I handed a recurring client report to someone who had been with us a few months, and the first two went out with correct numbers in a layout the client found confusing. The client emailed me, politely, and I spent an uncomfortable afternoon wondering whether I'd made a mistake. The mistake was mine, but a different one from the one I suspected. I had handed over the task without handing over the one thing that client cared about, which was that the summary at the top had to answer a specific question. I'd known that for years and never said it because I'd never needed to.
The third report was fine. By the fifth it was better than mine had been. I came close to taking it back after the second one, and I'm fairly sure I would have if the client had been a bigger account.
Why does delegation keep failing?
When delegation fails, the owner usually concludes the person wasn't up to it. Having watched this go wrong many times, I think the more common cause is that the owner delegated the task and kept the decision. The person does the work, then waits for sign-off on every judgment call, and the owner experiences this as "they keep needing me." They keep needing you because the structure told them they weren't allowed to decide.
A hand-off that holds up has four parts, and I put them in a message that takes about five minutes to write:
- What the outcome is and how I'll know it's good.
- What they're allowed to decide without asking, including money and what to say to the customer.
- What I'd like to hear about afterward, which is different from before.
- A date when I'll look at it, so nobody has to wonder whether silence means approval.
The third part matters more than it looks. "Decide, and tell me after" gives someone real authority and gives you the information you wanted, without turning you into a gate.
What should you not delegate yet?
Not everything moves at seventy percent. I keep a short list of things that stay with me, and it has a particular shape: decisions that are hard to undo, or that tell the team who we are. Letting someone go. Changing prices. Anything with a legal signature. Taking on a client I have a bad feeling about. Even on that list, the preparation can be handed over. Someone else can gather the facts, draft the recommendation, and tell me what they would do, and then I make the call without having to dig.
I look at that list again every six months. It has gotten shorter each time. When an item survives, I try to say out loud why, because "only I can do this" is a belief far more often than it is a fact.
What is freed time for?
When I say delegate more, I'm not arguing for a smaller role. The hours you get back have a better use than rest, though rest is fine too: paying attention to people. Who is ready for more? Who is bored? Who is stuck on something a ten-minute conversation would clear? In a business where the owner is busy doing tasks, those questions go unasked, and good people leave without ever having complained about anything.
If I were starting over, I'd cut my own task list down to what only I could do. I'd spend some of the freed time on one conversation per person about what they'd like to own next. And I'd expect some of the first hand-offs to wobble, and let them. I would have scaled faster, and with less strain, if I had tolerated that sooner.
I'm not certain where the line sits for every business. A two-person shop with thin margins can't absorb a botched hand-off the way a team of fifteen can, and I'd be more careful there about what goes first. But I've seen far more owners wait too long than go too fast.
How do you know delegation is working?
Your own list is down to the calls on the keep-list and conversations about who owns what next. Work comes back to you finished, with a note on what was decided, instead of arriving for approval. When something goes slightly wrong, the person who owns it fixes it and tells you afterward, and you improve the hand-off message instead of taking the task back. The keep-list shrinks at each six-month review.
How can you test it this week?
Write down everything you did last week that someone else could have done at seventy percent. Be honest, because owners tend to list the tasks they dislike and quietly keep the ones they enjoy. Circle the three you'd least like to give away. Those are the most likely to be about identity rather than skill, and they are often where the most room for someone else to grow is sitting.
Then hand one over with the four parts above, and leave it alone for two weeks. If it comes back worse than you'd have done it, ask what the instructions were missing before you ask what the person was missing. It's the best test I've found for whether you delegated or merely offloaded.
Frequently asked questions
What if a customer complains about delegated work?
Answer the customer, then send the issue back to the person who owns the work along with whatever the hand-off message was missing. If the same complaint returns after the hand-off is fixed, that is a skill or fit question, and it deserves a direct conversation.
How do I delegate if I don't have anyone yet?
Then the first delegation is a hire, and the hand-off message doubles as the role description. Write the outcome, the decisions the person can make and what you want to hear about afterward before you post the job. That gives you a way to judge candidates and a first week that doesn't depend on you explaining everything.
How long until a delegated task stops needing me?
Give it a few rounds and a set review date rather than a fixed number of weeks. If each round needs less from you, it is working. If it needs the same amount, compare what you are still correcting with what the hand-off said they could decide. An unstated standard or a fuzzy limit usually explains it.
Which hand-off would free the most time?
A Planning Call gives us time to look at what still runs through you and choose the first hand-offs that would give you the most time back.
