Geordi is head of marketing for a new product introduction code named:
| Geordi: | According to the computer simulations, by running these ads at the target 18-24 year old group, we can expect an initial response rate of 19.2 percent. But we’re also interested in repeat purchases. By extrapolating the statistics from our experiences with the big-sky product, we anticipate repeat buying of two items a month by 5 percent of the target group. If the product gains momentum, long-run market share should stabilize around 35 percent. At this point, there is insufficient data to determine whether the product will be fully accepted. |
| Tasha: | In other words, we’re supposed to risk the entire project based on these numbers generated from the computer. I don’t like it. |
| Deanna: | Actually, we believe the values are low. In my discussions with the participants after the study, I sensed that they were much more enthusiastic than the numbers indicate. |
| Geordi: | In fact, Tasha, we deliberately chose conservative values to use in the simulations. If Deanna is right, this product could be hot. We might even show a major profit in the first six months. |
| Tasha: | But, that’s all speculation. This model has no relationship to reality. I could just as easily find 10 people who hate the product. I was talking to my niece the other day, and she says it’s revolting. |
| Deanna: | Now Tasha, you know we can’t make decisions based on the responses of one person. We scientifically selected the participants . . . |
| Tasha: | That was just an example. I’m much more concerned about these computer projections. I don’t understand how the computer produced these results. Instead of wasting time with simulations, I think we should talk to real people. |
| Geordi: | If you want, I can print out the computer model equations for you. But they’re from state-of-the-art market research, and they might be hard to follow. Plus, last time I checked, there were 50 pages of equations. |
| Tasha: | Are you saying that consumers have to compute 50 pages worth of equations to decide whether they want to buy this product!? |
| Geordi: | No, no. The equations are just used by the computer to describe how customers behave. |
| Tasha: | But how do we know the equations are right? Are you saying the computer can tell you whether I will buy the product? |
| Deanna: | No, of course not. But we don’t need to predict responses of individuals. We focus on how the entire target group will respond. |
| Tasha: | We don’t seem to be getting anywhere. I still don’t see how we know whether the computer is right. If it’s wrong, we lose a lot of money. I don’t see how the computer can analyze every possibility. For instance, what happens if the product appeals to the 15- to 18-year-old group? Won’t that create a negative status effect on our target group? |
| Geordi: | We talked about that possibility early on, but we don’t think it’s important enough to include in the model. There are hundreds of minor possibilities, but we don’t have time to include all of them. |
| Tasha: | In other words, your model only includes things you thought were important. That’s even worse. Why didn’t you just make up the sales projections, instead of hiding them behind some phony model? |
| Deanna: | I sense that we have reached an impasse. Perhaps we should take a break and come back later. Geordi, perhaps you could prepare a small demonstration of the model. And Tasha, remember that we still have to make a decision. We can’t always predict the exact future; sometimes even a little information is better than none. |
Questions