Call (+1) 650-212-1212


Is Risk Assessment Is A Risky Business?

Sean Salleh 14 Oct 2013 Risk and uncertainty

As if risks in general weren’t enough, the question is now about the quality of the risk assessment itself. Even more daunting, lists of ‘mistakes in risk assessment’ have been multiplying thanks to different authors – and they each seem to come up with new and even conflicting errors for organizations and risk managers to commit. While risk assessment models will depend for their accuracy on the people who build them, it’s fair to say that Analytica as a modeling environment makes it easier for users to get things right. Here’s a list compiled from some of the errors being discussed today, and an indication of how Analytica can help to take the risk out of risk assessment.


Military risk assessment - but how would you consider 'catastrophic' and 'occasional'? Image source:

Should You Assess Extreme Events or Assess Their Consequences?

Focusing simply on extreme events could be risky in itself. Records show that past efforts to predict extreme events are somewhere between mediocre and lousy. In addition, obsessing over particular events may blindside us to other eventualities. The trick is to look at the consequences of such extreme events - for example if drops in demand or supply could adversely affect a business and by how much. Analytica lets you work both angles by modeling the consequences, but keeping events and drivers on record in meaningful influence diagrams.

History is Bunk – Except When It isn’t

Some hold that observing the past is futile if our aim is to predict the future. Others warn against re-inventing the wheel in risk assessment or starting from scratch. In fact, this sounds like the ‘light is a particle, no, it’s a wave’ debate. Both sides are right if the domain under discussion is defined appropriately. Extreme events do tend to defy past patterns. On the other hand, statistics show clearly that your IT services are most likely to break because of server failure, not because of a new Sandy (tropical storm). Make one model in Analytica for each if you want and join them up using Analytica’s Intelligent Array technology.

The Rush towards Residual Risk

Residual risk is what is left after risk management has been applied. However, comparison with the level of inherent risk (the risk beforehand) is important too. If it costs $1 million to bring an inherent risk of $100,000 down to a residual risk of $10,000, there’s clearly something wrong. Use Analytica to model both kinds of risk assessment and also to experiment with solutions to see how much risk you can truly eliminate for the smallest financial outlay.

Trying to Assess Everything

The risk in trying to cover every possible base is rather more subtle. As a risk manager, you don’t want to miss anything (big) in your risk assessment. On the other hand, those who try to do too much end up unable to finish their analysis and will necessarily miss some aspects of risk. With Analytica you can however assess the sensitivity of a risk assessment to particular elements and refine it accordingly. You can also evaluate how uncertainties in those elements combine in the overall result.

Faulty Handling of the Math

Sometimes faulty handling may be as simple as presenting a formula to an audience that expects a simple statement about the severity or the impact associated with a risk. Playing catch with standard deviations may seem simple to mathematicians, but business or functional managers may not grasp the significance behind statistical symbols or equations. As well as the option to show the underlying analytical machinery, Analytica allows you to present conclusions graphically and intuitively. That way, to paraphrase Thomas Harris, senior management may draw a line under your name rather than through it.

If you’d like to know how Analytica, the modeling software from Lumina, can help you take the risk out of risk assessment, then try a thirty day free evaluation of Analytica to see what it can do for you.

» Back

Sean Salleh

Sean Salleh is a data scientist with experience in guiding marketing strategy from building marketing mix models, forecasting models, scenario planning models, and algorithms. He is passionate about consumer technologies and resource management. He has master's degrees in Operations Research from University of California Irvine and Mathematics from Northeastern University.

Leave a Comment