Learning how to run a small business without doing everything yourself is not a discipline problem. I want to say that plainly, because most owners who ask me this question show up apologizing for something — that they're disorganized, that they can't let go, that they're bad at delegating. Almost none of that is true. What's actually happening is a design flaw: the business was built with exactly one person as the default handler for every unassigned task, and nobody ever went back and reassigned them.
Here's the through-line for everything below. Every hour on your calendar is either a decision only you can make or a default nobody ever assigned. That's it. Two categories. The first category is your job. The second category is a design defect you can fix in about six weeks. Every move in this list exists to separate one from the other and then move the second pile off your plate.
Seven moves. They're in order for a reason — doing number five before number one is how owners end up with a shelf full of unused procedures and the same 62-hour week.
1. Log a real week in 30-minute blocks. Do not estimate it.
You do not know how you spend your time. I didn't either. When I finally logged a week honestly, I found eleven hours of work I would have sworn took three.
The rule is simple: for five working days, at the end of every 30-minute block, write one line about what you actually did. Not what you meant to do. A phone note, a legal pad, the back of a job ticket — the medium doesn't matter, the honesty does. Log the interruptions as their own blocks: "tech called about the wrong part," "customer wanted to move Thursday," "approved a quote."
Two things make this fail. The first is reconstructing the log on Friday from memory, which produces a flattering fiction. The second is logging a "good" week — pick a normal one, ideally a busy one, because the busy week is the one that's actually breaking you. If you want the longer version of this exercise with the arithmetic attached, I walked through it in detail in the six-Saturday audit.
You'll end up with roughly 80 to 100 blocks. That list is the raw material for everything else. Without it, the next six moves are guesswork.
If this sounds like your week, see how owners hand this off.
2. Label every block with one of four letters
Go through the log and write a single letter next to each block. No paragraphs, no agonizing — first instinct.
- O — Only me. Genuinely requires the owner. Be stingy here; we'll test it in move three.
- T — Transferable today. Someone on the team could do this now, with a five-minute explanation.
- S — Needs a system first. Transferable, but only after somebody writes down how it works.
- K — Kill. Shouldn't exist. Rework, a report nobody reads, a step that survived a software change three years ago.
The K pile is usually the surprise. In most service businesses I've looked at, somewhere between 10% and 20% of the owner's logged blocks are pure rework — re-measuring because the estimate was vague, re-scheduling because the confirmation never went out, re-invoicing because the change order lived in a text message. That's not delegation material. That's a broken upstream step, and handing it to a new hire just gives them your headache. If your K pile is fat, read the difference between firefighting and root cause before you hire anyone.
3. Write the "only me" list — then cut it in half
Take your O blocks and consolidate them into a written list of functions. Most owners' first draft has twelve to fifteen items. The real list is usually four to six.
Apply one test to each item: if I were unreachable for two weeks, would the business be materially damaged by someone else doing this, or would it just be done differently than I'd do it? "Differently" is not damage. "Differently" is the price of getting your life back.
What survives that test, in my experience, is a short and fairly consistent set: signing legal and financial commitments, setting the pricing floor, hiring and firing, the handful of relationships that genuinely trade on you personally, and defining what "good work" means for the crew. Notice what isn't on it — quoting individual jobs, approving individual purchases, answering the phone, chasing an unpaid invoice, or being the person who knows where the spare key is.
What can a small business owner actually hand off first?
Hand off the recurring, date-driven work before the judgment-driven work. Scheduling confirmations, invoice follow-up, payroll deposits, contractor onboarding paperwork, and job-completion photos are all governed by a rule and a calendar, not by taste. They can move to another person with a one-page procedure and no loss of quality. Estimating, pricing exceptions, and personnel decisions transfer later, after the person has demonstrated judgment on the rules-based work first.
4. Clear the date-driven admin off your calendar before you touch anything else
This is the move most owners skip because it feels small. It isn't. Date-driven compliance work is the most delegatable work in your entire business and the most expensive when it slips, and it sits on the owner's plate almost everywhere I look.
The recurring items in a typical service business with a crew:
- Federal payroll tax deposits, on a monthly or semiweekly schedule set by your lookback period, plus Form 941 at the end of the month after each quarter.
- Form I-9 completed by the employee no later than their first day and by you within three business days of the start date.
- 1099-NECs to subs and to the IRS by January 31 — and if you file 10 or more information returns in total, electronically.
- Time and payroll records retained under the FLSA: three years for payroll records, two years for the time cards and schedules the wage computations rest on. The Department of Labor's recordkeeping fact sheet spells out exactly which records and for how long.
- State-specific insurance obligations — worth noting that Texas is the only state where private employers may generally opt out of workers' comp, so if you operate across state lines you cannot run one checklist.
None of that requires the owner. All of it requires someone, on a date, every time. Assign each item to a named person and a recurring calendar entry this week. If nobody on the team can hold it, that's the argument for an outsourced back office — clean books, current filings, and a real records trail are the plumbing of a well-run company, not a luxury. I broke the functions down station by station in the 8-station back-office audit.
