How Many Employees Should I Schedule Each Shift at My Cocktail Lounge?

This vs. the common alternatives

Most lounge owners arrive at a shift count one of four ways, and only one of them survives a bad week.

The habit schedule. "We always run six on Friday." It is fast, it is stable for the staff, and it is wrong roughly half the year. Habit schedules are calibrated to the busiest month you ever had, which means you carry three months of February at October headcount. The tell is a labor line that swings between 18% and 34% of sales with no change in the schedule — you didn't get worse at scheduling, the volume moved and the grid didn't.

The fixed-ratio schedule. One bartender per fifteen seats, one server per twenty. Ratios are a real improvement over habit because they at least respond to room size, but they respond to *capacity*, not *revenue*. A 90-seat lounge at 40% occupancy on a Tuesday still triggers the full-house ratio. Ratios also ignore check average entirely — the same fifteen guests drinking $17 stirred cocktails and $9 highballs produce wildly different revenue per body but identical staffing under a seat ratio.

The labor-percentage schedule. You set a target — say 24% of sales — and back into hours from your projected revenue and average wage. This is genuinely good math and it is what most POS-connected scheduling tools do natively. Its weakness is that it is a *whole-shift* number. Hitting 24% across a nine-hour Saturday tells you nothing about whether you were catastrophically thin from 10pm to midnight and three people deep at 5pm. Percentage targets average away exactly the volatility that ruins a lounge's guest experience.

Sales per labor hour (SPLH). You set a per-employee hourly revenue target, then divide each *daypart's* projected sales by it. This is the method that holds up, because it is granular enough to place people inside the night rather than across it, and it is denominated in the thing you actually manage — hours on the floor.

The practical difference is where the errors land. Habit over-staffs the slow half of the week and under-staffs the surge. Ratios over-staff big rooms. Labor percentage gets the total right and the shape wrong. SPLH gets both, provided you're honest about the target.

That target deserves scrutiny. A high-volume beer-and-shot bar might sit at $110–$150 SPLH because service is pour-and-go. A craft cocktail lounge where a single drink involves a fine strain, a fresh-pressed juice, and an expressed peel realistically lands lower — $55 to $75 per labor hour is a defensible working floor, and if your program leans heavily on stirred spirit-forward builds with high check averages you may sit at the top of that band or above it. What you should never do is inherit somebody else's number from a forum post. Derive it: take four strong recent shifts you'd call well-staffed, divide the sales by the total labor hours worked, and see what your own room produces when it's running right. That's your baseline.

The same logic ports across service businesses, which is why the method is worth learning rather than the tool. A car wash schedules attendants against cars per hour. A logistics shop schedules dispatchers against loads booked. A RevOps team sizes an SDR bench against pipeline coverage rather than headcount tradition. The unit of demand changes; the division doesn't.

How to choose between them

Choosing a scheduling method is really choosing how much data discipline you're willing to carry. Here is the honest decision path.

Work the branches in order.

No history? You cannot compute what you haven't measured. Spend a quarter running conservative ratios while logging two columns per hour: sales rung and bodies on the clock. Ninety days is the minimum for a lounge because you need at least one slow stretch and one holiday surge in the sample. Six months is better if you're seasonal — a rooftop lounge's June and its January are effectively two different businesses and averaging them produces a schedule that's wrong in both.

Unstable check average? If your 5pm aperitivo crowd drinks $14 spritzes and your midnight crowd drinks $19 stirred classics, one blended SPLH target will systematically over-staff the early window and under-staff the late one. Split it: run separate targets per daypart. Most lounges need two or three — early, prime, late.

The minimum-viable crew floor is non-negotiable. The math will occasionally tell you a Tuesday at 6pm needs 1.4 people. You cannot run a lounge with 1.4 people. You need someone behind the bar, someone who can leave the bar to run a card or check an ID, and a plan for what happens when both are occupied and a four-top walks in. For most rooms the practical floor is two, and three once you're serving food or running a door. Set that floor explicitly and accept that on dead shifts your labor percentage will look ugly — that's the cost of being open, and the real decision hiding behind an ugly Tuesday number is usually "should we open at all on Tuesday," not "can we cut the barback."

