The short answer: real weekly capacity is not the hours in your week. Start from the hours you can actually be on the tools, subtract travel, quoting, admin and the jobs already committed, and what is left is what you can honestly promise. Overbooking damages more reputations than poor workmanship, because a missed date is visible to everyone.
"We can squeeze it in."
Five words, said on the phone in a good mood. Three weeks later they have turned into this: you are texting one client at 7 a.m. to push their start date again, finishing another job in the dark, and driving between two half-done sites in the same day because neither could be told no. Nobody got bad work. Everybody got a bad experience.
Here is the part contractors underestimate: overbooking damages your reputation more efficiently than bad workmanship does. A flawed tile line gets fixed and forgotten. A blown start date gets retold at every dinner party: "he's good, but good luck getting him to show up." The market forgives imperfect work far more readily than it forgives unreliability, and unreliability is rarely a character flaw. It is an arithmetic error, made weekly, by contractors who never computed their real capacity.
How many hours a week can you actually work?
The root error is simple: you plan as if a week contains 40 (or 50) hours of production. It never has.
Track an honest week and it comes apart like this for a typical solo contractor:
- Driving: between sites, to the supplier, back to the supplier for the thing that wasn't in stock (6 to 8 hours)
- Quoting and site visits: the unpaid work that fills next month (3 to 5 hours)
- Supply runs and staging: 2 to 4 hours
- Admin: invoices, texts, ordering, scheduling (3 to 4 hours)
- Slippage: the client conversation that ran long, the inspection window, the rain
Out of a 45-hour working week, the hours that actually move a job forward (tool-in-hand, on-site production) land somewhere between 26 and 30 for most solo operators. Crews do somewhat better per person, but the overhead never disappears; someone still quotes, someone still drives.
This is the number that matters: site hours, not paid hours, not working hours. If you schedule 40 hours of production into a week that physically contains 28, you are not ambitious. You are pre-booking apologies.
The drive cost calculator shows what the drive to a job costs you, per job and across a year.
How do you calculate your real weekly capacity?
Do this exercise one time and your scheduling changes permanently:
- Track two normal weeks. Every block: production, driving, quoting, supply, admin. Phone notes are fine. Do not track an unusually good week; track a real one.
- Take the average weekly production hours. Suppose it comes out at 28.
- Subtract a buffer, before you feel like you need one. Hold back 15–20%. Call it 5 hours. This is not slack. It is the landing area for everything you cannot foresee but can absolutely predict: the discovery behind the drywall, the failed inspection, the client who adds "one small thing," the flu.
- The remainder is what you sell. In this example: 23 hours a week of committed production.
Twenty-three hours sounds shockingly low the first time you compute it. It is also why your last six months of scheduling felt like a rolling emergency. You were selling 35 against a true 23, and the difference was paid in pushed dates, night work, and clients who now describe you with the word "but."
Buffer discipline: the hardest easy thing
Everyone agrees with buffers in theory. In practice the buffer is the first thing sold, because it looks like free money: an empty Thursday afternoon three weeks out, and a client on the phone right now who wants it.
Three rules keep the buffer real:
The buffer is spent on existing commitments, never new ones. Its whole purpose is to absorb overruns on jobs already promised. The moment it absorbs a new booking, every downstream date inherits the risk.
Rebuild it weekly. When Tuesday's discovery eats Thursday's buffer, next week's schedule gets adjusted (one job slides now, deliberately, with a phone call) instead of five jobs sliding later, chaotically, with excuses.
A job is not "a day." It is its hours plus its tail. A "one-day" bathroom vanity swap is six site hours plus a supply run plus the return trip when the tap set is wrong. Book jobs by their honest hours including the tail, and days stop mysteriously overflowing.
The deeper shift is emotional, not mathematical: an empty slot in your calendar is not waste. It is the asset that makes every promise on either side of it true.
Saying "three weeks out" like it's good news
Contractors do not overbook out of greed. They overbook out of fear of the pause after "I can't start until the 24th." It feels like watching the job walk away.
But watch what actually happens when a booked-out contractor says it with confidence:
"Right now I'm booking about three weeks out: I protect my start dates, so when I give you one, we actually start that day. I can put you in for the 24th and confirm materials the week before. Does that work?"
Two things land in the client's mind. First: this person is in demand, which reads as quality. Nobody wants the contractor who can start tomorrow because nobody else called. Second: this person keeps dates, which is precisely the thing every client has been burned on before. You have converted the wait from a cost into evidence.
Some clients genuinely cannot wait, and they leave. Let them. Those were exactly the jobs that would have been squeezed into hours you did not have, becoming the next round of apologies. A waitlist you honor is worth more than a calendar you don't.
