GMROI, gross margin return on inventory investment, is gross margin dollars divided by the average cost of inventory you held to earn them. A GMROI of 2.4 means every dollar tied up in stock returned $2.40 of gross margin over the period. Retailers trust it because it joins the two numbers a product business lives on, margin and inventory, into one figure that can be compared across SKUs, categories, channels, and years, and because it punishes the two ways a seller loses money without noticing: thin margins and slow stock.
The formula
GMROI equals gross margin divided by average inventory at cost.
Gross margin is net revenue minus cost of goods sold for the period. Average inventory at cost is the beginning inventory value plus the ending inventory value, divided by two, both valued at what you paid, not at retail. Use the same period for both halves. A 12-month gross margin over a 12-month average inventory is the standard version; a quarterly GMROI is fine as long as you label it and do not compare it to an annual figure.
Two things about the inputs decide whether the number is worth anything. COGS must be real landed cost per unit, including freight, duties, and inbound handling, not the factory invoice. And inventory must be valued under a consistent method. The IRS lays out that requirement in Publication 538: where inventory is necessary to account for income, you value it at the beginning and end of the year and apply the method consistently. FIFO is the common choice for ecommerce because unit costs drift upward with each container and FIFO keeps the ledger honest about it.
A worked example with real numbers
Take a seller with one SKU, a kitchen scale, over a calendar year.
- Units sold: 9,600
- Net revenue after refunds and marketplace fees: $172,800 (an average of $18.00 net per unit)
- Landed cost per unit: $7.40, so COGS of $71,040
- Gross margin: $172,800 minus $71,040 equals $101,760
- Inventory at cost on January 1: $22,200 (3,000 units)
- Inventory at cost on December 31: $31,080 (4,200 units)
- Average inventory at cost: ($22,200 plus $31,080) divided by 2 equals $26,640
GMROI equals $101,760 divided by $26,640, which is 3.82. Every dollar held in scales returned $3.82 in gross margin over the year.
Now the same scale after a supplier price increase and a slow second half. Landed cost rises to $8.60, sales hold at 9,600 units, and the seller over-ordered in Q3, ending the year with 7,500 units.
- COGS: 9,600 times $8.60 equals $82,560
- Gross margin: $172,800 minus $82,560 equals $90,240
- Ending inventory at cost: 7,500 times $8.60 equals $64,500
- Average inventory: ($22,200 plus $64,500) divided by 2 equals $43,350
GMROI equals $90,240 divided by $43,350, which is 2.08. Revenue did not move. Margin fell 11 percent. GMROI fell 46 percent, because the seller was earning less on each unit while holding far more of them. That gap between the margin change and the GMROI change is the whole reason the metric exists.
Why retailers prefer it to margin alone
Gross margin percentage tells you how much you keep per sale. It says nothing about how long the cash sat in a warehouse before the sale happened. A 55 percent margin product that turns twice a year can be a worse use of capital than a 30 percent margin product that turns nine times. Inventory turnover captures the speed but ignores the margin. GMROI is the product of the two, which is why a merchandiser can rank an entire catalog by it and the ranking means something.
It also survives the mix problem. A category with high-margin slow movers and low-margin fast movers might show a middling average margin and a middling turnover. Its GMROI, computed per SKU and then compared, shows which half of the category is carrying the other.
What GMROI does not tell you
It ignores operating costs below gross margin. Advertising, storage fees, returns processing, and payroll all sit outside the numerator. Two SKUs with identical GMROI can have contribution margins that are nowhere near each other if one needs $4 of ad spend per unit to sell and the other sells on organic search. Use GMROI to allocate inventory dollars, then check contribution margin before deciding a product is a winner.
It also depends on average inventory being representative. A seller who receives one giant container in November and ends the year with a warehouse full of it will show a distorted year-end value. Monthly averages, twelve data points instead of two, fix that for anyone whose stock levels swing.
And it is only as accurate as the COGS feeding it. Marketplace sellers who post revenue from settlements but estimate COGS as a flat percentage of sales will get a GMROI that is arithmetic on a guess. Tools that carry landed cost per unit and post COGS as units ship, ConnectBooks being one example for sellers on Amazon, Shopify, Walmart, TikTok Shop, and eBay syncing into QuickBooks or Xero, exist for exactly this reason: the inventory metrics are worthless unless the cost layer under them is real.
How to use it week to week
Compute it per SKU on a trailing 12-month basis, refreshed monthly. Sort descending. The top of the list is where the next inventory dollar should go. The bottom of the list is a set of decisions: raise price, cut the reorder, liquidate, or drop the SKU. Then compute it per channel, because the same SKU can carry a different net revenue per unit on Amazon than on Shopify once each platform’s fees come out, and the channel with the lower net is the one dragging the number down.
Retailers who run this discipline for a year usually find the same thing: a small share of SKUs produce most of the gross margin return, a long tail ties up cash for almost nothing, and the catalog they thought was balanced was not. GMROI does not fix that. It makes it impossible to keep ignoring.
