How to Price Jewelry Repairs Without Guessing

Price a repair as two separate halves, labor and metal, and put each half on its own schedule. The labor half comes from what an hour at your bench actually costs you, not from a national average. The metal half comes from today’s metal cost, not from the sheet you wrote when gold was cheaper. Most stores get the first half wrong once and the second half wrong continuously.

Search for how to price jewelry repairs and almost everything you find was written for the person handing you the ring. It tells them a resize runs $40 to $90 and a chain solder starts around $30, so they know what to expect at your counter. Useful to them. Useless to you. This is the other side of that counter.

What should a jewelry repair cost?

There is no national answer, and the spread is the proof. David Geller, the bench jeweler who spent nearly three decades publishing repair pricing for the trade, found that a straightforward sizing job retails anywhere from $20 to $85 depending on the store, and that every one of those stores closes about nine out of ten repair jobs anyway. Same work. Four times the price. Same close rate.

So a table of average repair prices is not a benchmark. It is noise from thousands of stores with different rent, different benches and different courage. The only price that means anything is the one built from your own cost of doing the job. That is what the rest of this is about.

One caution before you start copying figures. The “charge $100 to $125 an hour” number that circulates on jeweler forums and gets republished as current advice comes from an article Geller wrote in January 2019. It was sound then. It is seven years old now, and the cost of running your shop, keeping a bench employed and buying metal has not held still since. Take the method from that era. Do not take the number.

Why the repair counter is the most underpriced part of your store

Because it is the part jewelers are most afraid to touch, and the fear does not match the evidence. Geller’s comparison, published in INSTORE in June 2022, is worth reading twice. Ten people look in your showcase and you sell three or four of them. Ten people come in with a repair and typically nine leave the item with you, a win rate he puts at 80 to 90 percent.

Now compare what each of those close rates costs you to hold. The showroom needs hundreds of thousands of dollars in inventory sitting there aging. The shop, Geller notes, usually runs on under $20,000. Your highest-closing, lowest-inventory department is the one with the price list nobody has revisited since the last owner wrote it.

His explanation for why the close rate holds across that $20 to $85 spread: the shop is trust-sensitive, not price-sensitive. Someone handing you a ring their grandmother wore is not shopping four quotes. They are deciding whether they trust you with it. Price is a distant second, and a suspiciously low price can even work against you.

Then there is the asymmetry Geller kept running into, and you probably recognize it. The same jeweler who will not raise a sizing price by $5 or $10 after several years will raise a replacement tennis bracelet by $500 without a second thought. The bracelet feels like the market moving. The sizing feels personal. It is not. It is a service with a cost that has gone up.

The two halves of a repair price

Every repair price is labor plus metal, and the two halves fail in opposite directions. Labor gets set once and then frozen out of fear. Metal goes stale quietly, without anyone deciding anything. Pricing them as one lump is why a store can raise prices and still lose ground on the jobs that use the most gold.

Geller’s worked example from April 2018 shows both halves in one job. A new 3mm half shank. He figured the sizing stock at 1.13 pennyweights, and with 14K gold at $41 a pennyweight at the time, that was $46.33 in material. He marked materials up three times, giving $139 retail on the metal. Then the labor: installing a half shank means soldering at two points, which he priced as two sizings at $42 each, plus shaping the stock to match at half a sizing, $21. That is $105 in retail labor including polish. $139 plus $105 is a $244 retail price, and the bench, paid 26 percent of retail labor, cost him $27.30 of it.

Follow the structure, not the figures. The one input in that example that has moved hardest since 2018 is $41 a pennyweight, and it is the input a stale price sheet never updates. If you are still charging the 2018 answer, you are absorbing every dollar gold has moved since, on every metal-heavy job, out of your own margin. Our post on pricing custom jewelry when the gold price keeps moving works through the metal arithmetic and the reprice trigger in detail.

The labor half: build a rate from your own bench

Start from what an hour of bench time actually costs you, then correct for the hours that are not billable. Geller’s time studies found that a bench jeweler produces roughly five and a half billable hours in an eight-hour day. The rest goes to setup, cleanup, walk-ups, the phone and the jobs that come back. His fix was blunt: whatever a job should take, add 25 percent.

That correction is where most homemade price sheets break. A jeweler times a resize at twenty minutes, prices twenty minutes, and never accounts for the fact that only about two-thirds of the day is chargeable at all. The overhead does not disappear. It just comes out of the store instead of the job.

Decide how you pay the bench in the same breath, because it sets your floor. Geller paid his jewelers a percentage of retail labor rather than an hourly wage, 26 percent in that example, which means the labor cost of a job scales with what the job is worth instead of how slowly it went. Whether you copy that or pay salary, you need the number, because a labor price you cannot trace back to what the bench costs is a guess wearing a price tag.

The metal half: why a stale price sheet leaks quietly

Because a service cannot be repriced after the fact. Jonathan Geller, a 15th-generation jeweler who now produces the Blue Book his uncle started, made the sharpest version of this argument in INSTORE in June 2026: “Services don’t have turn like inventory, so you don’t get another shot at it.”

