The Owner's Personal Operating Manual
Why this matters
Everyone who works for you has already built a model of how you operate. They built it from guesswork, from the last time you snapped at someone, and from a rule of thumb about which hours you are approachable. It is roughly half right, it took them months to assemble, and every new hire starts over from nothing.
The cost lands in two places. Things that should reach you fast do not, because someone guessed wrong about whether it qualified. And things that should never have reached you arrive anyway, because asking you is cheaper for them than being wrong. A written manual replaces the guessing with a specification. It is not a document about self-improvement. It is an interface spec for the people who have to work through you.
What this is, and what it is not
It is team-facing. Everything in it is written for someone else to act on. Your own system for structuring a week, protecting planning time and managing your energy is a separate thing that lives with you, and mixing the two produces a document that is uncomfortable to share and useless to read.
It is operational, not confessional. "I can be short with people when I am behind" is a confession. "If I answer in one word, I am behind, and it is not about your question - ask again after lunch" is operational, because it tells someone what to do.
It is short. One page, read in under three minutes, or it is not read. Roughly 15 statements is the working size, and each one has to be worth the space.
The seven sections
1. How to reach me, ranked, with a response time on each channel. This is the section with the most value per line and it is the one owners hedge. Commit to numbers. A workable default set: phone call for anything that has stopped work, answered inside 15 minutes during working hours; text for same-day matters, answered by end of day; email for anything that can wait, answered inside 2 working days. Tune those to your shop, but publish numbers, because "as soon as I can" gives nobody a basis for deciding whether to wait or escalate.
2. What I want to know immediately, and what waits. List the immediate ones explicitly and keep the list to five or six items - an injury, a customer threatening to walk, damage to a property, a crew member who cannot work today, a supplier failure that stops a job. Everything not on that list waits for the daily huddle or the weekly. The value is in what is left off, so being generous with the list defeats it.
3. How to bring me a decision. Name the format. What you need before you ask: what the options are, what you would do, what it costs to be wrong. A person who arrives with those three has a two-minute conversation. A person who arrives with "what do you want to do about this" has a twenty-minute one, and you have taught them that arriving unprepared is efficient for them.
4. What I decide fast and what I sit on. People read a delay as disapproval when it is usually just how you process a particular class of thing. Saying "anything involving hiring takes me a few days and that is not a bad sign" removes an entire category of anxiety and stops the follow-up asking.
5. What I am reliably bad at, and what to do about it. Not a list of flaws. A list of failure modes with a workaround attached. "I say yes too quickly when someone catches me between calls, so if you get a yes from me in the yard, put it in writing before you act on it."
6. How to disagree with me, and how to tell if I have changed my mind. Most people in a small shop have no idea whether pushing back is welcome. Say it plainly, and give the tell: what does it look like when you have actually reconsidered versus when you have gone quiet to end the conversation.
7. What happens when I am unreachable. Who decides, up to what ceiling, and what waits. This is the section that gets skipped and it is the one that determines whether a week away works.
The rule that makes it credible
Every statement in the manual is a promise that can be measured, and every response commitment carries a fallback for when you break it.
The fallback is the part that makes the whole document safe to rely on. "Phone answered inside 15 minutes during working hours" is a promise you will miss sometimes, and without a fallback, the person who missed you is stuck holding a stopped job while wondering whether to try again. With a fallback - "if I have not answered in 15 minutes, call the field lead and decide it between you, and tell me after" - the missed promise costs nothing, because the manual already routed around it.
A manual without fallbacks is more dangerous than no manual, because it concentrates dependency on you and then removes the improvisation people were using to cope with your absence.
An owner's manual, four sections filled
This is a real shape, not a template to copy word for word. The value is in the specificity.
How to reach me. Call me for anything that has stopped work. I answer inside 15 minutes between 7 and 5. If I have not answered in 15 minutes, call the field lead, decide it between you, and tell me after - I will not second-guess a call made in that window. Text me for same-day things, I answer by end of day. Email me for anything that can wait, and expect 2 working days. I do not read email after 6 or on Sundays, so anything sent then starts on Monday.
