How Many Employees Should I Schedule Each Shift at My Watch and Jewelry Repair Shop?

Signals you actually need this

Most watch and jewelry repair owners do not schedule; they inherit a schedule. Three people on weekdays, five on Saturday, because that is what the previous owner ran and nobody rebuilt it after the customer base changed. The tell is not that the schedule feels wrong — it is that nobody can explain *why* it is what it is. If you ask your store manager why Tuesday has four people and they say "that's just how we've always done it," you are staffing on inheritance, not on math.

Here are the specific signals that your shift counts have drifted away from your revenue:

Your labor percentage swings more than a few points week to week without a matching swing in sales. Pull twelve weeks of payroll dollars against twelve weeks of gross profit. If labor as a percentage of gross profit bounces from the low twenties one week to the mid thirties the next while sales stayed flat, the schedule is not tracking the money. A stable ratio means the plan is working even when volume moves; a jumpy ratio means you are staffing to habit and letting the ratio land wherever it lands.

Bench work orders age past your promise date on the same weekdays every month. If watch overhauls consistently slip when they enter the queue on Thursdays, that is a coverage problem disguised as a capacity problem. The bench is not too slow — the bench is understaffed on Thursday intake and the backlog compounds into Friday and Saturday.

Your counter has visible dead air. Walk the floor at 10:15 on a Tuesday. If two associates are reorganizing a case that does not need reorganizing while a third stands at the register with nobody in front of it, you are paying three people to produce one person's gross profit. That is not a people problem. That is a scheduling problem you created.

Saturday customers leave without being helped. The inverse signal. If you have a browsing rush at 1pm on Saturday and one associate is stuck doing a ring sizing consult while three people wait to drop off batteries, you are losing transactions you already paid to acquire. Count walkouts for two Saturdays. Owners are consistently surprised.

You cannot answer "what should this shift produce?" This is the deepest signal. Every scheduled shift should have an expected gross-profit output attached to it. If Saturday has eight people and your floor is $350, that Saturday is expected to generate roughly $2,800 in gross profit. If it does not, you now have a real diagnostic question — was the target wrong, was the traffic wrong, or was the execution wrong? Without a number attached, you cannot even ask.

The same signals show up in adjacent trades that run a skilled bench behind a retail counter — shoe repair, custom framing, guitar shops with a luthier in the back, small-engine repair, optical shops with an on-site lab. Anywhere skilled labor and walk-in retail share a footprint, the counter's rhythm and the bench's rhythm are different, and a single flat schedule serves neither.

What good looks like versus what bad looks like

Bad staffing is flat. It puts the same headcount on Tuesday as on Saturday because that is administratively simple, and it treats the bench jeweler and the counter associate as interchangeable bodies. Good staffing is shaped — it varies by weekday, it splits the count by role, and it slots people into the hour bands where the receipts actually ring.

Here is the difference in practice at a single storefront.

Bad: four people every day, Monday through Saturday. Twenty-four person-days per week. The owner feels covered. Monday's gross profit is $1,700, which means four people produced $425 each — fine, slightly above floor. Saturday's gross profit is $2,800 across the same four people, meaning each person carried $700 while customers waited and some left. The owner sees a "good Saturday" and never learns that Saturday could have been $3,400 with eight people on the floor.

Good: the same twenty-four person-days redistributed. Monday five, Tuesday four, Wednesday four, Thursday four, Friday five, Saturday eight — that is thirty person-days, so you either add hours or you trim the slowest midweek day to three and find the rest in part-time Saturday coverage. Either way the shape changes, and the shape is where the money is.

Bad splits roles by whoever is available. A certified bench jeweler pulled to the counter to ring up a $40 battery is a $95-an-hour resource doing $18-an-hour work. Good staffing protects bench time with a hard fence: bench people do bench work, and the counter absorbs intake, quoting, and retail sales.

Bad ignores the intake-to-completion lag. Jewelry and watch repair is not a same-transaction business. The gross profit from a Saturday intake posts when the work is finished and picked up, sometimes ten days later. If you schedule against *completion* dates, you will chronically understaff intake days and overstaff pickup days. Good staffing schedules the counter against intake and pickup volume, and schedules the bench against work-order hours in the queue — two different signals, two different curves.

Bad has no minimum. Good sets a floor of two on the floor at all times regardless of what the math says, because one person cannot take a lunch, cannot leave the register to open a case, and cannot handle a security situation. The math tells you how many above the minimum, never below it.

Real cost and ROI ranges

The cost of getting this wrong is measurable in both directions, and the two failure modes cost different amounts.

