Delivery · The Cascading Leak
One Job Ran 20 Minutes Long. Your Whole Afternoon Just Went With It.
In This Article
8:04am. The first job of the day should take forty minutes. It takes an hour — a part wasn't in the truck, or the customer had one more question, or the job was just a little more than it looked like from the driveway. Nothing went wrong. That's just what a real job does sometimes.
Except now it's 9:04, not 8:44, and your 9am customer is standing in their kitchen wondering where you are. By 3pm you're the fourth apology of the day, and the reason has nothing to do with how fast you worked.
That's the buffer leak: a schedule with no room in it for a job to be anything other than exactly what you estimated. And the fix isn't going faster. It's giving the day somewhere to put the twenty minutes you were always going to lose eventually.
The leak: zero-buffer scheduling
Most schedules are built on one quiet assumption: every job takes exactly as long as you said it would. Forty minutes means forty minutes, every time, for every customer, on every kind of day.
Real jobs don't work that way. There's variance built into the actual work — extra scope you find once you're in there, a customer who needs five more minutes of your time, a supply run, traffic between stops. None of that is a mistake. It's just what happens across enough appointments, often enough, that it should be planned for instead of hoped against.
A zero-buffer schedule doesn't plan for it. So the first job that runs long doesn't stay contained to that job — it gets inherited by the next one, and the next, and the next, until the last customer of the day is the one paying for a delay that started six hours earlier and had nothing to do with them.
What the cascade actually costs
There's no report that puts a dollar figure on this one, so it's easy to shrug off as "just a rough day." Try the math anyway — as an illustration, not a claim about your specific numbers:
Imagine five jobs booked back-to-back, forty-five minutes apart, no room between them. The first one runs twenty minutes long — completely normal. That twenty minutes doesn't disappear; it rides along into every job behind it. Add just a few more normal minutes on job two or three — one more question, one more spot you can't find to park — and by job three, you're thirty-five minutes behind. By job five, the customer's been waiting the better part of an hour, for a delay that has nothing to do with anything you did to them personally. Some of them will be gracious about it. Some will quietly decide not to book you again — not because the work was bad, but because the day felt disorganized before you ever picked up a tool.
That second group is the expensive one. They rarely complain; they just stop calling. And replacing a customer who quietly leaves costs far more than keeping the one you already had — acquiring a new customer runs 5 to 25 times more expensive than retaining an existing one (Harvard Business Review, 2014, citing Bain & Company/Reichheld research). A chronically late schedule doesn't just cost you goodwill on a Tuesday. It's a slow leak in the customers you already spent real money to win.
Why it happens
The instinct, when a day starts sliding, is to work faster — rush the current job, shave corners on the next one, try to "catch up." It feels productive. It isn't. Working faster doesn't remove the variance that caused the problem; it just trades quality for a few saved minutes, and it still breaks completely the moment a job legitimately needs more time than you're willing to give it.
The actual problem is structural, not personal. Nobody built a place in the schedule for a normal job to be a little abnormal. The fix isn't hustle — it's a deliberate gap whose entire job is to absorb that variance before it reaches your next customer.
The Buffer Block — the fix
Three moves, and none of them require new software.
Size it from real data, not a guess
Pull your last ten job days. For each one, check how often the first appointment ran past its estimate, and by how much. That number — not a round one you picked because it sounded reasonable — is your real buffer. For most local service jobs, it lands somewhere between 10 and 20 minutes, but let your own history set it.
Put it between every appointment, not just at the end of the day
A buffer stacked at 5pm doesn't stop your 11am customer from waiting. It has to sit between every single job, because the overrun that breaks the day can happen on job one just as easily as job four.
Protect it like it's already booked
The moment a buffer becomes "room to squeeze in one more quick one," it stops being a buffer. It's not extra capacity waiting to be filled — it's the slot on your calendar with the most important job of the day: keeping every other appointment honest.
Pull your last ten job days. Check how often job #1 ran long, and by how much. If you don't have a number, or the answer is "pretty often" — that's your buffer leak, and now you know roughly how big it is.
The rule that protects it
One rule matters more than the exact number of minutes: the buffer is never "found time." The instant it becomes the place you squeeze in a favor, a walk-in, or "just five more minutes" with a chatty customer, it can't do the job it exists for — and the cascade is back, just starting from a different job.
Treat it the way you'd treat a paying customer's slot: it's on the calendar, it's spoken for, and nothing goes in it except the overflow it was built to catch.
Make it automatic
You don't need new tools for this — you need one new rule inside the tool you already use. Almost every scheduling or booking system lets you set a fixed gap between appointments. Set it once, sized from your own data, and every future day gets built with the room already in it. Nobody has to remember to leave space by hand; the calendar just won't let a job get booked without it.
Picture the version of the day where the first job runs long and nothing downstream even notices — the buffer quietly absorbed it, and your 11am customer never knew there was ever a problem. Now picture the version most local businesses actually run: no room anywhere, one normal overrun at 8am, and five apologies by 3pm for a delay none of those customers caused and none of them will forget. The first version isn't a faster business. It's a business with somewhere to put a completely normal bad twenty minutes.
You can build the Buffer Block yourself this week with the three moves above. If you'd rather we look at your actual schedule and tell you exactly where the leak is and how big it's really running, that's exactly the kind of thing we find in a free report.
Frequently Asked Questions
Enough to absorb what actually happens on a normal day, not an ideal one — usually 10 to 20 minutes, depending on the job. Don't guess: pull your last ten job days and check how often the first appointment ran long. That number, not a round one you picked out of the air, is your real buffer.
Because the schedule, not your speed, is the problem. A calendar with zero room between jobs assumes every appointment takes exactly its estimated time. The first one that doesn't — a part that's not in the truck, a question that takes five extra minutes — pushes every appointment after it. Working faster just trades quality for a few minutes; it doesn't remove the gap that's actually missing.
Build a deliberate buffer between every appointment — not just at the end of the day — and protect it like it's already booked. The buffer's whole job is to absorb one job running long before it touches the next customer's time. The moment you let it get filled with "one more quick one," it stops doing that job.
Yes, because it's solving the actual mechanism: lateness compounds when nothing absorbs the first delay. A buffer stops the compounding at job one instead of letting it ride through the rest of the day. It won't stop a job from running long — it stops that overrun from becoming everyone else's problem too.
You're not losing hours — you're trading a slot you were already going to lose to lateness for one you control. A day with five jobs and no buffer that runs an hour behind by 3pm isn't more productive than a day with four jobs that all start on time; it's just more apologizing. Size the buffer from real data, build it into every gap, and most businesses find they're not actually giving up a bookable slot — they're giving up the illusion of one.
