A feasibility study has two possible audiences. The first is the committee that must approve the project. The second is the management team running it eighteen months later, trying to work out whether the thing is still on course. Studies written only for the first audience are rarely any use to the second.
Feasibility, viability, sustainability
The distinction matters. Feasibility asks whether the project can be done. Viability asks whether it produces an acceptable return once it is. Sustainability asks whether it still does so over the long run, after the enthusiasm and the initial funding have both been spent.
A study that answers only the first question will pass a board and fail a business. The long-run test is the one that separates a project from a commitment.
Assumptions are the deliverable
The model is not the product of a feasibility study. The assumptions are. A spreadsheet whose inputs cannot be traced to a source, a document or a stated judgement cannot be revisited when conditions change — and conditions always change.
Written properly, each assumption carries three things: what it is, where it came from, and what would have to happen for it to stop being true. That third element is what converts a static approval document into a live management instrument.
Stress the case that worries you
Sensitivity analysis has a tendency to test the variables that are easy to flex rather than the ones that would actually break the project. The discipline is to identify, before modelling, the two or three assumptions on which the whole case rests — and to test those to destruction.
Viability that holds only in the base case is not viability. It is optimism with a discount rate attached.
Build it to be re-read
The best evidence that a feasibility study was done well is that someone pulls it off the shelf two years in and finds it still answers questions. That only happens when it was written for the operator rather than for the approval.
