It's easy to talk about innovation, reinvention, vision, and all the other business concepts that are supposed to offer new solutions to old problems. You just need a novel approach to a problem and you'll bypass whatever has held your company back.

But there's a problem that these variations on a theme don't address. To do something novel, you need to break out of old patterns, associations, and habits. Over time, we all become like carpenters who have become expert with a hammer and handsaw, but who may never have laid eyes on routers, power saws, nailing guns, and other tools that can help speed construction. It's fine to talk about making smarter decisions, but how do you do that until you learn to think differently?

A McKinsey Quarterly article by McKinsey consultants Olivier Leclerc and Mihnea Moldoveanu has an interesting take. The two discuss "flexible objects for generating novel solutions," for which they've coined the term flexon. Flexons aren't meant to be a magic solution to make problems go away. However, they are strategies that can at times help uncover and illuminate problem-solving approaches you may not have previously considered.

1. Use your networks.

Every part of your business involves a network of people. There are people who are thought leaders, others who are receptive to ideas, those who produce products and services, and those who consume them. Think of these networks as a graph, in which there are points, or nodes, connected by lines. The lines would represent such factors as trust, friendship, working partnerships, and influences.

Leclerc and Moldoveanu offer an example of a pharmaceutical company that wanted to know which physicians and researchers could best influence doctors about a new drug. By identifying a group of doctors who had co-authored papers in the appropriate area of medicine, the company could essentially find clusters of like-minded doctors who could then further influence readers of medical journals. This approach could work well in almost any area of marketing and sales, particularly with the data available on social networks to help pinpoint influencers.

2. Let it evolve.

The success of a product line, service, brand, or marketing strategy can hinge on any number of different factors. Testing all of them is unrealistic, and a roll-the-dice bet is risky. Instead of trying to find the optimal solution in once step, use an evolutionary approach. Break the problem down into a series of tests, each one focusing on a factor that seems important. Over time you cull out the dead ends and advance.

Samsung has used this strategy brilliantly. The company didn't come up with the idea of its popular Galaxy product line overnight. Over time, the company tested different models and features to learn what consumers liked. If you were building a new menu for a restaurant, you might try some representative dishes with a sampling of your best customers to see if there is a type of cuisine, level of spice, or set of ingredients that make them respond most, using a series of invitation-only test meals.

3. Identify the decision agent.

This approach looks at the dynamics of interactions among individuals, each trying to get the best outcome. It's like game theory, but more general because you don't assume that all the parties are acting in totally rational ways.

Also, rather than assume you're necessarily talking about people, you can consider groups and organizations as well as individuals. Then you model the interactions as a set of competitions and cooperation among all involved:

You ascribe to them beliefs and motives consistent with what you know (and think they know), consider how their payoffs change through the actions of others, determine the combinations of strategies they might collectively use, and seek an equilibrium where no agent can unilaterally deviate from the strategy without becoming worse off.

The decision agent approach lets you consider a problem in terms of the relevant players, what they're looking for, and how they behave. It may be that with the proper appeal to the right entities you can find an alternative solution to a problem. An auto repair shop might be unable to get the parts necessary to fix a customer's car. The customer may be highly emotional and difficult to deal with because he needs his car to get to work. The parts shop might be waiting for a back-order. Do you beat your head against a wall trying to make the vendor move faster than it can, find another vendor, offer a loaner to the customer, or some combination?

4. Look at the system dynamics.

In engineering, you can get great value from describing a design problem as a complex system and looking at the interactions, causes, and cascading effects. There is an ultimate result, but you may have a wide choice of ways to affect that result.

A simple example from manufacturing could be addressing a lack of product inventory necessary to close sales. You could be constrained by getting parts in house, the type of manufacturing equipment you have, issues in distribution from the factory to stores, or bad demand forecasting. Looking at this interconnected system, you can identify where the biggest problems are and then see what different combinations of changes might do to availability.

5. Parse your information processing.

This isn't an analysis of IT per se, but instead examines the movement of information through the company. Look at who creates information, who uses or transforms it, or where it gets bottled up by employees. If there are problems with decisions, it may be that you need to examine these flows of information, making sure that everyone is getting the full set of information they need to operate.

None of these approaches is simple to implement. However, investigating the concepts may get you out of a jam and onto the road toward success.