Q&A with Glenn Koller (BP Amoco)

 

MarketingNPV (MNPV): Risk management seems to have a pretty good foothold in most parts of the organization except marketing. Any theories as to why?

Glenn Koller: In fact I don't think it is true. If you come from a marketing background, you tend to see the deficiencies in that particular area. But if you were in IT or operations or law or some other area of the company, once you get under the covers you see a lack of risk assessment and management in all aspects of the organization.

Most big companies are made up of lots of little enclaves or cul-de-sacs within the company that may operate as fiefdoms and they tend to do things the way they want to do them. So the view of risk differs significantly by your area of concern within the company. For example, in finance a certain type of risk is a good thing — no risk, no reward. But in the health and safety area, to the extent it can be, all risk must be wiped out. Marketers have to see clear benefits. Unless they can see the benefits, why would they do it?

MNPV: So what kind of benefit would a CMO of a mid-to-large-sized, reasonably complex marketing operation derive from a better risk assessment process?

Koller: Most companies are pressed to make difficult decisions about allocating resources between two or more very attractive options. Well, if I'm forced to decide between Option A and Option B, one of which I can more reliably risk-assess and the other less so, I will either choose B and hope that the risks are minimal, or perhaps select A over B if I can't get my mind around the risk/reward of B. Unless all my choices have been comprehensively risk-assessed, I can't possibly compare their expected economic impacts. So if I'm the CEO or CFO, and IT is making a strong, risk-adjusted resource request while marketing is asking for funding without a thorough assessment, I can 1) ask marketing to go do a risk assessment quickly so as not to hold up the decision; 2) give the money to IT and tell marketing "tough luck"; or 3) give the money to marketing and hope for the best. In today's environment, it's hard to imagine selecting #3.

MNPV: What kind of questions do you get most often from marketing executives?

Koller: Many people think that if you put a range of uncertainty around things then you've taken care of risks. They have a spreadsheet and in that spreadsheet there is a cell for a number reflecting an uncertainty factor but the actual number will be different. So they build distributions around their best guess. But risk goes beyond uncertainty. There is a risk that your brakes may fail on your drive into the office. There is uncertainty surrounding the probability of that happening, and uncertainty around the cost of that — "Will I damage the fender, total the car, wind up on disability?" etc.

Most questions I get are about assessing the uncertainties and working them back into big spreadsheets with lots of numbers that few people ever actually take the time to understand.

MNPV: So what type of questions should you be getting?

Koller: Managers should focus more on the chances of abject failure Īݠthe probabilities that the project will be a complete failure with a negative financial outcome or that it will never get off the ground in the first place. Lots of projects fail either technically or economically. If you set an economic threshold for success, like a minimum ROI or NPV or customer acquisition volume, there is always a chance (and often a good chance) that you won't clear the threshold and the project will fail. There are all kinds of risks of abject failure — ranging from the risk the project will fail to capture the imagination of the CEO who will kill if before it gets started, to technical, environmental, and political problems. All sorts of things can cause failure at various points in a project lifecycle. Many people don't even think of those outcomes so they don't spend too much time assessing them before the projects begin.

MNPV: Is that linked somehow to that pathological optimism that marketing executives are often accused of suffering?

Koller: Culturally we plan on success. We count on success. But a lot of times we don't get success. Yet psychologically, we don't like to think about things going wrong. Most managers compete for resources within the company and they find it difficult to see how a long list of things that could go wrong would help them make their case for their project. So unless the person up the ladder asks for a risk-adjusted proposal, it behooves them to not make their project look worse than it could.

But the bottom line is that when we look at these projects for the chance that things could go haywire, we can much better judge the expected outcome than we could just by examining the positive scenarios. Every project in the company should be evaluated against the likelihood that it does not go exactly according to plan. Because chances are, it won't.

 

Read a review of Koller's book Risk Assessment and Decision Making in Business and Industry: A Practical Guide.