Supplier costs, margin math, and keeping Shopify's numbers honest. Written for stores that buy wholesale and resell — no fluff, no growth-hacking.
Three increases of about 2% each add up to 6.4% and four margin points, and no alert fires for any of them. Eight steps, three queries and the sort order that puts the money at the top — plus why the worst-looking 38 lines are the wrong place to start.
A 15-point duty change lifts landed cost 13.3% and takes 8.0 margin points — three different numbers from one event. How to read your own entry, decode a supplier's absorbed pass-through from the price change alone, and find the rate where your floor actually breaks.
A 6% early-order discount plus 150-day dating is worth 11.8% against the in-season price — but only on units that sell. The breakeven sell-through that sizes the order: 43.9% on a carry-over staple, 77.0% on a model-year line that has to be cleared.
Three quarters of the cost of a coil is metal, so an eighty-cent move in the index takes ten margin points at a fixed retail. How to split base from adder using two editions of your own file — and find the $3.00 base increase hiding inside a price cut.
A $1.30 move on a $7.40 box is 17.6% and $1,170 a year on one SKU. Two sizes out of five drop through a 30% floor while the size everybody spot-checks looks fine — and a fixed annual contract price turns the whole thing into a locked loss.
A MAP cut with no cost change costs 7 margin points and never appears in a price file. A promo cost left in the cost field reports 50.3% on a line earning 40.7%. Four money flows called "rebate", only two of which belong in cost per item.
Discontinuation arrives as a missing line, not an email — and most diffs are structurally blind to absence. The three things called "discontinued", and last-time-buy math where 420 units nets +$1,150 and 700 nets −$3,130.
Four pricing layers, and the two that move most often never appear in the file. A multiplier change that lifts cost 6.8% with every printed list price identical, and one job quote that puts a 12.3-point hole in a product group.
Twelve hundred lines, 148 cost changes, nine decisions that need a human. An archive rule you can't build retroactively, an intake log, the four normalisations before any diff, and why the unmatched pile is where price processes die.
The same part delivered two ways: 28.0% stocked against 19.0% dropshipped — and a slow bulky line where that reverses by 3.5 points. Plus the thing nobody costs: stock is a hedge against your supplier's price file, and dropship has no runway at all.
Price is the concession a supplier defends hardest and the one small buyers ask for first. What terms, freight rules, increase notice and a rotation allowance are each worth in dollars — and why the three cheapest asks add up to more than a point of price.
A pack discount has a peak, and past it the carrying cost eats the concession: 20% off buys nothing, 8.6% off buys 1.3 margin points. The effective-unit-cost formula, a tail SKU that loses 8 points to its pack, and the pack change that hides a 25% increase.
A supplier quoting 2% under can cost 2.1% more delivered. Spend-weighted cost drift instead of the file average, fill rate and lead-time spread converted to dollars, what 2/10 net 30 is really worth, and a six-dimension weighting that survives an argument.
The cheapest list price is often the dearest delivered — pack sizes and freight thresholds decide it. Which lines deserve a second source, how to match the same part across two files, and the tier-discount math where saving 5% loses $173 a year.
Three costs with three different shapes. What one return costs by disposition — $20.70 even when the item goes back at full price — how to size a blended allowance, how shrink converts to margin points, and the four ways smearing it all into cost per item breaks your supplier diffs.
A 45% margin at 1.2 turns earns less per dollar than a 22% margin at 9 turns. GMROI as a ranking tool, one tail SKU costed properly, the attach-rate check that reverses the answer, what three exit routes actually cost, and why a 12.9% cost increase cuts return on inventory by a third.
A discount is quoted on the price and paid out of the profit. A 15% code against a 40% margin destroys 37.5% of the gross profit and needs 60% more units to break even. Two formulas, a discount-vs-margin table, and the maximum depth your floor allows.
The threshold is a margin calculation, not a round number borrowed from a competitor. Break-even and target-contribution formulas, a threshold table by label cost and margin, why a 6% cost increase moves a $36 threshold to $44, and the basket that clears $39 and earns 64 cents.
Three costing methods give three different numbers for the same SKU — ten margin points apart in a rising-cost year. What each method is for, worked on one six-month timeline, and why replacement cost is the strongest fit for Shopify's single field.
Payment fees, shipping labels, packaging, returns and chargebacks stand between product margin and order profit. The per-order stack worked on three orders — including a $19 order that loses money at a 40% margin — and why a 10% cost increase can cut order profit by a third.
Gross profit = net sales − recorded cost, and nothing else. Payment fees, shipping labels, packaging, missing and stale costs never enter it — and every error runs in the flattering direction. How to read the reports honestly.
The floor is the alarm line where a SKU stops paying its way. How to derive it from your cost-to-serve, why fast movers and the long tail need different floors, and what to do the day a price file pushes a SKU through it.
Hardware and industrial suppliers don't send a price — they send a list and a rule. Multipliers and column pricing, per-C and per-M quoting, standard packs, effective dates, and the surcharges that move your cost with no new price book at all.
Fixed catalogs, contract pricing, open quotes and volume breaks keep selling at yesterday's cost. Why the channel with the thinnest protection is usually your biggest one — plus five post-price-file checks and the clauses that prevent a repeat.
How warehouse distributor pricing actually works — line codes, jobber vs dealer net, ACES and PIES, supersessions, update cadence — plus a review routine that survives a catalog of tens of thousands of part numbers.
A refundable deposit that isn't cost, a price floor set by the brand, and part numbers that change identity mid-year. Three distortions generic profit tools get wrong — and how to handle each.
Quantify it in dollars before you feel anything about it, verify the letter against the actual file, pull the levers that exist, reprice selectively — and make the next check automatic. Includes the forward-buy math.
List vs jobber vs net, chained discounts, per-case pricing, part numbers Excel silently mangles, and the rows that quietly disappear. A field guide to matching a supplier's spreadsheet to your catalog.
Freight, duty, brokerage, FX and shrink turn a $9.00 line into $11.13 — a 35.7% margin into 20.5%. How to build a landed cost, how to allocate it, and when it isn't worth the effort.
Bulk editor, CSV import, apps and the Admin API — plus the empty-column trap that blanks thousands of costs in one import, and why you always export first.
Four formulas that matter: your real margin now, the hold-margin price, the monthly dollar impact, and how much extra volume you'd need if you chose not to reprice. With worked examples.
What the cost field actually feeds, its three structural limits, and the subtle reason stale costs permanently corrupt your profit reports instead of failing loudly.
A 50% markup is a 33% margin. The confusion only ever runs one way — it always flatters you. Formulas, a conversion table, and the two places it costs real money.