Then place the hours inside the night. The count answers *how many*; the hourly sales curve answers *when*. Pull transaction timestamps, not just daily totals. A lounge whose tabs cluster from 9pm to 1am should open light — one bartender, one server — ramp a second bartender and a barback in at 8:30, add floor coverage by 9:30, and stagger the outs so you're not paying four people to watch two guests close out at 1:45. Staggered starts are where most of the savings actually live; a schedule with the right headcount and flat 5-to-close shifts still bleeds money at both ends.

One more filter: how much execution machinery do you need on top of the math? If your crew is six people and everyone reads the group chat, a shared sheet and a calculator are enough. Once you're past a dozen part-timers with availability constraints, swap requests, and split shifts, the administrative overhead of publishing and enforcing the schedule outgrows the math itself, and that's the point where dedicated scheduling software earns its price — not because it computes better, but because it stops the phone calls.

Costs, timelines, and expected impact

Let's put numbers on what this exercise actually costs you and what it returns.

Time to first schedule. If your POS exports hourly sales, building the first daypart-level model takes two to four hours of focused work — pull the export, bucket by day of week and hour block, average the trailing quarter, divide. If you're reconstructing from paper tickets or a POS that only gives daily totals, budget a weekend, or start logging forward and accept that your first real model lands in ninety days.

Ongoing maintenance. Fifteen to thirty minutes a week to re-run the division with the newest week folded in, plus a bigger reset quarterly when seasonality shifts. The trap is building a beautiful model once and never touching it. A model calibrated in March and still running in July is just a habit schedule with extra steps.

Software. Free tiers exist and are genuinely usable for a single room — several major hospitality scheduling platforms offer no-cost plans covering one location. Paid tiers generally split into two pricing shapes, and the shape matters more than the sticker: per-location pricing (a flat monthly fee regardless of roster size) versus per-user pricing (a smaller per-head fee). For a lounge running eighteen part-time bartenders and servers, per-location almost always wins. For a lean crew of six on stable full-time-ish schedules, per-user can be cheaper. Run the arithmetic on your actual roster before you sign anything; the same platform can be the cheap option or the expensive one depending entirely on how many names are on your list.

What you should expect to recover. Be skeptical of anyone promising a specific percentage. What's defensible is the mechanism: if your labor line currently swings 15+ points between your best and worst weeks, that swing is the recoverable territory. Over-staffed slow shifts are pure loss — you're paying wages against revenue that isn't there. Under-staffed peak shifts cost you differently and more quietly: longer ticket times, guests who leave before the second round, and a bar team that burns out and quits, which is the single most expensive line item in this entire discussion. A bartender who walks in month four costs you the recruiting time, the training weeks, the drop in drink consistency, and the regulars who followed them out.

Timeline to seeing it. Two weeks to feel it on the floor — staff notice immediately when the schedule matches the room. Four to six weeks before the labor percentage moves visibly, because you need a full cycle of slow and busy weeks to see the smoothing. One quarter before you can honestly say whether the target you picked was right.

Second-order effects worth budgeting for. Better scheduling changes your hiring math. A lounge that used to need eight people because two were always idle can often run seven who each get more hours — which makes the job better and retention easier. It also changes your inventory rhythm: if you know Thursday 9pm is your consistent peak, prep, batching, and ice production get scheduled against that rather than done reactively. And it changes what you can promise: a room that knows its own demand curve can commit to a private-event booking on a Tuesday without wondering whether it'll wreck the week.

