I have a rule of thumb I give every founder who asks me about building their own warehouse: if you're doing under 30,000 orders a month and you're already pricing racking, you haven't finished the math. Last year a DTC skincare founder in Eastvale, California showed me his plan — 20,000 square feet, a team of twelve, 18,000 orders a month, and he wanted to double the space and put in a conveyor line. His logic: the 3PL charged him over four dollars per order, and his own warehouse would come in under three. "The savings are pure profit," he said. I asked him to lay six months of 3PL invoices next to his actual warehouse spend, line by line. He went quiet. His real per-order cost running his own building was a dollar higher than the 3PL invoice.
This is not an unusual story. Over the past decade I've run this calculation for at least twenty companies, and the error is always the same direction: people see the 3PL's variable cost clearly and underestimate how hungry a fixed-cost warehouse is. The numbers below are ranges I've seen in the market, for modeling purposes only — 3PL quotes and local rents vary widely, so use your actual quotes.
Compare per order, not per month
The most common mistake happens at step one: comparing "the 3PL charges me $80,000 a month" against "my own warehouse costs $60,000 a month in rent and labor" and concluding the warehouse saves $20,000. That comparison is meaningless because the denominators don't match. The 3PL's $80,000 moves with volume — in a slow month at 40,000 orders, you pay for 40,000 orders. Your warehouse's $60,000 is carved in stone. Halve the volume and you still pay the full $60,000.
The only honest comparison is per order. Your true 3PL cost per order is simple: last month's total invoice divided by order count — receiving, storage, pick and pack, packaging, accessorials, all of it. Your own-warehouse cost per order is (monthly fixed cost + monthly variable cost) divided by monthly orders. Fixed cost means rent, the baseline crew you keep on payroll, equipment depreciation, amortized WMS fees, insurance, the miscellaneous property bills. Variable cost means piece-rate labor, packaging, and any parcel rate differential.
Back to the skincare brand: $52,000 a month in fixed cost on that 20,000-square-foot Eastvale building (small-bay industrial in the Inland Empire area runs roughly $1 to $1.50 per square foot per month, triple net, depending on location and lease term), $21,000 in variable cost, 18,000 orders — $4.05 per order all-in. The 3PL invoice averaged $3.90. His own building cost more, and it consumed twelve people's worth of management attention. Do the math per order and the answer surfaces on its own.
Three costs the spreadsheet always misses
When the own-warehouse math comes out cheap, it's missing at least one of these three costs. Usually all three.
First, idle cost in the slow season. E-commerce has peaks and valleys — Black Friday through Christmas can run triple a normal month. A warehouse staffed and sized for the peak sits half-empty for four months, and that dead weight gets amortized into your annual per-order cost where it looks terrible. A 3PL spreads peaks and valleys across dozens of clients, so you only ever pay for what you use. I know a seasonal swimwear brand whose own-warehouse math came to $5.60 per order annualized; after moving to a 3PL it was $3.80. The entire gap lived in those five slow months.
Second, the management tax. Hiring, training, workers' comp, scheduling, WMS maintenance, equipment repair — a 3PL spreads a dedicated team across dozens of accounts, and you have to build all of it yourself. One seller I know went from eight warehouse staff to thirty, and the founder started losing two days a week to warehouse personnel issues. "Everything I saved on fulfillment, I paid back in my own time," he told me.
Third, the tuition cost. Your first warehouse layout, your first process design, your first SOPs — they will be wrong, and every revision costs months of efficiency. A 3PL has made two orders of magnitude more mistakes than you have. I put it this way: the tuition for running a warehouse has already been paid by the 3PL, and you're choosing between amortized tuition (their service fee) and paying it fresh yourself. The former is almost always cheaper.
Where break-even actually sits
So where is the line? Here's a formula you can use directly:
Break-even monthly orders ≈ own-warehouse monthly fixed cost ÷ (3PL cost per order − own-warehouse variable cost per order)
Example: $80,000 a month in fixed cost, $4.20 per order at the 3PL, $1.80 in variable cost per order running it yourself. Break-even = 80,000 ÷ (4.20 − 1.80) ≈ 33,000 orders per month. Sustain volume above that and your own building starts winning; below it, stay with the 3PL.
Three adjustments are mandatory. First, average order value and margin. At a $15 average order with 30% margin, an extra dollar of fulfillment cost is life or death — push the break-even up by at least 30%. At $120 average order in a high-margin category, fulfillment is a small slice and you can build earlier for control over the customer experience. Second, SKU count. With over a thousand SKUs and a long tail, your own pick efficiency will structurally lag a 3PL's (they get pick density from multi-client volume) — inflate the "own-warehouse variable cost per order" in the formula. Third, growth trajectory. If orders grow 15% a month, you can build six months ahead of the curve. If volume is flat, never build for "maybe someday" — I've seen too many warehouses built for projected growth end up as the founder's private warehousing museum.
The hybrid answer
By now some readers expect the conclusion to be "build once you're big enough." Wrong. In practice, the setups that run smoothest are hybrids — at least half the cases I've worked on landed there.
Play one: own the winners, outsource the tail. Keep the 20% of SKUs that drive 70% of orders in a small building you run — fast shipping, controlled experience — and hand the long tail plus returns to a 3PL. A pet-supply client runs exactly this: an 8,000-square-foot building handling 300 hero SKUs, with 2,000 long-tail SKUs and all returns at a 3PL. Blended cost per order runs $0.70 below pure self-operation.
Play two: split by coast. Run the West Coast yourself (close to your suppliers and your core market) and use a 3PL as an East Coast forward stocking location. Two to three days of parcel transit difference between coasts is an experience gap that fulfillment savings can't buy back.
Play three: own building for B2B bulk, 3PL for B2C parcels. Bulk and parcel operations have completely different rhythms, and mixing them in one building drags both down.
Here's my suggested next step, and it's concrete: pull six months of 3PL invoices and compute your true per-order cost. Then run the formula above with real local numbers — actual rent quotes for your market, actual wage quotes (warehouse associate pay in the Inland Empire has been running roughly $18 to $22 an hour in recent years, depending on the hiring channel). Put the two numbers side by side. The answer is usually clearer than people expect. And if the break-even sits at 30,000 orders while you're doing 15,000, stop agonizing — put the energy into traffic and product and let specialists run the building. That's the real value of this math: not telling you which option is cheaper, but telling you when it's even worth asking the question.
