If you want to know whether a front desk is run well, do not watch it at noon. Watch it at six. Noon flatters everyone. The lobby is quiet, the phones are slow, and even a thin team looks competent. Six in the evening is the truth. The flights have landed, the meetings have let out, and everyone checking in that day seems to want their key in the same twenty minutes. A desk that is calm at six was staffed on purpose. A desk that is drowning at six was almost always staffed by the clock instead of the wave.
That distinction is the whole subject of this piece. Most desks are covered with what I call flat coverage, the same number of agents on the floor from open to close, evenly spread, because it looks fair and it is simple to write. It is also, quietly, one of the most common ways to run a front office badly. Good front desk staffing does not spread people evenly across the day. It matches them to the shape of the day, and the shape of the day is a curve.
What is the arrival curve?
The arrival curve is just the honest picture of when guests actually check in. It is not a straight line and it is not a mystery. On most properties it is low in the morning, because the people in those rooms are leaving, not arriving. It dips into a genuine lull around the middle of the day. Then it climbs, gently at first and then steeply, from midafternoon into the early evening, and it usually crests somewhere in the dinner hours before it tails off late.
The core idea deserves a plain definition.
Once you can see the curve, flat coverage stops looking fair and starts looking like a fight with the shape of the day.
The exact shape shifts with the building. A convention hotel spikes when a group's room block opens. An airport property moves with flight banks. A resort in Waikiki, where I ran night audit, has a different rhythm than a downtown business hotel in San Francisco, where I lead rooms operations now. But the underlying truth holds everywhere I have worked: arrivals are not evenly distributed, so staffing that is evenly distributed is fighting the shape of its own day.
Here is the part people underrate. That midday lull is not dead time to be minimized. It is prep time, and it is the most valuable stretch on the floor. What you do in the quiet is what makes the peak survivable. So the curve is not only telling you when to add bodies. It is telling you when to take them off the counter and put them to work on the afternoon.
Why is flat coverage so tempting and so wrong?
Flat coverage survives because it is easy to defend. Everyone works a normal shift, the schedule is symmetric, and nobody has to explain why one agent starts at two in the afternoon. It feels equitable. The problem is that guests did not agree to arrive equitably.
When you spread coverage evenly across a curve that is anything but even, you make two mistakes at once and pay for both. In the morning and at midday you are overstaffed. You have agents standing at an empty counter, which reads as expensive to anyone watching labor and, worse, wastes the exact hours when that labor could be preparing the building. Then in the evening you are understaffed against the actual volume, so the line backs toward the door, checkin quality drops because everyone is rushing, and the guests who waited longest are the ones arriving tired at the end of a travel day. You have managed to spend too much and deliver too little in the same shift.
A flat schedule is not neutral. It quietly overpays the quiet hours and underfunds the only hour that matters.
The fix is not more people. Often it is the same people, placed differently. I would rather have a lean midday and a deep evening than a comfortable, uniform staff that is comfortable at exactly the wrong time. The goal is to shape the labor to the demand, so the counter is thin when the lobby is empty and deep when the wave hits.
Flat coverage versus curve coverage, side by side
Put the two schedules next to each other and the difference is stark. Say you have the same pool of agent hours to spend on a two hundred arrival day. Flat coverage parks, for example, three agents on the counter from open to close, every hour looking like every other. At ten in the morning that is two agents too many; at six in the evening it is two agents too few. The hours were real, they were just standing in the wrong place. Curve coverage takes the same total hours and stacks them differently: one or two agents through the slow morning and lull, then four or five overlapping across the evening crest, then back down as it tails off. Nobody worked more hours. The building simply got its help when the guests actually arrived. Same cost, opposite outcome, and the only variable that changed was when people were standing at the counter.
The same pool of hours, drawn two ways, makes the point at a glance.
Nobody worked more hours; the curve schedule simply moved them off the empty noon counter and onto the six o'clock crest.
How do you actually build the schedule?
It starts where every front office day starts, with the arrivals report. That report is my forecast. Before I schedule a single agent I want to know how many rooms I owe, and roughly when those guests are coming, because that is the raw material of the curve for the specific day I am staffing. A Tuesday with two hundred arrivals stacked into a convention block is a different building than a Tuesday with the same count spread across the afternoon.
The specific building matters here too, and it changes the curve in ways a generic template never captures. A property that runs heavy on groups spikes when a block opens and goes quiet between blocks. A property that runs on transient business travel builds a smoother, later evening. Weekends do not look like weekdays, and the night before a big citywide event does not look like an ordinary Tuesday. I do not schedule to an average. I schedule to the day in front of me, because averaging a lumpy month into a smooth week is just a slower way of arriving at flat coverage. The arrivals report is what keeps me honest about which kind of day I am actually staffing.
