Every founder hits the same wall at the same place: the hours run out. Skill can grow, prices can rise, systems can tighten — but the day stays twenty-four hours long no matter how good you get, and a business capped at your personal hours isn’t a company. It’s a job with extra paperwork.
Delegation is the only exit. And here’s the direct answer to how to delegate effectively: delegate outcomes, backed by written process — then inspect what you expect, on a schedule, and budget for the mistakes of the learning period like the tuition they are. Delegation almost never fails because people are bad. It fails because the instructions lived in your head, and your head doesn’t ship with the task.
I’ve been handing off work since I was a kid — I ran my first real company on other people’s hands before I could drive myself to the job sites. Which is also how I learned the failure mode, in writing, from the federal government.
Why delegation feels impossible
The founder’s math error goes like this: they’ll do it at 80% of my standard, so I’ll just do it myself. The comparison feels rigorous and is completely wrong, because it compares quality-per-task when the actual variable is capacity. A lawn cut at 92% of your standard, done without you, beats the 100% lawn that caps the company at your calendar — I learned that with literal lawns, and it’s stayed true through every business since. You’re not choosing between good work and slightly worse work. You’re choosing between a ceiling and a company.
Underneath the math error is an identity one: nobody cares like I do. True — and irrelevant. Nobody will ever care like you. The job was never to find people who care like founders; it’s to build process sturdy enough that ordinary, decent effort produces your standard anyway. Systems carry the caring. That’s what they’re for.
The letter from the Postmaster
Here’s my tuition receipt. As a kid running my lawn company, I printed hundreds of flyers — genuinely well-designed, I was proud of them — and sent friends out to hang them on doors and mailbox posts. Simple instruction. Off they went; flyers everywhere; leverage achieved. I’d multiplied myself.
Except the crew, optimizing for speed the way every crew optimizes for speed, stuffed the flyers inside the mailboxes. Which — as I then learned — is federally illegal. Mailboxes belong to the postal system, and the postal system noticed. A letter from the Postmaster arrived, and let me tell you, there is a specific feeling attached to your first piece of official government correspondence being about your marketing department.
Read the failure precisely, because it’s the whole lesson: nobody was lazy and nobody was dumb. The instruction — “hang these on doors” — was complete in my head, where it lived alongside everything I knew and never said: not in mailboxes, doors specifically, here’s why it matters. In the field, the unsaid parts became their judgment call, executed at my expense. Everything you fail to specify becomes a decision someone else makes for you, at full speed, in your name.
Delegate outcomes, document process
The fix has two halves that people keep trying to use separately. Delegating the outcome gives the work an owner — one person, a definition of done, a deadline. Documenting the process gives the outcome a route: the steps, the standards, the boundaries, and — critically, per the Postmaster — the failure modes. What wrong looks like. The two or three ways this task goes sideways and the one rule that must never be broken even if everything else improvises.
A process that lives in your head isn’t a process; it’s a mood with seniority. The test for whether it’s actually documented: could a smart stranger execute this from the document alone, without a single “quick question”? Until yes, you haven’t delegated — you’ve just distributed the confusion. And write it once, properly: the hour a checklist costs gets repaid every single time the task runs without you, which is the entire arithmetic of leverage.
Cheapest documentation trick I know: annotate a finished example. A photo of the job done right, marked up with the three things that make it right and the one thing that would make it wrong, outperforms a page of prose — because people don’t execute descriptions, they copy references. Give the crew the reference. “Make it look like this, never like that” survives the field better than any paragraph ever written.
Inspect what you expect
Handing off the work is half the discipline. The other half is the cadence nobody enjoys: inspection. Early on, tight loops — spot-check the actual output, not the summary of it, while mistakes are still cheap and correctable. As the track record accumulates, loosen deliberately: weekly becomes monthly, checking everything becomes sampling. The loop never quite reaches zero, because inspection isn’t distrust — it’s the maintenance cost of the leverage, the same way checking the oil isn’t an insult to the engine.
Two rules make inspection livable for everyone involved. Praise in public, correct in precise private — vague criticism teaches anxiety, not standards. And when the same mistake shows up twice from different people, stop blaming people: the process is the defect. Two identical failures from two different hands is the system confessing.
Price in the tuition
Delegation has a tuition phase, and pretending otherwise is how founders talk themselves back into doing everything. The first weeks of any handoff run slower and rougher than doing it yourself — that’s not the experiment failing; that’s the price of the asset being built. Budget for it the way you’d budget for equipment: expected, capped, and paid on purpose. My flyer campaign cost me a federal warning and some pride. It bought me a rule I’ve used in every company since. Cheap, honestly.
The alternative has a price too — it’s just deferred and much larger: every task you refuse to hand off is a permanent lease on your hours, and your hours were supposed to be buying something bigger.
What never gets delegated
Honesty about the boundary. The standard itself stays with you — what good means, what the customer must feel, which unscalable touches survive growth on purpose. Culture stays with you, because culture is just the founder’s behavior, photocopied. And accountability never transfers at all: you can hand off the task, the route, the whole department — but when the letter comes, it has your name on it. It always has your name on it.
That’s not the fine print of delegation. That’s the deal. You get the leverage; you keep the blame. Founders who accept both halves scale. Founders who want the leverage without the letter stay small, safe, and personally exhausted.
Write it down. Check the work. Keep the blame. That’s the entire technology of getting big.
Somebody eventually has to hand over the mower. Do it with a checklist, a calendar, and the full understanding that the Postmaster writes to the owner — never to the crew.
Tyk Sullivan
Serial entrepreneur in Tampa, Florida. Founder of Plutus Capital and NUMEN. More about Tyk →