+31 (0)850654710
    NLDEFREN

    The hows of the warehouse · Part two of five · Throughput

    A day's work in two hours

    Throughput is never a rate. It is a requirement inside a window — and almost every decision about method, labour, racking and machines is an answer to how wide that window is.

    The argument: Which peak are you actually solving

    Part one of this series measured demand — the seven figures that describe what a business has to handle. Part two is the first set of choices: how you will work, with how many people, on what equipment.

    It starts by rejecting the way throughput is usually discussed. Picks per hour, orders per day — these sound like design targets and they are not. Throughput is a quantity of work that has to happen inside a window, and the first job is working out which window actually binds.

    There are three, nested inside each other. A seasonal peak spanning weeks. A day-of-week peak spanning days. And an intraday peak spanning hours, which almost nobody measures and which usually decides the size of the building.

    Take two businesses shipping 1,200 orders a day. One receives them evenly from eight in the morning. The other receives most of them between three and five in the afternoon, against a five o'clock cut-off.

    Required pick rate, spread against compressed The same 1200 orders spread across eight hours requires a modest rate. Compressed into two hours it requires four times that rate, and everything sized against it multiplies with it. Spread across 8 hours 150 orders/hour Compressed into 2 hours 600 orders/hour 8 hours 2 hours Four times the pickers, four times the equipment, four times the pack benches, four times the despatch lanes — idle for six hours a day.
    Identical daily volume. Completely different building. The window sizes headcount, handling equipment, pack benches, despatch lanes, staging area and dock doors all at once — and none of them scale against the daily total.

    The conclusion: Move the window before you size for it

    This is the part of the paper I would want someone to take away even if they read nothing else.

    Building capacity to service a self-inflicted constraint is the most expensive mistake in this series, and it is permanent. Try to widen the window before you buy anything that exists only to survive it.

    Most narrow windows are not physical laws. They are the consequence of a cut-off nobody has revisited, a carrier collection time nobody has renegotiated, and customers ordering late because nothing rewards ordering early. Moving the cut-off, splitting it by service level, or discounting orders placed before a certain hour are all cheaper than the building you would otherwise need — and the orders you shift are usually the ones that had no reason to be late in the first place.

    Only when those are exhausted is the window a genuine ceiling. And then you design for it deliberately, knowing what it costs, rather than discovering in year three that you built a warehouse to service a promise nobody had priced.

    What the full paper covers: Nine sections

    • The three nested peaks, and why seasonal and intraday peaks point toward opposite answers on automation
    • Why picks per hour is not a target — the congestion ceiling, and the diagnostic that tells you when hiring has stopped working
    • Throughput's second ceiling — you cannot pick an order that has not arrived, and what that does to headcount planning
    • Picking methods mapped to the order profile from part one — batch, zone, cluster, wave, goods-to-person, and which profile each suits
    • The system and the standard — why a temp on directed work reaches 80 to 90% of standard inside a week, and the two places that claim stops holding
    • Labour and its two ceilings — recruitment and seasonality, and why your peak is everybody's peak
    • Storage media as a throughput decision — density against accessibility, and the pallets-per-SKU number that decides it
    • Method, racking and equipment as one decision, not three vendor conversations
    • What automation is actually for — peak shaving and labour reduction size in opposite directions, the unit of economics is the pick rather than the product, and the comparison almost nobody makes: automation against the cost of the move it defers

    Eight diagrams, the arithmetic worked through, and the field examples — including a new warehouse that opened 15% short of picking locations, and a distributor whose floor would shrink by four fifths.

    Get the rest of the series

    The hows of the warehouse runs to five parts. Leave your details and the remaining four arrive one a week for four weeks — a link to each paper, never an attachment. After the fourth the sequence ends: this is consent for this series only, not a newsletter.

    How we handle your data

    Where the series goes

    Part two produces an operating design: picking method, storage media, equipment class, headcount at peak, shift pattern, cut-off and automation. Seven decisions, each of which constrains the building.

    Part three puts a shell around them — footprint, clear height, column spacing, dock doors, fire class. Permanent things, sized against the choices made here.

    Part two of The hows of the warehouse. Willem ten Asbroek — BizBloqs Management Solutions B.V., Netherlands. September 2026.

    Elsewhere in this series