The assumptions that decide a feasibility study

A feasibility study is a model of a business that does not exist yet. Every input is an estimate, which means the output is an estimate too — and the difference between a useful study and a decorative one is whether it says so.

## The inputs that actually move the answer

In most F&B models, the result is dominated by a small number of assumptions. The rest matter, but they do not decide.

**Covers or transactions per day.** This is usually the single most sensitive input and the one most often set by ambition. A study that assumes a full room from month one is not modelling a restaurant; it is modelling a wish.

**Average transaction value.** It compounds directly with volume, and it is frequently set from the menu's mid-price rather than from a realistic ordering pattern. People order differently at lunch and dinner, alone and in groups.

**Rent and fit-out.** Fixed, unavoidable and usually the largest single commitment. An error here cannot be recovered through operations.

**Labour.** Modelled as a percentage far too often. Labour is a schedule, and a schedule is driven by opening hours and by the demand curve, not by a target ratio.

**Ramp-up period.** How long until the venue reaches its steady state. Studies that assume a short ramp and studies that assume a long one produce very different cash requirements even with identical steady-state numbers.

## The question a study should answer

Not "does it work?" but "under what conditions does it work, and how likely are those conditions?"

That means every model needs at least three versions: a conservative case, a base case and an optimistic case, with the assumptions that differ between them stated explicitly. If all three cases produce an acceptable return, the decision is easy. If only the optimistic one does, you have learned something far more useful than a single number would have told you.

## Sensitivity is the real deliverable

A sensitivity analysis answers the question that matters: how much can each key assumption be wrong before the answer flips?

If the project stops working when daily covers fall five per cent below plan, that is a fragile project regardless of how good the base case looks. If it still works at twenty per cent below, that is a robust one. Two projects with the same headline return can differ enormously on this measure, and it is invisible unless the analysis is run.

## Cash, not just profit

A model that shows profitability in month eight and never shows the cash balance is incomplete. Restaurants fail on cash timing far more often than on the absence of eventual profit. The model should show the lowest point of the cash curve, when it occurs, and what funds it.

Pre-opening costs, deposits, initial stock, the payroll that starts before revenue does, and the first slow months after the opening novelty passes all need to be in the cash view.

## What honest looks like

An honest study is uncomfortable in specific ways. It states each assumption and its source. It labels an estimate as an estimate rather than presenting it in the same typeface as a known cost. Where an input genuinely cannot be known, it models a range instead of choosing a comfortable point. And it is willing to conclude that the project does not work, or does not work at this rent, or does not work at this size.

A study that only ever confirms the plan it was commissioned to test is not a study.

## Practical recommendations

1. List every assumption on one page, with its source and whether it is known, estimated or guessed.
2. Build three cases and state exactly which assumptions differ between them.
3. Run the sensitivity on the top three inputs and report how far each can move before the answer changes.
4. Model the cash curve monthly, including the pre-opening period, and identify its lowest point.
5. Ask what happens if the ramp-up takes twice as long as planned. This single test catches most under-funded projects.

## In short

The number at the end of a feasibility study is the least interesting thing in it. The value is in knowing which assumptions carry the result, how fragile they are, and what would have to be true for the project to work.