← Back to analysis

WMSTurkiye analysis · ANL-04

Why do WMS projects fail? Documented cases and early warning signs

Failures in warehouse system projects rarely come from a single software bug. Public cases and patterns that recur in the field point to the same things: unreliable master data, cut-short testing, going live at the wrong time and involving operations too late.

WMSTurkiye editorial team · Published:

Three public cases

The cases below come from different countries and from projects that were not WMS-only; but what happened in warehouse and distribution processes clearly shows the risks a WMS project can face. Figures are only as stated in the cited sources.

CaseWhat happenedLesson for your WMS project
Hershey (US, 1999)The system, spanning ERP, CRM and supply chain software, went live as Halloween orders were pouring in; the company said the problems would keep it from delivering about USD 100 million of product [1].Do not time go-live for the peak season; set the date against readiness criteria, not the calendar.
Target Canada (2013–2015)The team investigating the problems estimated that product information in the system was accurate only about 30% of the time (98–99% in the US). Distribution centres were bursting with stock while store shelves were empty; some items could not be processed for shipping. All 133 stores closed [2].If master data such as item dimensions, weights and units is wrong, the warehouse system cannot work. Make data cleansing the project's first task and add validation at entry.
ASOS (United Kingdom, 2019)Struggling to embed new automation software at its Berlin site, the company faced stock availability problems in the US and EU; the CEO called it “a failure in execution”. The company booked GBP 50.5 million of transition and restructuring costs [3].Launching a new site, automation and software at the same time multiplies risk. Plan a phased ramp-up and allow time to reach full capacity.

Seven common failure patterns

  1. Wrong or vague scope. A system is chosen before processes are written down; exceptions (returns, partial shipments, quarantine) surface in production.
  2. Missing or wrong master data. Without dimensions, weights, barcodes and units, putaway and picking rules cannot work [2].
  3. A project team that does not know the operation. The project is treated as an IT job; the warehouse team is not part of the design and does not own the system.
  4. Cut-short testing. When the schedule tightens, testing is the first thing cut; the system goes live without end-to-end and load testing.
  5. Going live at the wrong time. Going live in peak season, or with many changes at once, removes the time needed to fix problems [1][3].
  6. Unnoticed integration errors. Faulty messages from the ERP pile up silently; stock differences appear later.
  7. Unmeasured goals. Because accuracy and productivity were not measured before go-live, neither success nor failure can be proven.

Early signs of the “return to Excel” after a WMS

Failure does not always look like a crisis. Often the system seems to work while operations quietly start working around it. Watch for these signs in the months after go-live:

If two or more of these signs are present, measuring the current state with the operational diagnostic is a good way to separate software, process and data problems.

Reducing the risks

If you would like to share an anonymous lesson from your own project, write to info@wmsturkiye.com. Shared cases are published only with explicit permission, with company and personal details removed.

Sources

  1. CIO, “Supply Chain: Hershey's Bittersweet Lesson” — cio.com (accessed: 2026-09-19)
  2. Canadian Business, “The Last Days of Target Canada” — canadianbusiness.com (accessed: 2026-09-19)
  3. RTÉ / Reuters, “ASOS warns on profit again after botching warehouse revamps”, 18 July 2019 — rte.ie (accessed: 2026-09-19)

Method and corrections: editorial policy.