Logistics — AI implementation
A logistics business was adopting AI tools without first defining the operational requirement. The work began with the business itself.
20 September 2026
The situation
The business was already experimenting with AI tools, but the starting point was the technology rather than the operating requirement.
That matters because a logistics process can appear to offer obvious automation opportunities while still depending on exceptions, timing, judgement and information that sits outside the apparent workflow.
The work
The assignment started with the business itself: what people were trying to achieve, where information came from, what decisions were being made, which steps created delay or duplication, and which problems were actually worth solving.
Only then was AI considered.
The practical question became where does AI genuinely improve the operation, and where would it merely add another layer?
The conclusion
AI was treated as one implementation option inside a wider business requirement, not as the purpose of the project.
That distinction changed the nature of the implementation from “use AI” to “solve these operating requirements, using AI where it is appropriate”.