The cost of getting it wrong in the tight direction. Under-staffing looks great on the labor line for exactly one month. Then ticket times stretch, the well gets disorganized because nobody has ninety seconds to reset it, quality drops, tips drop, and your two best people start answering recruiter texts. Cocktail lounges are unusually exposed here because the product is craft — a slammed bartender at a beer bar pours faster, but a slammed bartender at a cocktail lounge starts cutting the steps that make the drink worth $18. Guard the peak. If you're going to be wrong, be wrong heavy on Saturday at 10pm and lean on Tuesday at 6.

Implementation and handoff details

Building the model is the easy half. Handing it to a management team so it survives you is the half that fails.

Step one — write the target down and say it out loud. The SPLH number only works if the whole team knows it. Tell your crew plainly: "In this room, one person on the floor should comfortably move about sixty-five dollars an hour in sales without the ticket line backing up." That's a floor, not a ceiling — the people chasing real tips clear it and then sell the second round and the bar snacks. What the number actually buys you is an end to argument. Nobody negotiates for a friend's shift when the count comes off a division.

Step two — split the count into roles before you publish. Fifteen labor hours is not a schedule; it's a budget. Allocate it: two bartenders on the well, one barback covering ice, glass, and batch pulls, three cocktail servers on the floor, one door host during the surge. The barback slot is the one most often cut and most often wrong to cut — a barback keeps two bartenders at full speed, so cutting one $18/hr body to save money frequently costs you more in slowed drink output than it saves.

Step three — publish fourteen days out and mean it. Late schedules cost you good people. Many jurisdictions now have predictive-scheduling ordinances with real penalties for last-minute changes, and even where the law doesn't reach, a bartender who can't plan two weeks ahead takes the job that lets them. Fourteen days is also long enough that the model's projections get tested honestly rather than adjusted into being right after the fact.

Step four — build the variance loop, and give it an owner. Every week, put projected sales next to actual sales for each daypart. When a block runs more than 15% off in either direction for two consecutive weeks, that's a signal, not noise — the neighborhood changed, a competitor opened, a new menu landed, or your target was wrong. Recalibrate. Without this loop the model calcifies within a season. Name one person responsible for it: usually the bar manager, not the owner, because the manager sees the floor and the owner sees the spreadsheet.

Step five — the handoff document. If this method lives only in your head, it dies the week you take vacation. Write a one-page runbook: where the sales export lives, what the current SPLH target is and when it was last set, what the minimum crew floor is per daypart, who publishes, and what triggers a recalibration. Every operations discipline worth having has this document, whether you're staffing a lounge or a RevOps team — the model isn't the asset, the documented method is.

Step six — sanity-check against reality, weekly. Walk the floor at your projected peak. If the count says five and the room feels frantic with five, your target is too high for the shape of your business and you should lower it. If the count says five and two people are folding napkins, it's too low. The math should be argued with by the people standing in it; a schedule nobody on the floor believes in gets quietly undermined within a month.

What to do about events and outliers. A private buyout, a holiday, a neighborhood festival, the night the bar down the street closes for renovation — these break the model and should be handled outside it. Don't let a New Year's Eve into your trailing average; it will inflate every subsequent Tuesday. Flag outlier dates, exclude them from the baseline calculation, and staff them off a bespoke estimate. The same goes in the other direction for a blizzard or a power outage.

Handling the human side of a cut. When the model says a shift needs four and you've been running six, you are about to take hours from two people. Handle it directly: explain the method, show the numbers, and offer the hours back on the shifts where the model says you're thin. Most staff would rather work four busy shifts than six dead ones — tips track volume, not clock hours. The operators who lose their teams over this are the ones who cut silently and let people discover it on the posted schedule.

Related questions

How does this change if my lounge also serves food?

Food adds prep, runners, and expo, and it usually lowers your blended sales per labor hour because kitchen hours ring no sales directly. Track front-of-house SPLH separately from total labor, or the kitchen will make your bar look inefficient every night.

What if I'm opening a brand-new lounge with zero sales history?

Use conservative fixed ratios for the first ninety days and log hourly sales from night one. Staff slightly heavy during the opening month — service failures during a launch are far more expensive than a few extra labor hours, and word travels fast in a new room.

