Automation is often seen as a shortcut to efficiency. However, when implemented before operations are stable, it can amplify problems instead of solving them.
One of the most expensive examples of this in modern business history is the collapse of Target Canada — a case that offers an important lesson for growing companies considering large system upgrades.
In 2013, Target expanded into Canada with strong brand recognition and significant capital investment. The company opened 133 stores in a rapid rollout. On paper, the expansion strategy looked solid — demand existed, the brand was trusted, and the stores were well designed.
The failure was not caused by pricing, product selection, or customer interest. The problem was automation introduced before operational reality was stable.
Before fully stabilizing its Canadian operations, Target deployed a complex automated infrastructure that included a new ERP system, automated inventory management and replenishment, large-scale migration of product and pricing data, and a fully automated supply chain operating from day one.
From a systems perspective, the design was modern and scalable. In practice, it relied on data and processes that were not yet reliable.
The automation depended on accurate data across thousands of products and supply chain points. The data was incomplete and inconsistent. Once the automated systems went live, the errors quickly multiplied. Stores began experiencing inventory appearing in the system that did not physically exist, warehouses reporting stock-outs while shelves were full, replenishment orders triggering incorrectly, and prices in the system not matching shelf labels.
Employees were instructed to trust the system. Human judgment was overridden by automated processes. Because the system controlled the network centrally, errors were not isolated — they were replicated across the entire operation.
Within two years of launch, Target shut down all 133 Canadian stores, wrote off approximately USD 7 billion, and exited the Canadian market completely.
Target automated before key operational foundations were ready — specifically, before processes were proven, data quality was reliable, exception handling procedures were understood, and human override mechanisms were clearly defined. Automation locked flawed assumptions into the system and amplified them across the network.
This case is often dismissed as a "large enterprise problem." It is not. Mid-sized companies frequently repeat the same mistake when they implement ERP systems before processes are stable, automate reporting before decision ownership is clear, remove manual checks too early, or assume dashboards are more accurate than operators.
The scale may be smaller, but the consequences are still significant. Automation does not remove operational problems. It multiplies them.
Automation should never be the first step. It should come after process clarity, reliable data, defined decision rules, and clear escalation paths. Without these foundations, automation does not create efficiency — it institutionalizes error.
Avanor works with leadership teams before and during automation initiatives to identify which processes are ready to automate and which are not, stress-test assumptions embedded in systems, define where human judgment must remain, and prevent scalable failures caused by premature automation.
Sometimes the most valuable automation decision is a simple one: not yet.
Talk to us before you automate — not after something breaks.
Get in touch