Sit with that for a second, because it is the whole case. A ring in your case that you bought badly can be repriced next month, or next year, or marked down and moved. A resize you quoted on Tuesday at last year’s metal cost is finished. The margin you gave away is gone the moment the customer says yes, and there is no second chance at that job. Stores that write repair pricing once and revisit it when they remember are not leaving money on the table. They are handing it back, one job at a time.

His practical move is to build the higher tier before you need it. Keep a second price list for the metal-heavy work, the sizings, shanks and clasps, ready to go. When the market justifies it, you switch to it rather than improvising under pressure. And when a long-standing customer balks, you have somewhere to go that costs you a discount you chose rather than a discount you conceded.

The simplest version, if you are running your own sheet and have not touched it in years, is the one David Geller gave: raise everything 10 percent now, and do it again three months later. Two moves of that size are absorbed almost invisibly at a counter closing nine jobs in ten. One large correction after five years of nothing is the one your regulars notice.

Not every service deserves a price. Some deserve a no.

Some service lines lose money at any sane price, and the trade has already told us which ones. In INSTORE’s 2024 Big Survey, an anonymous form that drew more than 700 responses from American jewelry store owners and managers, respondents were asked which parts of their business were bad profit centers. Watch repair topped the list, especially work sent out of the store, with one respondent putting their margin excluding batteries at around 15 percent.

The rest of that list is worth knowing before you build a menu. Loose diamonds came second. Appraisals, engraving and pearl stringing were named for the time and the equipment they demand, though several stores keep them deliberately as loss leaders that bring people through the door. And the one that should stop you: some jewelers named custom design itself, not because the margin is bad but because they underestimate labor and materials and end up breaking even on work they assumed was their most profitable.

There is a pattern under all of it. Margin leaves when the work leaves your bench or your wheelhouse. Sent-out watch work, jobs needing equipment you do not own, repairs to badly made pieces bought elsewhere, specialty work beyond your shop: every one of those is a job where you carry the customer relationship and the risk while someone else does the part that earns. Price those to reflect what they really cost you, or stop taking them.

One honest caveat from that same survey. A store that dropped watch repair entirely brought it back after customers got angry. A loss leader you choose, priced with your eyes open and defended as a service that keeps good customers walking in, is a legitimate decision. A loss leader you never noticed is not. And if custom is the line quietly breaking even in your store, that is a quoting problem rather than a pricing one: we wrote about quoting custom jewelry faster without underpricing your work separately.

Four questions before you write a repair price

Run these in order on your top ten jobs and you will have a defensible sheet in an afternoon. The order matters, because each one changes the answer to the next.

One. Does this job belong at my bench?If it goes out, or needs equipment you do not have, or sits outside what your jeweler does well, you are the middleman on someone else’s margin. Price it as such or decline it warmly. This is the question that removes jobs rather than pricing them, which is why it comes first.

Two. What does an hour at my bench actually cost me? Bench pay, benefits, tools, supplies, bench-related overhead. Your number, not an article’s. Write it down, because everything below multiplies against it.

Three. How long does the job really take? Time it, then add 25 percent for the day that is not billable. If you skip the correction you will underprice every job by roughly the same margin you were trying to protect.

Four. When did I last raise this price?If the answer is more than a year, or you cannot remember, that price predates your current metal cost and your current payroll. Put a date on the sheet and a reminder in the calendar so the next answer is never “whenever.”

Then put the price somewhere a customer can find it

A price list nobody can see still costs you the job. Most jewelry store websites say “we offer repairs, call us,” which asks a stranger to phone during business hours to learn whether you can size a ring and roughly what it runs. The stores winning that search are answering it on the page. We made the broader version of this argument in our post on how independent jewelers compete with online diamond retailers: personal service is not a strategy until you name it, price it, and let someone start it without calling you.

It is also a demand problem, not only a pricing one. Every repair is a person handing you a piece of jewelry they care about, which makes your repair counter one of the best sources of custom work in the store. That is the argument in how to get more custom jewelry orders, and a published service menu is how that door opens for people who have not walked in yet.

This is where JewelerStudio fits, and only after the arithmetic above is done. Studio AI renders a piece from a description in seconds, so a customer asking what their old setting could become sees it before they commit, and Ring Prolets a shopper build and price a ring on your site. Both stay white-labeled to your store, and the enquiry lands with the details attached instead of as a voicemail saying “call me back about my ring.” Plans run $149 to $849 a month and every plan includes a 7-day free trial.

No tool will set your prices for you, and you should be suspicious of one that claims it can. Do the four questions first. If a plain page listing your services with honest starting prices and a booking link does the job for your store, that is a real answer, and it is free. Book a demo if you want to see the rest of it running on your own site, or see pricing first.

The short version

Your repair counter closes around nine jobs in ten on under $20,000 of inventory, and it is trust-sensitive rather than price-sensitive, so the fear that has kept your sizing price flat for five years is not supported by how your own customers behave. Price the labor from what your bench costs you, corrected upward 25 percent for the hours that are never billable. Price the metal from today, and keep a second tier ready so you are not improvising when it moves again. Say no to the jobs that leave your bench, or charge for what they really cost. Then put the list where a stranger can read it at eleven at night, because that is when they are looking at the ring and wondering whether it can be saved.