Executives looking for a deeper understanding of how business operates should look at birds flying in formation or at exotic species thriving in an ecosystem, according to proponents of complexity theory.

The theory's growing number of backers say companies have traditionally limited themselves to the approach of classic science. Companies focus on a few variables and assume that A-leads-to-B-leads-to-C.

But complexity's proponents argue that it has made great progress over the past several years in showing how the complicated interactions of many autonomous agents—such as birds and exotic species—can be explained with surprisingly simple rules. Backers say those observations can be extended to the world of business to explain the complicated patterns of actions and reactions among competitors, suppliers, and customers. As a result, executives should be able to make more realistic predictions about more diverse kinds of human behavior.

For now, complexity theory is the management trend that dare not speak its name. Writers apparently feel that complexity seems too, well, complex to command the attention of already-harried corporate executives.

Still, complexity theory is finding its way into management. The theory even influenced the U.S. Justice Department's antitrust suit against Microsoft, because of an economist's work on how organisms (or technologies) can become dominant in a competitive landscape.

Based on early attempts to apply complexity theory, here are some practical suggestions on new ways to look at business systems and at business problems:

PLACE EVEN MORE EMPHASIS ON BOTTOM-UP MANAGEMENT.

In 1997, the Marine Corps tested a core tenet of complexity theory, that small and autonomous beats large and centrally controlled. In a 10-day field trial of new tactics, small squads with state-of-the-art communications were allowed to make operational decisions, instead of waiting for orders from higher up. The squads were pitted against a much larger, mechanized force. The squads won. The larger force was "outflanked and outmaneuvered," says Ron Schultz, co-author of Open Boundaries: Creating Business Innovations Through Complexity.

The principle behind the success isn't just that decentralization is important. Traditionally, decentralization has been based on the idea that the person dealing directly with a customer or supplier will make a better decision than a senior executive who is more remote from the situation. Instead, complexity theory argues that organizations function like animal colonies in an ecosystem, adapting successfully to changing circumstances based on the spontaneous actions of individuals.

KEEP UP WITH THE COMPETITION WITH EVOLUTIONARY LEAPS.

Danny Hillis, a fellow at Walt Disney, once used a computer simulation to demonstrate that in an eons-long evolutionary contest between two species—host and parasite—both keep improving in fitness to keep up with each other. Compressing the equivalent of centuries into seconds, the simulation illustrated how species (and businesses) succeed for a time on developmental plateaus, then leap to higher performance levels as conditions change. So, businesses need to be prepared to fully exploit current core competencies, but they also have to be ready to shift to radically new ways of doing business on little notice.

IT'S A JUNGLE OUT THERE, SO PROMOTE BUSINESS DIVERSITY.

Tropical jungles sustain more species than temperate zones. So, as the business climate heats up and becomes more like a jungle, firms need to keep seeding new products and ventures to sustain themselves. St. Paul, Minn., may hardly be the tropics, but it remains a business hot spot as 3M keeps reinventing itself. The company insists that at least 25% of sales come from products less than four years old. 3M also fosters new ideas by telling company scientists to spend 15% of their time on whatever they like. No wonder Post-it notes and Thinsulate were invented there.

THINK PATTERNS, NOT RULES.

Complexity theory helps locate patterns and relationships that may not be obvious, including a variety of organizing principles. Michael McMaster, author of The Intelligence Advantage: Organizing for Complexity, cites the case of AMEC, a North Sea oil exploration contractor building offshore platforms worth billions of dollars. AMEC used complexity theory to understand what customers wanted and to change its bidding priorities. Emphasizing an unexpected combination of ideas—including sharing gains, developing oil company personnel, and beating the time budget—AMEC "went from a string of failures to consistent winning of bids," Mr. McMaster says.

REWIRE YOUR ORGANIZATION FOR INNOVATION.

Complexity theory recommends self-organizing information networks, rather than ones that impose structure from the top down. Self-organizing networks may sift data in unexpected ways, adding to an organization's creativity. Xerox, for instance, has added a way for its service technicians to provide each other feedback. Before, service tips used to take months to compile and propagate; now, a mediated Web site immediately posts the latest fixes. Additionally, the new site is the hub of a vast virtual nervous system that can quickly discern patterns in Xerox products' fitness in varied market niches, perhaps finding that cold weather may cause problems for one model but not another. John Seely Brown, the firm's chief scientist, calls this approach using "the social mind of an organization."

In addition to different ways of thinking, complexity theory offers new mathematical techniques for business analysis. Among them is:

CULTIVATE GENETIC PROBLEM-SOLVING.

Using genetic algorithms, computers can actually develop solutions for otherwise intractable problems involving lots of resources and constraints. They start with a possible solution. Then they have the program mutate, much as genes do. The computers test the mutations to see if they're moving toward a better solution or further away. Over time, through trial and error, the computers produce a better solution than is possible using standard algorithms. At IBM's new Institute for Advanced Commerce, scientists use genetic algorithms to test the future of Web-based companies. So far, the only winners have been those able to carve out niches of their own. Otherwise, the companies created for the simulation stage fierce battles for dominance, leading to price fluctuations reminiscent of service- station price wars.

In general, complexity's developers say that executives need to go beyond their tendency to focus on a few key variables and look more at the behavior of whole systems. After all, introducing a product is more like changing the way you participate in a business ecosystem than like lobbing a stone and calculating where it will land.