The schedule in five steps
From there I work in a rough sequence, and it is worth spelling out as steps because the order is what keeps the plan honest.
- Pull the arrivals and shape the curve. Read how many checkins the day owes and when they cluster. This is the forecast everything else hangs on.
- Set a realistic checkin time. A warm, complete arrival is not instant. I refuse to schedule as if it were, because pretending checkins take two minutes is how you end up understaffed at the crest.
- Find the compression. Look at how much of the volume stacks into the peak window. That compression, not the daily total, is what decides how many agents I need at the counter at once.
- Stagger the starts. Bring agents online as the curve rises rather than all at once. Some open early for the departure push, a middle wave lands ahead of the afternoon build, and the heaviest coverage overlaps the evening crest.
- Protect the top with a little slack. Overlap the peak slightly more than the raw math demands, so one callout or one early group does not fold the plan.
The schedule, drawn as a line, should look a lot like the arrival curve itself, only shifted a little earlier so the help arrives before the guests do. That shift is the quiet trick of the whole craft: you are not staffing the wave, you are staffing the moment just before it.
What the good schedule protects
- The peak, above everything. If I have to be thin somewhere, it is at noon, never at six. The evening crest is the shift, and it gets the deepest bench I can give it.
- The prep window. The midday lull is when the hard rooms get pre-blocked and the VIPs get verified. That work is scheduled, not hoped for.
- A little slack at the top. A peak staffed to the exact minimum has no give. One callout or one early group and the whole plan folds.
- Quality, not just throughput. Enough hands at peak so an arrival still feels like a welcome, not a transaction under a line.
The lull is a workshop, not a break
The part of curve based staffing that gets missed most often is what the middle of the day is actually for. A flat schedule treats noon as ordinary coverage, the same as any other hour. A tired desk treats it as a break, a chance to breathe before the evening. Both are wrong. The lull is the most productive stretch on the floor, and the schedule should be built to exploit it, not to sit through it.
When the lobby is empty, I want most of my hands off the counter and into the afternoon. This is when the hard rooms get pre-blocked, the connecting families and accessible rooms and VIPs sorted while there is still time and options. It is when a newer agent gets walked through a folio question by someone who is not rushing. It is when a couple of arriving rooms get physically verified, so nobody discovers at four o'clock that a room the system called ready is not. Every one of those tasks is invisible to guests and decisive for the evening, and every one of them needs people who are not standing at a counter serving a line.
So the curve is really telling me two things at once. It tells me when to add bodies to the counter, and it tells me when to pull bodies off the counter and put them to work. A schedule that only staffs the counter, even a curve shaped one, is leaving the prep window empty. The best schedules keep a thin service presence through the lull and route everyone else into preparation, because the calm evening I want is manufactured in exactly those quiet midday hours that look, from the outside, like the desk has nothing to do.
The peak is where you either have slack or you do not
The single most common scheduling mistake I see is staffing the peak to the perfect number. On a perfect day, the perfect number works beautifully. There are no perfect days. Someone calls out. A flight bank lands late and dumps forty arrivals into a window you thought was tapering. A group's block opens an hour early. If you built the evening to the exact minimum, every one of those ordinary surprises becomes a line at the door, because you left yourself no room to absorb them.
So I build a small amount of intentional slack into the top of the curve. Not a lavish overstaff, just enough give that the shift can take a punch and keep standing. That slack comes from a few places. It comes from overlapping the evening coverage a little more heavily than the raw math demands. It comes from having a manager or lead who can step onto the counter when the line grows instead of watching it from the side. And it comes from the bench, which is why I care so much about reading the lobby early enough to pull that slack forward before the line ever forms.
The bench matters as much as the schedule. A front desk where only the scheduled agents can work the desk is fragile by design. When I have cross-trained people, someone from reservations or a supervisor who still knows the counter cold, I have options at peak that a rigid roster does not. Staffing to the curve is not only about the shape of the schedule. It is about how much the schedule can flex when the day refuses to match the plan, which it regularly does.
Staffing is a service decision, not just a labor decision
It is easy to frame all of this as a cost exercise, and labor is real, so part of it is. But I do not think of the arrival curve primarily as a way to save hours. I think of it as a way to protect the guest experience at the exact moment the guest is most likely to judge the hotel. The evening arrival is often a guest's first real contact with the building. They are tired, they have traveled, and the length of the line and the warmth of the welcome set the tone for everything that follows. This is where the whole front office gets tested, at the counter, at the crest of the curve.
Flat coverage sacrifices that moment to make the schedule tidy. Curve based coverage protects it, and it usually does so without spending a dollar more, because the hours were always there. They were just parked at noon when they should have been standing at six. That is the entire craft of it. You do not need more people than the day requires. You need them present in the shape the day requires, thin in the quiet, deep at the crest, and always with a little more give at the top than the math alone would tell you to keep.