Warehouse Productivity Metrics: Why One Unit of Measurement Is Never Enough

In this episode of Rozmowy Logistyków, Piotr Skobało explains how to properly measure warehouse productivity. He covers when to use UPH, LPH, or OPH metrics, and why B2B and B2C warehouses should use different KPIs.

FOUR BASIC UNITS OF MEASUREMENT

Units per hour (UPH) works well in e-commerce, where a customer orders a single product. Lines per hour (LPH) — order lines per hour — better reflects a worker's real effort with a varied assortment, because one line can hide either one unit or 50 of the same product. Orders per hour can be useful, but with high diversification across product groups stored in different zones (some in automation, some on racking), it can average things out and blur the picture. Pallets or cartons per hour work well in large distribution warehouses with full-pallet turnover. None of these metrics is universal — the choice depends on the type of business, and it's often worth using several in parallel.

FTE OR ACTUAL PRESENCE

A metric calculated per FTE (full-time equivalent) always carries inaccuracy — sick leave, vacation, and absences dilute the result. A more precise method counts actual presence in the warehouse (time cards, WMS login, scanners), though even a handwritten log is better than nothing. A key rule: once you've chosen a denominator, it has to stay constant — changing the counting method partway through generates misleading conclusions about a rise or drop in productivity that's actually just a result of the measurement method changing.

WHERE TO GET DATA WITHOUT AN ADVANCED WMS

The most reliable measurement comes from a dedicated system with end-to-end scanning, generating timestamps at every stage of the process — but even without such a system, you can start with the basics: order or unit counts matched against employee attendance. That's enough to get a sense of where a company stands, before investing in more advanced tools like a Labor Management System. A warning from practice: an employee who finds a gap in the system's measurement logic (e.g. how scan sequencing works) can "optimize" the result without any real process improvement — which is why good measurement also requires staying alert to what's actually being measured.

ROUGH BENCHMARKS FOR DIFFERENT BUSINESS TYPES

E-commerce/B2C: 80-150 lines per hour with a good layout and scanning, 150-250+ with voice picking or pick-to-light; packing 60-100 shipments/hour. B2B: 40-80 lines/hour, with a 20-40% increase after deploying scanning — a good basis for justifying a WMS business case. Wholesale picking: 15-30 lines/hour. FMCG/grocery: 20-40 pallets/hour for pallet picking, 100-200 units/hour for piece picking. Industrial/spare parts: 15-35 lines/hour, lower due to greater precision needs and SKU fragmentation. Logistics operators serving multiple clients: usually 30-100 lines/hour, closely tied to per-client profitability accounting.

THE SIGNAL TO EXPAND YOUR METRICS

The moment worth rethinking your measurement system is when e-commerce orders start growing as a share of a warehouse built for B2B — at a 2-3% share, the difference is negligible, but once it grows to 5-10% and beyond, picking individual units instead of whole pallets starts needing entirely different productivity benchmarks.