The main methods used by managers and investors to analyze projects and companies—the DCF method and the residual income method—are based on the discounting of forecasted cash flows from existing and projected project baskets of the company. In practice, when the company examines the feasibility of projects, it analyzes the net present value of each project (or, in the case of a company valuation, the value derived from the models in comparison to the price of the company) and invests in the projects which demonstrate a positive present value. Using an excessively high interest rate in the discounting process results in many projects being mistakenly seen as uneconomical. In addition, the projection of cash flows from projects (or a company) is exposed to mistakes in the projection, which again may cause many projects to be rejected although they may create positive value. Furthermore, in many cases, when the analysts use the traditional projection methods, they do not give due consideration to the effect of the examined project on other projects (synergies) or to the fact that such projects could affect the company's growth possibilities even if they do not directly affect the growth of the company.

The basic assumption in models based on discounted cash flows (or residual income) is that the level of such cash flows is uncertain. Therefore, the discount rate is determined in a manner which "indemnifies" the investor for the risk entailed, from his perspective, by such uncertainty. The problem with this method is that it does not relate to the possibilities which are opened to the managers after the investment, namely, to the changes in the system of possible investments and sources of revenue, which become clearer after a certain period of time. The company and its investors always face different options, and the most basic among them is the option which most investors in the company benefit from: the possibility of investing based on the company's development. In other words, investors could decide to discontinue the financing the company if they are not pleased with the interim results. This is an example of what is called real options.

Valuations could incorporate these real options facing the company, which are usually not fully addressed by the traditional valuation methods. This underpricing is particularly evident in startups, especially those operating in high tech fields.

Types of Real Options

• Options to switch— These are options to use the company's tangible and intangible assets for other uses than those represented in its current line of products. For example, options can change the blend of raw materials or the blend of products as a response to changes in market conditions or divert the use of a technology to another field.

• Options to abandon— These are options to abandon an investment which turns out to be uneconomical before it consumes all of the resources needed in order to launch the product on the market. For example, an investment in an R&D project may be terminated based on its results, without bearing the costs of market penetration and production if it is clear at that time that the product is likely to fail.

• Options to delay— These are options to change the timing of necessary investments or of the implementation of irreversible decisions in accordance with changes in market conditions. For instance, if the launching of a product can be postponed until such a time that the price of the product will economically justify its production, such delay would significantly reduce the risk entailed by just assuming that launching the product regardless of the prevailing market price for it at the launch time.

• Options to scope and scale— These are options incorporating the probabilities for higher-than-forecasted growth rates, which enable the investment to be increased or the company's fields of activity to be broadened while utilizing the company's existing infrastructure or technology. For instance, a company may increase its investment in a chip factory it owns and increase the volume of production two years down the road, in accordance with the demand for such products over the course of the coming year.

• Options to stage— These are options not to commit to invest all of the necessary amount at once, and constitute in fact a series of options to abandon an investment. These options actually represent the venture capital industry on the whole. Investors are always examining the cost involved in a smaller investment pending the meeting of certain milestones, opposite the prospects presented by a larger investment at an earlier stage.

Pricing Real Options

Real options may be priced by using models from option theory while taking into account specific parameters of the company and of the industry in which it operates. In many cases, customized models are used to price each real option individually, and the resultant values are then summed. It is important to note that the arbitrary use of commonly used models, such as the Black-Scholes model or the binomial option pricing model (explained below), may at times produce meaningless results since the model is fundamentally unsuited to the examination of investments when the possibility of creating similar investment portfolios synthetically (which is an underlying assumption in most of these models) is limited. Furthermore, the Black-Scholes model assumes a possible price distribution which does not necessarily correspond to the distribution of the results of each project. That said, almost any option which is desired to be taken into account may be priced by various algorithms based, among other things, on the use of simulation tools.

In practice, all option pricing methods inherently assume decision trees that are based on a succession of binominal decisions, and the quality of the computation is dependent only upon the scope and accuracy of the information available to the valuator. Binomial models (of which the Black-Scholes model is a case in point) assume the ability of short selling of the asset under examination, and of sales or purchases of risk-free bonds. Consequently, the models assume that any distribution of possible results may be reconstructed by stock and bond portfolios whose composition is changed from one stage to the next in order to provide the investor with a risk-free portfolio. This is not to say that projects' forecasted cash flows should be discounted with risk-free interest, but rather that in the structuring of the pricing model, stock and bond portfolios are constructed in a manner that creates an environment where anticipated cash flows should be discounted with risk-free interest.

In many cases, the value of real options—despite the complexity of their computation—is so crucial that ignoring them and making exclusive use of models such as discounted future earnings or cash flows models, may produce a valuation that is dramatically lower than the company's true underlying value. Over and above the deficient valuation, disregarding real options incorporated in a technology could cause an investor to decide not to invest in a company, whereas their inclusion in the valuation process would have presented the startup in a more favorable light.

Usually, the real options facing entrepreneurs and investors are too complicated to describe by a series of options which are priced by the model. Simulation methods are therefore the most efficient and useful tool for estimating and pricing such options.