How should you value a company? Three common approaches

23 July 2015 Matt Owen

Valuations are big news. Barely a week goes by without a hot startup completing a round of funding that values it in the billions. Air BnB is worth more than Hilton, Uber is worth more than Mercedes…

But how are these valuations calculated? There’s undoubtedly an element of hype at work with some of these, but what really makes business A worth more than business B?

Before we get started, it’s worth mentioning that no valuation model is perfect (or standard for that matter). Valuation is often as much art as science. While ad revenue might be the most important measure for an online publisher, it’s not going to matter for a company manufacturing pipe cleaners, so different metrics apply.

Let’s take a look at some of the more popular valuation methods.


We’ll start with precedents because it’s probably the most straightforward method of valuation (although it also has the potential to be the least accurate in many cases).

It’s also worth mentioning that it’s relatively unusual to rely exclusively on this method, which is far more likely to be used as part of a comparibles analysis (more on those in a moment), but it is still a valuable procedure.

Put simply, precedents involve seeing if a similar business or asset has been sold in the recent past, and looking closely at the elements of that deal. What was worth the most to the buyer? Why exactly were staff considered more valuable than hardware? As you can imagine there are a thousand and one points you could consider here.

Precedents are reasonably straightforward as, for the most part, the data you need is publically available, and the number of previous deals can help you asses market demand. It’s worth remembering that while public, some data may be incomplete, and as always, past prosperity isn’t always a guarantee of continued success.

Precedents can also be useful if you are valuing a business with a number of separate parts – a company that manufactures different products for example – as it will be difficult to find a directly comparible company for a business that makes sausages but also as a spinoff accounting arm. A number of separate precedents can help you find a ballpark figure to start with, although a more detailed analysis will be required.


Or ‘Company Comparibles Analysis’.

This is a fascinating approach, as it’s a method of valuation that in many cases is based on the individual skill and knowledge of the appraiser as much as the figures. In essence, Comps is the practice of comparing similar companies in order to find a value, but in practice the skill set required to do that goes far beyond simply knowing who your main competitors are.

To construct a really thorough model, your appraiser must have intricate knowledge of the industry, the sector, and the individual companies. A consultant’s consultant. Comps are a large part of the reason that top bankers earn those salaries.

With that said, there’s no reason you can’t begin constructing a comps model yourself. Assuming you have a list of your close competitors, then you can use DueDil’s company information to see how they stack up against one another. In order to begin constructing a model, you’ll need to use Valuation Multiples. This involves taking a given asset and finding a statistic – cash flow for example – that relates to its market value.

In trading, price-earnings ratios are popular because they relate to share price, but for enterprise measurements you need enterprise values, EBITDA being the most obvious place to start.

Again, the best figures to use change according to the business model. Although rarer, it would be quite acceptable to use subscriber numbers as a valuation multiple for a SAAS company.

The main thing to be aware of here is that experience really does count for a lot. Comps require a lot of year-on-year comparisons t be accurate, so your initial figures may be quite different from the final outcome a seasoned pro will provide.

DCF: the Discounted Cash Flow

The DCF model can be used to value all sorts of things, from single assets, all the way up to an entire corporation.

Essentially, this model involves adding up all the future expected cash flows (within reason) from whatever it is you are trying to value. How much will an asset be worth to you over one, three, or five years?

Next, these amounts are discounted to their present rate. There are a few ways to do this, but usually the weight average cost of capital (WACC) from the business is used. This involves some rather complex math that is probably best left to appraisers rather than shmoes like me, but essentially WACC is the cost of borrowing money, depending on the amount of equity or debt a company has.

Suffice to say that the higher the WACC, the higher the risk.

Getting back to DCF, add all those discounted cash flows together and you’ll have a reasonably good idea of the company’s value.

For the mathematically-minded among you, here’s how that looks on paper:


How does it work?

This model is particularly attractive to investors. If the number this valuation gives you is higher than the total amount you’re being asked to invest, then you could be on to a good thing.

Anything else I need to know?

Note that I used the word ‘could’ in that last paragraph. The trouble with projecting cash flow is… well, you’re projecting. As any good risk manager will tell you, anything could happen in the next five years, from a flattening of the market to alien invasion.

Unknown unknowns can wreak havoc with the most diligent projections, and as the timescales get longer, the higher the risk that your numbers are wrong becomes. Once you get a decade out, it’s common to apply a terminal value, but these timescales can be radically different from business to business, particularly in the era of growth-hacking and disruptive start-up models.

When you are projecting out, the smallest changes to integers can cause huge shifts in the final figures, so you need to be very sure of those projections.

These are the probably the three most commonly-used methods of valuation, although there are multiple other models available which make a qualified appraiser an absolute must for larger deals and investments.

As I mentioned earlier, methods tend to change radically across sectors, and even between seemingly similar assets. Likewise, if you are the one selling your business, then it makes sense to provide the highest possible figure by combining methods, but if you are a buyer, then something like liquidation value (how much would all the assets of a company be worth in a liquidation sale?) would be a far more beneficial approach.