Probability weighted valuation

Some businesses do not have a single future that a forecast can describe. A drug either gets approved or it does not, a licence is either granted or refused, a case is either won or lost. The value in each of those states is very different, and there is nothing in between. A single forecast built on risk adjusted revenue smooths a switch into a dial and produces a number that describes a company existing in none of the possible worlds.

The alternative is to build each branch as its own valuation, on whatever method suits that branch, and then weight the results by the probability of reaching them. The output looks similar to a conventional answer and means something quite different: it is the expected value across outcomes rather than the value of an expected outcome.

Where there is a sequence of gates rather than a single one, the same idea extends into a decision tree, with a probability at each node and a value at each terminal branch. That structure has the useful side effect of showing where the value is created and destroyed, because it makes visible how much of the answer depends on clearing the very first gate.

Be honest about what the method does not fix. The probabilities are judgements, usually drawn from historical success rates for comparable situations, and small changes in them move the answer a long way. What the method buys is not precision. It is that the shape of the analysis matches the shape of the uncertainty, and that the assumption doing the work is visible instead of buried inside a revenue line.