Tell me immediately. Anyone hurt, on the job or off. A customer saying they want someone else, or saying they are not paying. Property damage of any size. A crew member who cannot work today. A supplier failure that stops a job today. That is the list. Everything else goes in the morning huddle, and if you are unsure whether it belongs on the list, it does not, but tell me at the huddle that you nearly called.
Bringing me a decision. Come with three things: the options you see, the one you would pick, and what it costs if you are wrong. If you have those, we will be done in two minutes and I will usually go with yours. If you do not have them yet, that is fine, but say so at the start so I know we are thinking rather than deciding.
What I am bad at. I say yes too fast when you catch me between calls. If you get a yes from me in the yard or in a doorway, put it in writing before you act on it, and I will confirm or correct it the same day - that is on me, not you. I also go quiet when I disagree instead of arguing. If I go quiet, ask me directly what I do not like about it. I will tell you. I just will not volunteer it.
Notice what those four have in common. Each one hands the reader an action, and three of the four hand them an action for the case where the owner fails. That is what makes a manual load-bearing rather than decorative.
Keeping it honest
The manual is written by the person with the least accurate view of the subject. Every owner writes a slightly better version of themselves, not out of vanity but because you remember your intentions and other people only see your behaviour.
So the manual gets reviewed before it is published. Give the draft to two people who work closely with you - typically the office lead and the field lead - and ask them to mark every statement as true, sometimes, or not true. Not to comment. To mark. Commenting produces diplomacy, marking produces data.
The threshold: on a manual of about 15 statements, if 2 or more distinct statements come back marked "not true" by either reviewer, the manual is a wish list and gets rewritten before it goes out. Distinct is the operative word - two reviewers marking the same statement is one problem, not two.
One owner ran this on a 16-statement draft. The office lead marked 3 as not true, the field lead marked 4, and 2 of those overlapped, so 5 distinct statements failed. Well over the line of 2.
He fixed them in two different ways, and the split is the interesting part.
Three were fixed by changing the statement. He had written that he answers texts within two hours; both reviewers said end of day was the honest number, so end of day is what the manual now says. He had written that he is happy to be interrupted for customer complaints at any time; the field lead marked it not true and was right, because the last three interruptions of that kind had gone badly at exactly the wrong hour. It now reads: call me about a complaint any time before 4, after 4 text me the summary and I will call the customer first thing.
Two were fixed by changing his behaviour, because the statements were things he actually wanted to be true and was not doing. He had written that he tells people when he has changed his mind. Both reviewers marked it not true. He kept the statement and started saying it out loud, because deleting it would have removed the one commitment in the manual that made pushing back on him worth attempting.
The general rule from that split: change the statement when it describes a preference you do not actually hold, and change the behaviour when it describes something you do hold and are failing at. Deleting a statement to make the review pass is how the document becomes a description of a person nobody needed a manual for.
Re-run the review once a year, or after any change that alters your availability - a new hire in the office, coming off the tools, a season that moves your hours.
The two ways this document fails
It becomes a shield. "It says in my manual that I do not take calls after five" used in the middle of an actual emergency turns a useful document into an excuse. The manual describes the default, and defaults exist so that exceptions are visible. If someone breaks one of these rules and they were right to, say so publicly, or you will not get the next one.
It is published and never enforced against yourself. A manual that promises a 15-minute phone response, from an owner who routinely takes two hours, is worse than silence: people stop trusting the specific claims, then stop trusting the document, then go back to guessing - except now they are guessing while assuming you think you are being responsive. If you cannot hold a number, publish the number you can hold. A manual full of modest, kept promises beats an ambitious one that is contradicted by Tuesday.
References
- U.S. Small Business Administration (SBA), small business communication and management practices
- Trade-standard practice, escalation and on-call protocols in small service shops
- See related: How to Document How You Work for Your Team, How to Stop Being the Answer to Every Question, Building a Personal Operating System, The Phone Boundary Worth Setting as an Owner, How to Take a Week Off Without the Shop Stopping