Overstaffing costs you payroll on a linear curve. One unnecessary eight-hour shift at a counter wage plus payroll burden runs roughly $150 to $200 all-in. Repeat that on three weekdays and you have spent $450 to $600 a week, $23,000 to $31,000 a year, on dead air. For a single-storefront jewelry shop, that number is frequently the difference between a good year and a flat one. The correction is nearly free: reassign those hours to Saturday or trim them entirely.

Understaffing costs you on a curve you cannot see. A walkout does not appear in any report. The customer who came in for a watch battery, saw three people ahead of them, and left is a lost $15 to $30 transaction — but the real loss is that battery customers convert into sizing customers, sizing customers convert into repair customers, and repair customers convert into the $800 custom redesign. You lost the top of a funnel, not a battery. That is why understaffing a Saturday is the more expensive mistake even though it looks like the safer one on the payroll report.

The bench math runs differently from the counter math. Bench labor carries higher gross margin — the material cost on a sizing or a solder is small relative to the labor charge, so a productive bench jeweler can clear well past a $350 daily floor on routine work alone. That is exactly why the floor should be role-specific. A common structure is a higher daily floor for the bench and a lower one for counter and intake roles, because the counter's job is partly to *feed* the bench, and feeding is not directly monetized on the day it happens.

The tooling cost is close to trivial relative to the payroll it governs. Scheduling software in this category is priced two ways, and the shape of your shop determines which is cheaper. Per-user pricing tends to land in the low single dollars per user per month for scheduling-only tiers, rising as you add time and attendance. Per-location pricing typically runs a flat monthly fee per storefront regardless of headcount, and several vendors offer a genuinely usable free tier for a single location. For a shop with two bench jewelers and six rotating part-time counter staff, per-location pricing is dramatically cheaper — eight people on a per-user plan versus one flat location fee is not a close call. For a lean crew of three full-timers, per-user pricing usually wins. Check current published pricing before you commit; these plans change.

Where the ROI actually comes from. Not from the software. The software publishes a schedule to phones and handles swap requests, which is real but modest value. The ROI comes from the discipline of attaching an expected gross-profit number to every shift, which changes behavior long before it changes headcount. When a bench jeweler knows the shift is expected to produce a specific number, the marginal hour gets spent quoting the overhaul instead of tidying the case. That behavioral shift is the return, and it costs nothing.

A realistic payback frame. If redistributing your existing person-days recovers even two lost Saturday transactions a week at an average ticket well above a battery, and simultaneously removes one unnecessary midweek shift, the annualized swing on a single storefront is comfortably in the five figures. You did not add payroll. You moved it.

How it plugs into your workflow

This is a RevOps problem wearing a retail apron. The method is the same one revenue operations teams run on sales headcount — set a per-head productivity target, divide the required output by the target, get the headcount, then place capacity against the demand curve. The only thing that changes is that the "quota" is a daily gross-profit floor and the "territory" is a weekday.

Here is how it slots into an operating rhythm you can actually sustain.

Monthly: refresh the divisor. On the first of the month, pull trailing 90-day gross profit by weekday from your POS or repair-management system. Recompute the headcount for each weekday. Most months nothing moves. When something does move — a new competitor opened, a mall anchor closed, wedding season started — you catch it within thirty days instead of within a year.

Quarterly: re-set the floor. If your team consistently clears the per-person target, raise it. A floor that never moves stops being a floor and becomes a ceiling, because people optimize to the number you gave them. Raising it from $350 to $400 is not a productivity demand; it is an acknowledgment that the mix shifted toward higher-margin work. Say that out loud when you raise it or it reads as a squeeze.

Weekly: place the shifts. The monthly math gives you counts. The weekly pass gives you placement. Pull hourly transaction and work-order-intake data and find your real peaks — for most bench-and-counter jewelry shops that means a Saturday browsing wave and a weekday lunch band of quick batteries and sizings, with consults and appraisals clustering midweek when people can book an appointment. Staff the counter to those bands and keep bench coverage steady rather than peaky, because bench throughput is governed by queue hours, not by who is standing in the store.

Daily: check one number. At close, gross profit divided by bodies scheduled. Compare to the floor. That is the whole daily ritual. You are not looking for perfection — you are looking for a persistent gap in one direction on one weekday, which is the signal to re-divide.

Feed the schedule downstream. The headcount plan should drive your hiring plan, not the other way around. If the division says Saturday needs eight and you have six people willing to work Saturdays, you have a recruiting requirement with a number attached, which is far more actionable than "we should probably hire someone." It also drives your training plan: if the bench is the constraint on three weekdays, apprenticing a counter associate onto basic bench work is a capacity investment with a computable return.

Where the data lives. Repair-shop management systems and retail POS platforms both export transaction-level data with timestamps; either one gets you the weekday and hourly curves. If your system will not export cleanly, ninety days of manual daily-total entry in a spreadsheet is one afternoon of work and it is enough to run the whole method. Do not let tooling be the reason you keep staffing on inheritance.

