Much of the failure of long-term strategizing arises not from its difficulty but from confusion about how to do it. When hard strategic choices loom, executives ask their experts what the future will look like. The experts — middle management, analysts, engineers, consultants — aren’t asked to provide a comprehensive view of possible futures and their probability of occurring. They’re asked for a description of a single future or several possible futures for consideration.
With exquisite mock precision, they describe these highly specific futures, shrugging off uncertainty on the grounds that the future is ultimately unknowable. Risk is relegated to a qualitative discussion of potential problems, often resulting in an upward adjustment of the discount rate to compensate for the lack of accounting for uncertainty in the overall analysis.
But the uncertainty of the future is no excuse for less rigor or clarity. A consistent, quantitative perspective on uncertainty not only builds the best foundation for making good management decisions but also provides a platform for developing a shared understanding of trade-offs, bridging disagreement and establishing accountability.
Consider a multinational, consumer brand company that was trying to rebalance its sourcing locations for the coming decade. It had long outsourced most of its manufacturing to a limited number of Asian countries, with many factories in each. But with sales volume expected to double in 10 years, it was considering whether to outsource to additional countries to spread its risks more evenly.
Sponsored by AccentureThe technologies and processes that are transforming companies.
Those risks were many. The environmental and labor practices of subcontractors could damage the brand. Supply could be disrupted by natural disasters, political unrest, labor strife, or lack of infrastructure investment. And for the highly visible brand companies — companies in industries like footwear, luxury goods, and consumer electronics — reliability of supply is paramount. Their margins on products far exceed manufacturing cost and empty store shelves mean huge lost opportunities.
But after almost a year of discussion, the company’s leadership team was no closer to a decision than when it started. Each of several courses of action had passionate advocates: 1) keep manufacturing in the current locations and expand capacity there; 2) plan to contract for a significant percentage of future manufacturing in a somewhat less politically stable country (which we will call Country A); 3) return to another Asian country (Country B), where a tentative, previous outsourcing operation had stalled; or 4) press ahead with automation experiments that would make some reshoring feasible.
Frustrated by their inability to resolve their differences, they turned to a rigorous probabilistic approach to decision making that has its origins in the centuries-old work of the mathematicians Daniel Bernoulli and Pierre-Simon Laplace, who first tried to figure out how to make rational decisions in the face of uncertainty, as in games of chance.
The team began by bringing in internal and external experts to provide their perspectives on the risks and uncertainties surrounding each sourcing option. Those risks and uncertainties were then assessed and converted into a range of probabilities for each.
For example, they concluded that a strike at factories in one of the existing locations had a 90% probability of lasting at least two days, a 50% probability of being resolved within two weeks, and less than a 10% chance of requiring more than four weeks for resolution. They also assigned probabilities to the likely strike frequency across all of the factories in each country.
This probabilistic assessment approach was used for every aspect of the sources of value and the sources of uncertainty: the threat of earthquakes, flooding, civil unrest, the possible introduction of the Trans-Pacific Partnership (TPP) trade agreement, and much more. In each case, the team related the uncertainty to the scale of the supply-chain disruption it caused, creating a quantified assessment of the uncertainty involved. The team then quantified each of those probabilities in terms of the scale of supply-chain disruption they would cause.
What emerged was a clear and quantified picture of the likely consequences for each alternative, including how exposure to new locations could increase or decrease supply security, how new partners might expose the company to additional brand risk, and where the greatest geopolitical and environmental risk most probably lay.
The team was then able to translate those probabilities into supply loss or gain and, eventually, into margins. That enabled the team to calculate a range of upside and downside financial potential for each alternative. Initially, they found that entering Country A and B gave the company a 70% chance of making tens of billions of dollars more than with any other alternative. Further, they determined that they could ramp up fairly quickly in Country B, where they already had a small presence, allowing them to wait a year to see how the political situation played out in Country A. At the same time, they could see how rapidly their work on automation progressed and they realized they could revisit the possibility of reshoring when they again took up the question of moving into Country A.
With a carefully quantified range of potential value, the company’s leaders were able to break their impasse. They were confident that they now understood and could quantify the many risks and uncertainties, compare the various alternatives, and match the choice to the risk appetite of the company. And clear quantification of likely value and uncertainty allowed a hybrid solution to emerge that otherwise might not have been considered. They were not just making a “good enough” decision but the best possible decision they could.
The great irony is that by embracing and quantifying uncertainty, rather than evading it, they were able to achieve much greater clarity and utility in their view of the future.
via HBR.org http://ift.tt/1TfzcKv