Last November Intel Corp. made worldwide headlines when it announced it was investing $1 billion—more than triple the amount it originally planned—to build the world's largest semiconductor plant in Vietnam. With assembly and testing operations in China, the Philippines, Malaysia and Costa Rica, Intel could have chosen any country for its new facility. It selected Vietnam, according to Brian Krzanich, Intel's vice president and general manager for assembly and test, because of its "very vibrant population, increasingly strengthened education system, strong workforce and very forward-looking government," in addition to its labor costs, which are among the world's lowest.

When companies choose to move to low-cost countries, there are three basic decisions to make: what to move, where to move and how to move. We've found that many managers lack a framework for making these decisions, one that puts the benefits and risks in the proper context and allows executives to make informed decisions that are consistent with corporate strategy. In a Bain & Company survey, we canvassed 138 manufacturing executives in a range of sectors. While more than 80% say that cost migration is a high priority in their industry, fewer than two-thirds have launched significant cost-migration initiatives.

The fact is, deciding what, where and how to move is never easy—and rarely are they all-or-nothing propositions. They require careful analysis of each product line, focusing on issues like relative labor costs, logistics costs, customer requirements and time to market. We've identified the ways cost leaders are thinking through these decisions.

What to move. The price of shifting an entire production facility is often so high that it just doesn't make economic sense. That's why cost leaders think in terms of functions, not factories. They realize that just by shifting certain carefully selected processes or activities, they often can match the savings of moving facilities without having to bear the shutdown and startup costs. Companies examine specific functions—such as finance or marketing—and components on a case-by-case basis, identifying those ripe for migration and sidestepping those that are not. Boeing has a center that does design and technical work in Russia, a country with deep aerospace-engineering skills. Procter & Gamble has its payroll done in Costa Rica.

In deciding which functions to move, the leaders also carefully take into account opportunities to build new markets in the host country. Intel stands to benefit from Vietnam's technology market, which is expected to grow by 20% a year. For Emerson, China accounts for more than $1 billion in annual sales. The US-based company's divisions maintain 29 manufacturing operations in 14 locations in China.

Where to move. While China and India are the most popular destinations for multinationals seeking lower costs, leading companies are far more likely to look beyond those two countries. They recognize that each country has its own risk and benefit profile, and that profiles can change over time. As a result, they decide on a short list of target low-cost countries, based on their long-term competitiveness as well as current costs and capabilities. For example, when examining these factors, some companies feel China's political uncertainty and weak enforcement of property rights outweighs the benefits of low-cost labor. Leading companies hedge their bets by taking a portfolio approach, accepting higher unit costs in some countries to protect against currency risks, political risks or the impact of natural catastrophes. Hungary's labor cost almost quadruples China's, but its highly educated workforce and relatively low political risk makes it a better bet for some multinationals seeking skilled manufacturing. Leading companies also build portfolios that balance the risk advantages of diversification with the scale advantages of consolidation.

How to Move. A company's own organizational structure often poses a major hurdle. Leading companies drive their moves to low-cost countries from the top down, while laggards tend to leave decisions up to individual units. The latter encourages incremental decision making rather than a strategic approach to cost management. Also, it prevents companies from reaping savings across business units by pooling sourcing, jointly developing new suppliers or expanding economies of scale in low-cost countries.

By contrast, cost leaders' top-down, centralized approach allows them to use scale to their advantage as they build out their presence in low-cost countries. Also, sometimes it's the only way to overcome resistance to the redeployment of labor and resources. Germany-based Siemens's Osram division has set a target of increasing low-cost country production from 15% to 33%.

The move to a low-cost country requires a major organizational effort and strong leadership. But by taking a methodical approach to making smart what, where and how decisions, companies will be able to avoid many of the problems that can undermine even the best intentions. And they can sidestep perhaps the greatest roadblock of all: decision paralysis.

Suvir Varma is a partner in Bain & Company's Singapore office. Michael Thorneman is a partner in Bain's Shanghai office.