Should salaried managers count in the labor-hour math?

Count them for total labor cost, but exclude them from the SPLH divisor if they're primarily supervising rather than serving. A working manager who spends most of a shift behind the well should count as a body; one running admin in the office should not.

How do I handle bartenders with wildly different output?

Set the target off an average performer, not your best one. Then use the gap as a coaching tool rather than a scheduling one — schedule the count the math gives you, and pair a strong bartender with a developing one on peak shifts so the training happens under real pressure.

Does this method work for a multi-room or multi-location group?

Yes, but calibrate each room separately. A rooftop, a basement speakeasy, and a neighborhood cocktail bar in the same group will have genuinely different SPLH targets. Roll them up for reporting; never average them into one shared target.

FAQ

What is the best way to determine how many employees I need for a shift?

Divide that shift's projected sales by a pre-agreed sales-per-labor-hour target. For a craft cocktail lounge, $55–$75 per labor hour is a reasonable starting band because drinks take real time to build. Then apply a minimum crew floor so the math never leaves you with an unworkable skeleton crew.

How do I calculate my own sales-per-labor-hour target instead of borrowing one?

Take four recent shifts you'd honestly call well-staffed — the room ran smoothly, ticket times held, nobody was folding napkins. Divide each shift's sales by the total labor hours worked. Average the four. That number is your room's real capability, calibrated to your menu and service style.

What if my sales swing hard from day to day?

That's exactly the case this method handles. Pull three to six months of sales split by day of week and daypart, then run the division separately for each block. A Saturday late window and a Monday early window are different businesses that happen to share an address; give each its own number.

How far in advance should I publish the schedule?

Fourteen days is the practical standard. Some jurisdictions legally require advance notice with penalties for last-minute changes, and independent of law, a schedule staff can plan around is one of the cheapest retention tools you have in a business where replacing a trained bartender is genuinely expensive.

Do I need scheduling software to do this, or is a spreadsheet enough?

A spreadsheet handles the math fine for a single room with a small crew. Software earns its cost once roster administration — availability, swaps, time clock, labor-law compliance — outgrows the calculation itself. Buy for the administrative burden, not the arithmetic.

How often should I revisit the target once it's set?

Check variance weekly, recalibrate quarterly, and reset immediately after any real change: a new menu, a price increase, a service-model shift, or a new competitor on the block. A target set once and never revisited becomes a habit schedule wearing a spreadsheet costume.

Sources

• U.S. Bureau of Labor Statistics, Occupational Outlook Handbook — Bartenders: https://www.bls.gov/ooh/food-preparation-and-serving/bartenders.htm
• U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics: https://www.bls.gov/oes/
• U.S. Department of Labor, Wage and Hour Division — Tipped Employees: https://www.dol.gov/agencies/whd/flsa/tips
• National Restaurant Association — industry research and operations resources: https://restaurant.org/research-and-media/research/
• U.S. Small Business Administration — Manage your business: https://www.sba.gov/business-guide/manage-your-business
• SHRM — scheduling and workforce management resources: https://www.shrm.org/topics-tools/topics/talent-acquisition
• IRS — Tip Recordkeeping and Reporting: https://www.irs.gov/businesses/small-businesses-self-employed/tip-recordkeeping-and-reporting
• Cornell University School of Hotel Administration — Center for Hospitality Research: https://sha.cornell.edu/faculty-research/centers-institutes/chr/

Related on PULSE

• [How Many Attendants Should I Schedule Each Day at My Car Wash?](/knowledge/tl0067)
• [How Many Sales Reps Do I Need to Hire for My Logistics Company?](/knowledge/tl0058)
• [How Many Salespeople Do I Need to Hire for My Car Dealership?](/knowledge/tl0052)
• [How Many Producers Do I Need to Hire for My Insurance Agency to Grow My Book?](/knowledge/tl0015)
• [How Do I Figure Out How Many People to Schedule Each Day and at What Times for My Single Store?](/knowledge/tl0002)