Two Percent Over, or Thirty-Eight: What Where You Buy Actually Costs
Two stackers spend $2,000 on silver on the same afternoon, on the same kind of metal.
The first pays about 2 percent over what the metal is worth, which is roughly where a disciplined auction bid lands once the buyer's premium and the shipping are counted. The second pays about 38 percent over, the sort of figure a bank counter or a gift-shop style seller can reach on small pieces.
The first walks away with about $1,961 of metal. The second walks away with about $1,449. Same money, same day, same silver. Roughly $511 of the second stacker's money went into the wrapper, and the first got about 35 percent more metal for the identical spend.
That is the whole argument for tracking where you buy, and it is almost impossible to see one purchase at a time. A few dollars over on a single coin feels like nothing. Thirty of them at the same place adds up to a different stack, and no single purchase on its own would have shown it.
The number that shows it
You cannot find this by comparing what you paid. Cash prices are not comparable across time: a coin bought for fifty dollars in a quiet market and a coin bought for fifty dollars during a run are completely different deals, and the price alone will never tell you which was which.
What is comparable is the premium. Take what you paid and compare it against what the metal in that piece was worth on the day you bought it. That difference is what the seller charged you for the wrapper, and it holds its meaning across years, currencies and markets.
This is the number almost nobody has, because getting it means knowing the metal price on every purchase date going back to your first buy. It is the one lookup a spreadsheet genuinely cannot do, and it is the wall every stack spreadsheet hits.
Group it by where you bought
Once you have a premium per purchase, group them by seller and average each group. Now you know something no advice column can tell you: what each place you actually shop at charges you, on average, over the metal. Not what they advertise. What you paid.
Two things worth keeping in mind when you read it.
It is not a league table. Sellers price differently for all sorts of reasons that have nothing to do with how you were treated: rent, staff, insurance, how much volume they move, what they choose to carry, whether they will buy it back from you later, and what the card processor takes. An auction lot with a buyer's premium and shipping is a different product again from a shelf price you can walk out with. A high premium is not automatically a bad buy: a hard-to-find mintage, a slabbed coin or a piece you wanted in your hand the same afternoon are all real things to pay for, and the channel comparison goes through what each one is genuinely best at.
And one purchase is not a pattern. It takes a few buys at a place before the average means anything, and a fair premium is a range rather than a line.
What the gap does tell you is what the wrapper costs at each place you buy. Thirty-eight percent over on plain bullion is worth seeing, whatever you decide to do about it.
The other number: what your metal has cost you
Most stackers can say what they paid for a particular piece. Cost per ounce across everything they hold is a different question, and it is rarely close at hand, because working it out means going back through every purchase and the answer stops being current as soon as you buy again.
It is everything you have paid, divided by every pure ounce you hold, per metal.
Paid means all-in: premium, fees, shipping and tax, because that is what left your account. Pure ounces means metal content rather than coin count, so a 90 percent silver half is not half an ounce and a 100 gram bar is 3.2151 troy ounces before purity. The melt walkthrough has the math. Keep it per metal, because an average across gold and silver ounces is dominated entirely by whichever you hold more ounces of.
The average weighs everything the same. Shipping, the buyer's premium, the buys made in a currency you have since stopped using, all of it counted at full weight beside the purchases that stand out on their own. That is what makes it worth having next to the ones you remember clearly.
Watching it move
A single average is a fact. The same average over time is the thing worth looking at, because it only moves when you buy.
A step down means recent buying came in below your history. A step up means the opposite. Long flat stretches are months you did not buy, and they are meaningful too: the metal moved during them and your cost did not.
A falling average is not a score. Buying more when the price is lower pulls it down by math alone, and it says nothing about where the price goes next. What it does show is the shape of your own buying over time, steady or timed to the dips.
The two things it does not mean
It is not your break-even to sell. Your average cost is what you paid. Selling means somebody else's buy price, and dealers buy under spot rather than at it. A stack sitting exactly at its average cost is a little short of break-even, and the gap is whatever your buyer's spread turns out to be.
It is not what your stack is worth. Worth is the metal at today's price. Cost is what you paid. Two separate numbers that get blurred together constantly, and the space between them is the premium and the price movement pulling in the same direction or opposite ones.
Working it out yourself
The average is two columns and a division: pure ounces per row, all-in paid per row, sum both, divide. Any spreadsheet does it.
The premium comparison is the one that stops people, for the reason above: it needs the metal price on every purchase date, not today's.
If you keep your holdings in the vault here, both are already drawn. Your average cost per ounce over time, stepped so it only moves when you actually bought, one panel per metal so silver is not squashed flat under gold. Underneath it, what each place you have bought from charged over the metal on the day you bought. It is on every plan, the free one included.
The short version
The same money buys meaningfully different amounts of metal depending on where you shop, and the difference hides completely at the level of a single purchase. The only fair way to see it is to compare what you paid against what the metal was worth that day, then group that by seller. Alongside it, your average cost per ounce is the honest summary of your whole buying history: better than any purchase you remember, not your break-even to sell, and not what your stack is worth.