As value investors, we define an investment's risk by its business risk, not its stock price volatility. As such, we have discussed a number of potential business risks on this site, from customer concentration, to liberal use of leverage, to reliance on a single product.
But an exhaustive list of the risks a business can face is not possible. Every business is unique, and so every business faces a unique set of risks that cannot necessarily be identified unless the business is well understood. For this reason, it may be helpful for investors to think of a business' risks not by going through exhaustive checklists, but instead by thinking about the business' single points of failure.
In industrial, networking and other contexts, a single point of failure (SPOF) is "a part of a system which, if it fails, will stop the entire system from working." For example, a typical laptop has one keyboard, one screen, and one CPU, all of which represent single points of failure. A set of dual servers with redundant hard drives and multiple network connections is built for high-availability, however, and so it would take a lot more to go wrong before a failure is experienced.
This concept can be translated to the world of investing. That single point of failure to a company's ability to maintain its value could be a single salesperson or an ability to refinance debt at a particularly time, for example. By first understanding a company, and then by identifying potential single points of failure, the investor is better able to identify the unique risks that a business faces.
This doesn't mean that a company that has seemingly eliminated single points of failure will necessarily maintain its value, however. For one thing, it is very hard to identify and eliminate all single points of failure. For example, if the server system's ISP goes down in the example described above, even that supposedly robust system may no longer be able to serve its purpose. Furthermore, many things can and do go wrong at once, and so even companies with no single points of failure can be adversely affected if things don't go their way.
Conversely, even companies with easily identifiable single points of failure can be great investments. For example, though it has made many, many efforts to diversify with varying success, Coke has been quite reliant on selling its flagship product to remain successful. And yet it has done so for numerous decades. As such, estimating risk is a matter of not only identifying the single points of failure, but also determining the odds of such a failure occurring.
It's management's job to add redundancy to the system. Where points of potential failure exist, management should expend effort to reduce the risk. But as value investors, we cannot rely on the fact that everything will go smoothly in the future because it has done so in the past. We must identify those points of failure which increase downside risk, and avoid them when the odds of failure are such that we are not receiving enough of a discount for our principal. Finding and understanding the single points of failure is the key to limiting downside risk.