One caution: "three weeks out" must be true. Quoting long lead times while quietly slipping favorites in early, or padding dates so much that you sit idle, both corrode the system. The confidence trick only works because it isn't a trick.
Scheduling multi-day jobs without the domino run
Solo and small-crew scheduling fails in a specific pattern: jobs are booked back-to-back-to-back, so one overrun topples every start date behind it. Clients experience this as "he keeps pushing me," when the real cause was a decision made a month earlier.
Structural fixes:
Never butt two multi-day jobs together. A day of air between them costs you nothing if job one runs true (fill it with quotes, small work, or the buffer's debts) and saves the entire downstream schedule when job one runs long. Which, regularly, it will.
Keep one flex day a week for small jobs and returns. The two-hour repairs, the deficiency fixes, the "while you're nearby." Give them a home instead of letting them puncture production days at random.
Confirm start dates a week out, with materials in hand. Half of blown starts are actually supply problems wearing a scheduling costume. If the tub is not physically in your possession or confirmed on a truck, the start date is a hope, not a date, and the client deserves to know which they've been given.
Sequence trades and inspections with the buffer in mind. Anything that involves waiting on someone else (an inspector, a countertop template, another trade) is a hard boundary. Schedule your own work up to it, not through it.
Seeing your committed capacity
All of this is easier when you can actually see the week you have promised, rather than reconstructing it from memory and text threads. This is what a schedule tool has to earn its place doing. Zeus's week view shows every job and appointment against the days they occupy, and the Work Board's multi-day Gantt Planner lays committed jobs across the coming weeks. So when the phone rings, you are looking at your true remaining capacity, not guessing at it in a good mood. The clock-in timesheets close the loop, showing what jobs actually took versus what you booked, which is how your capacity number gets more accurate every month instead of staying a hopeful fiction.
The tool does not create the discipline. It just makes the honest answer visible at the exact moment you are tempted to say "we can squeeze it in."
The compounding payoff
Run at true capacity for six months and the effects stack up quietly. Start dates hold, so clients start planning around you, and telling people you show up when you say you will, which in this industry is close to a superpower. Jobs stop overlapping, so quality stops competing with hurry. Your evenings return. And your pricing improves, because a contractor with a protected three-week waitlist negotiates from an entirely different chair than one with a guilty calendar and a phone full of apologies.
You will still have chaotic weeks; the trade guarantees it. But there is a difference between a chaotic week landing on a schedule with margins and one landing on a schedule already stretched past physics. The first is a bad week. The second is your reputation, leaking.
Where Zeus fits
A capacity number is worth only as much as the record behind it, and that record has to be made on site, in the hour it happens, not reconstructed on a Sunday from memory.
Clock in when you reach the address and out when you leave, and each job carries the hours it actually took. Put the week's jobs and appointments on the schedule and the planner tells you when two of them land on the same person on the same day, before a client hears about it. Costs go on the job that caused them, so the three-day bathroom that ran four shows the hours and the money it took beside what you quoted. A month of that is the honest denominator for the arithmetic earlier in this article: site hours, not working hours. The stops, the week and the board they come off are the day side of it.
There is more on the features page than the schedule. It costs nothing to start, there is no card, and it does not expire, and each size is set out here. Put it on your phone and let two weeks of clocked hours set your real number.
"We can squeeze it in" is a sentence you say when you cannot see the week. Said while looking at the hours already promised and the ones actually left, it turns into what it should have been: sometimes yes, more often the 24th.
Frequently asked questions
How much buffer should I actually keep?
Start at 15–20% of your production hours and adjust from evidence. If you end most weeks with untouched buffer, trim to 10%. If jobs still overrun and dates still slide, push to 25%. Or notice that the real problem is estimating, not scheduling: jobs booked at their hoped-for hours instead of their historical ones will sink any buffer.
Doesn't turning work away just feed my competitors?
The work you turn away at true capacity is work you would have delivered late and stressed. That is the version of you that generates the bad reviews. A competitor doing that job adequately harms you less than you doing it badly. And a genuine waitlist raises your perceived value; scarcity reads as quality in the trades, provided the dates you do give hold.
What do I do when a great client begs me to squeeze them in?
Offer what you actually have, not what they are asking for. That might be the flex day for a small version of the work, the next real opening, or a referral to someone you trust. Squeezing them in by breaking another client's date just transfers the disappointment to someone who booked properly. It also teaches your best clients that your dates are negotiable, which is the most expensive lesson you can teach them.
How do I dig out if I'm already overbooked?
One honest round of calls now beats three months of rolling slippage. Re-sequence everything against your true capacity, call every affected client with a firm new date, and hold those dates absolutely. Clients are far more forgiving of one straight conversation than of serial small surprises. The discomfort of that afternoon of calls is the tuition for never scheduling this way again.
ZEUS