1 Million Square Feet. What Your DC Actually Costs.

July 2026. Element Logic and Lululemon cut the ribbon on a 1 million square foot automated distribution center in Ontario, Canada. That number is the wrong thing to benchmark. The right number is cost-per-unit-shipped. If you don't know yours, that's the problem.

What the Average Operator Is Actually Paying

Manual DCs run cost-per-unit in the $38 to $44 range depending on labor market and SKU velocity. Top-decile automated facilities are landing closer to $26 to $29. That gap compounds across every order cohort you ship this quarter. One million square feet sounds like a capital story. It isn't. It's a throughput density story. Lululemon is not paying for space. They are paying for picks-per-hour at scale, with predictable labor variance heading into peak. You should be thinking the same way, regardless of your facility size.

The Port Risk Number You Cannot Ignore

A Verisk Maplecroft report published this week puts one-third of the world's highest-volume ports at high disruption risk. Not theoretical risk. Operational risk tied to weather events, geopolitical friction, and infrastructure fragility. For warehouse operators, this is an inbound velocity problem. Disrupted ports mean irregular container arrivals. Irregular arrivals destroy the clean replenishment cycles that automated DCs depend on. Your DC's throughput performance is only as good as the inbound signal feeding it. If your receiving cadence is chaotic, your pick efficiency collapses regardless of what equipment you run.

Three Metrics That Separate the Top Decile

First: throughput density. Measure picks per square foot per hour, not total picks. Space is expensive. Every dormant aisle is dead NetPPM. Second: inbound dwell time. How long does a SKU sit in receiving before it is slottable? Top-decile operators average under 4 hours. Average operators run 18 to 24. That gap shows up directly in sell-through velocity during the first 72 hours of a product drop. Third: cycle count frequency on your top 20% velocity SKUs. If you are running monthly cycle counts on your fastest movers, you are operating blind for 29 days at a time. Weekly minimum. Daily if you can staff it or automate it.

Three Actions Worth Running This Quarter

Pull your cost-per-unit-shipped for the last 90 days by SKU cohort. Segment by velocity tier. You will find a small number of slow-moving ASINs consuming disproportionate pick labor. Those are your first automation candidates or your first discontinuation candidates. Either way, that decision is worth making now. Next, map your top five inbound port dependencies. Given Verisk Maplecroft's disruption risk data, identify which SKUs are single-sourced through a high-risk port. Build a 30-day landed cost model assuming a 2-week delay on each. The brands that did this exercise in early 2024 had buffer stock positioned before Q4 pain hit. The brands that did not ate the margin. Finally, if you are evaluating automation, start with your receiving and put-away process before you touch pick. The Lululemon model works because inbound is clean. Dirty inbound plus expensive automation equals expensive chaos.

Three Questions to Pressure-Test Your DC Strategy

Can you state your cost-per-unit-shipped by velocity tier without pulling a report first? If not, your team is managing square footage, not margin. Which of your top 10 inbound SKUs by revenue touches a port flagged in the Verisk Maplecroft high-disruption tier, and what is your buffer position today? If your facility scaled to 2x current order volume tomorrow, would your receiving process be the constraint or your pick process? The answer tells you where to spend next.