A new book, Secrets of The Moneylab, chronicles experiments that companies are using to guide decision-making and boost their bottom lines.

Most companies of any size are well-versed in market research techniques. Perhaps they’ve held focus groups to get participants’ input on a proposed new product or conducted a phone survey to determine consumers’ thoughts on a particular firm’s reputation. However, far fewer test how people actually behave when money is involved. Conventional wisdom long has held that humans act logically to better their own interests and pocketbooks – the rational man theory of economics.

Of course, the past few years have shown that humans often act anything but rationally. A case in point: Companies that made loans to individuals who were likely to default. Go back earlier, to the late 1990s, and you’ve got Internet firms with no revenue and flimsy business plans attracting millions in funding.

These and other, often lower-profile examples, such as sales incentives that fail to properly motivate, cost companies money — often, lots of of it. With this in mind, a handful of businesses are conducting laboratory tests to examine how individuals and groups tend to respond to various business propositions. They’re using the findings of these experiments, rather than just relying on assumptions embedded in forecasts, to guide decision-making. The results show promise in areas ranging from contract negotiation to sales and inventory planning to new product launches. A number of these tests and their applications in business are chronicled in the recently released Secrets of the Moneylab: How Behavioral Economics can Improve Your Business, by Kay-Yut Chen, lead economist with Hewlett-Packard’s labs, and science writer Marina Krakovsky (Portfolio Penguin, 2010).

Here are a few of the experiments that Chen and others have conducted:

Testing the notion of reciprocity: When we feel that we’ve been treated fairly, most of us want to act in kind. Conversely, the urge to retaliate when we’ve been treated unfairly seems pretty much hardwired into us, even if we don’t always act on it. What’s more, it appears that this tendency is even more pronounced than the tendency to reward generous behavior.

Consider this experiment conducted by University of Chicago researchers. Participants were paired with another player called the Dictator – actually a computer – that would (not surprisingly) dictate the rules. In the first go-round, the Dictator was programmed to split $100 evenly between itself and the participant, with each getting $50. Then, roles reversed and the participants had to decide how to split the $100.

Here’s the catch. With one group of participants, the experiment started with the Dictator holding the $100, and then giving each player $50. With another, the players possessed the $100, and the Dictator took $50 from them. So, while both groups were ahead by $50 after this round, the way they got there varied.

So, what happened when the roles reversed? Those who had received $50 from the Dictator’s stash split the money almost evenly, keeping $50.50 and handing over $49.50. Conversely, those from whom the Dictator had taken the $50 gave less – $42, to be precise.

In a follow-up experiment, some Dictators again split the money evenly, while others would, with the participants who already had the $100, take $30, leaving the participants with $70. Despite the fact that the participants with $70 clearly were better off than those with $50, they judged these Dictators more harshly than the participants that walked away with less from the other Dictators.

The take-away for business: When people decide how they’ll respond to another’s actions, they take into account not just what happened, but also what they believe was the other’s intention. Of course, it’s rare that anyone can be absolutely sure of another’s intention or of the complete context in which decisions are made, which is why actions so often are misinterpreted. Here, taking money was seen, not surprisingly, as more selfish than giving, even when the taker left the participant better off than the giver. This is one way of showing how two parties to a dispute can have completely opposing views of a proposed resolution.

Examining the consequences of anchoring: Say you’ve got to calculate just how many units of a product to order when demand fluctuates widely and randomly. You don’t want to be saddled with unsold inventory, but you also don’t want to run out of goods and lose potential sales. This formula can minimize those scenarios and optimize profits:

((price – cost)/price) x maximum demand

Any business student should know this. However, when this problem was put to several groups of them, they seemed not to use the formula. Two groups were asked to place orders for widgets for which demand could range from 0 to 300. All the widgets were priced at $12, while the cost was either $3 or $9, depending on the group. With a cost of $3, the optimal order was 225, or (($12-3)/12) * 300; when the cost rose to $9, the optimal number dropped to 75 ((12-9)/12 * 300.

The students, however, tended to place orders near the average demand of 150 units. In the experiment, participants whose product cost $3 placed orders for 178, while those whose product cost $9 ordered 140, on average.

One likely reason that their orders fell short of what was optimal: anchoring, or the tendency of most people to base their answer on a handy reference point – in this case, average demand. Even though average demand has little relevance when demand fluctuates widely and randomly, it’s a starting point.

Anchoring comes into play in most negotiations, as the first number thrown out tends to influence future numbers. That’s why, the authors assert, contrary to conventional wisdom, it pays to make the first move when you’re doing the negotiating.

The book contains a number of other examples showing how laboratory experiments can provide insight to guide real-world business decisions. The experiments show how humans often act in ways that confound others. By having some idea of how others – customers, competitors, and business partners – might respond to a given situation, your firm can benefit.

Of course, the authors point out that not every decision warrants a laboratory experiment. However, for those that do, the proper one can yield insight that often saves companies significant amounts of money or points them to actions that will allow them to make more. ###