5. Turn each T and S block into a one-page operating card — not a manual
Here's where good intentions die. Owners decide to "document everything," open a blank document, and quit on page four of a 40-page manual nobody will read.
Write one page per task instead. Five fields:
- Trigger. What starts this? ("Job marked complete in the field app.")
- Steps. Numbered, in the tool you actually use. Name the software — Jobber, Housecall Pro, ServiceTitan, QuickBooks Online, whatever it is. Screenshots beat prose.
- Done looks like. The observable end state. ("Invoice sent, photos attached, customer texted, job status moved to Invoiced.")
- Common exceptions. The three or four things that actually go sideways, and what to do about each.
- Escalation line. The exact condition under which this comes to you. Be specific: "any change order over 15% of the original quote," not "anything unusual."
Write it while doing the task, not from memory — narrate into your phone on the drive back and clean it up later. A card takes fifteen minutes. Ten cards covers most of what's clogging your week. If you want the fill-in structure, it's in the SOP template walkthrough.
Field five is the one that carries the weight. Most handoffs fail because the escalation line is undefined, so the employee escalates everything to be safe, and the owner concludes that delegation doesn't work. It worked fine. You just never told them where the edge was.
6. Hand off with a checkpoint, not a hope
Handing someone a document and asking "make sense?" is not a handoff. It's a wish. Use three touches instead, and put dates on them:
- Touch one — you do it, they watch and correct the card. They edit the page as you work. Their edits are the proof they understood it.
- Touch two — they do it, you review 100% of the output for two weeks. Review the work product, not the process. If the invoice is right, don't relitigate the clicks.
- Touch three — sampled review at a stated rate. One in five, then one in ten, on a fixed day. Written down, not "I'll spot-check when I think of it."
The checkpoint is what separates delegation from abandonment, and it's the piece owners drop when they're busy — which is precisely when the handoff rebounds and confirms the story that they have to do it themselves. A sampled review that survives contact with a busy week is worth more than a thorough review you skip. This is the same discipline behind a real quality control process: defined standard, defined sample, defined response when it misses.
7. Put a number on it, then re-run the log in 60 days
Before you start, write down two numbers: total hours logged, and hours in the O pile. Then set the target — not "work less," but "move 12 blocks off my plate by [date]."
Sixty days later, run the same five-day log. You're looking for three things: did total hours drop, did the O share rise as a percentage (it should — you're keeping the owner work and shedding the rest), and did anything bounce back? Bounced work is the most useful signal you'll get. It almost always points at one of two failures: the card had no escalation line, or the checkpoint got skipped. Fix that specific card. Don't take the whole function back.
The failure mode I'd warn you about hardest: the owner who does moves one through six, gets a clean 15 hours back, and immediately fills all of it with more selling. The hours don't come back on their own. Block them before they're free — one half-day a week for the work that has no deadline attached and therefore never happens. The owner's weekly operating rhythm is how you protect that block from the business.
Why this sequence, and not a bigger org chart
Owners often skip straight to hiring, on the theory that another body absorbs the overflow. It rarely does. A new hire in an undocumented business doesn't reduce your load — they add a training load on top of it, and six weeks later you're doing your old job plus supervision. The Bureau of Labor Statistics business survival data shows roughly half of new establishments make it to five years; the ones I've watched come apart mid-growth were rarely short on demand. They were short on a second person who could execute without the founder in the room.
The sequence matters because each move makes the next one cheaper. The log tells you what's real. The letters tell you what's transferable. The "only me" list keeps you from handing off the things that genuinely require you. The cards make the handoff teachable. The checkpoint makes it stick. The re-log tells you whether any of it worked.
Do the log this week. That's the whole ask. Everything else follows from an honest page of paper, and one of the reasons we built Turnkey Services the way we did is that most owners already know what's broken — they just haven't seen it written down in half-hour increments yet.
Frequently asked questions
How long before I actually get hours back?
Expect a net loss in weeks one through three, since documenting and training cost time before they save any. Most owners see the crossover in week four to six and meaningful relief by day sixty. If you're still net negative at ninety days, the cards were probably written too broadly to follow.
I only have two employees. Is this worth it?
Yes, and arguably more so. With two people the business runs entirely on undocumented memory, so one resignation or one bad flu week takes the whole operation down. Small teams get the biggest return from the date-driven compliance handoff alone.
What if I hand something off and quality drops?
That is what the two-week 100% review is for — it catches drift before a customer does. If quality slips after you move to sampled review, tighten the sample rate on that one task rather than taking the whole function back.
What should stay on the owner's plate permanently?
Usually four to six things: signing legal and financial commitments, setting the pricing floor, hiring and firing, the relationships that genuinely trade on you personally, and defining the quality standard. Almost everything else transfers with a written procedure and a checkpoint.