Related questions

How is this different from just scheduling to labor percentage?

Labor percentage is a rearview check — it tells you whether last week's schedule was affordable. The gross-profit-per-person divisor is a forward instruction: it tells you how many people to put on Thursday before Thursday happens. Run both. Use the divisor to build, use labor percentage to audit.

Should bench jewelers and counter associates share one target?

No. Bench work carries higher margin per labor hour, so a shared target either understates what the bench should produce or overstates what the counter can. Set two floors, divide bench gross profit by the bench floor and counter-driven gross profit by the counter floor, then add the two headcounts.

What if I only have one bench jeweler?

Then the bench is your constraint and headcount math applies only to the counter. Your real question becomes queue management: how many work-order hours are in the backlog versus how many bench hours you have per week. Fix the constraint by cross-training or hiring before you optimize the counter schedule.

Does this work for a mall counter versus a standalone storefront?

Yes, but the curves differ sharply. Mall locations inherit the mall's foot-traffic pattern, which pushes weight to evenings and weekends. Standalone storefronts skew toward lunch and late afternoon. Same division, different weekday and hourly shape — never copy one location's schedule onto another.

How long before the data is trustworthy?

Ninety days is the working minimum for weekday averages because it smooths single-event noise. Thirty days will show you obvious shape; it will not survive one bad weather weekend. If you are newly open, start with an estimate, divide, and correct monthly until you have a full quarter.

FAQ

What if my shop doesn't have enough history to compute gross profit by weekday?

Use whatever you have — even six or eight weeks shows the Saturday-versus-Tuesday shape clearly. If you are brand new, estimate expected weekly gross profit, split it across weekdays using a rough traffic assumption, and divide. The formula does not require good data to start; it requires you to keep correcting the input monthly. Accuracy compounds.

Is $350 a day the right per-person floor for my shop?

It is a starting point, not a rule. The right floor is the number your leadership team can defend out loud to the staff: what an average person doing an average mix of batteries, sizings, solders, and over-the-counter sales should produce on an ordinary day. High-ticket shops with heavy custom and appraisal work set it higher. Volume battery-and-sizing counters set it lower. Set it, watch it for a quarter, adjust.

How do I handle December and wedding season?

Do not use trailing 90-day data for a peak month — it will understaff you badly. Use the same weekday from the same season last year as your divisor input. If last December's Saturdays averaged double a normal Saturday, the division returns double the headcount, which is correct. Layer a small buffer on the specific hours your history shows surging, then remove it after the season.

Can I use this for split shifts and part-timers?

Yes, and you usually should. The division returns total bodies needed for the day, not a shift structure. An eight-body Saturday can be four openers, two mid-shift, and two closers overlapping the peak — as long as the coverage curve matches the transaction curve and you never drop below your two-person floor at any moment.

What if my numbers say I need fewer people than I currently employ?

That is information, not an instruction to cut. First check whether you are measuring the right output — if the counter's job is generating bench work that posts as gross profit ten days later, a naive daily division will undercount the counter. If the number holds after that correction, the answer is usually redistribution before reduction: move hours to Saturday, move hours into a second location, or invest the excess capacity into cross-training.

Does this apply to multi-location jewelry groups?

Run the division separately per location, always. Blending data across stores averages away exactly the local pattern you are trying to staff against. What does consolidate cleanly is the floor itself — a single company-wide per-person target keeps the standard consistent while each store's own weekday curve produces its own headcount.

Sources

• U.S. Bureau of Labor Statistics — Occupational Outlook Handbook, Jewelers and Precious Stone and Metal Workers: https://www.bls.gov/ooh/production/jewelers-and-precious-stone-and-metal-workers.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 — Fair Labor Standards Act overtime rules: https://www.dol.gov/agencies/whd/overtime
• U.S. Small Business Administration — Manage your business, hiring and managing employees: https://www.sba.gov/business-guide/manage-your-business
• SCORE — free small business mentoring and operations resources: https://www.score.org/
• Jewelers of America — industry association resources for retail jewelers: https://www.jewelers.org/
• American Watchmakers-Clockmakers Institute — watch and clock repair trade body: https://www.awci.com/
• IRS — Employer's Tax Guide (Publication 15), payroll and withholding: https://www.irs.gov/publications/p15
• U.S. Census Bureau — Monthly Retail Trade Survey, retail sales data: https://www.census.gov/retail/

Related on PULSE

• [How Many Attendants Should I Schedule Each Day at My Car Wash?](/knowledge/tl0067)
• [How Do I Figure Out How Many People to Schedule Each Day and at What Times for My Single Store?](/knowledge/tl0002)
• [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)