How Long Should You Give Them?

The last time your salespeople met, they looked to you for guidance about product demonstrations. They wanted to know how long they should allow customers to try your products, and you directed them to keep track of what they were doing and get the data back to you. You planned to create a chart comparing the proportion of customers who purchased the product with the length of time they were given in a demonstration. You expected to find that people given short trial periods would be less likely to purchase the product than people given longer trial periods. And you expected your chart to reveal a peak at the optimum demonstration period, when the largest proportion of customers purchased the product.

Unfortunately, that's not what your chart showed.

Instead of what you expected, your chart revealed two clear peaks, and the line looked like a 2-hump camel. It started low, rose twice, once at two hours and once again at twenty-four hours, and then finished low. What are you going to tell your salespeople?

If you've ever carried out this analysis in your business, then the chances are good that you've already seen this 2-peak pattern. It's a common finding. But it took some hard thinking and a couple of experiments to explain its cause. Amir Heiman, of the University of California, did the work. Here's what he learned.

Demonstrations facilitate sales for high-quality products because they correct mistakenly low impressions of the products' quality and usefulness. But when low impressions of a product aren't mistaken, then demonstrations will hurt sales. So demonstrations should be reserved for high-quality products.

Demonstrations serve to teach and to convince customers that a product is right for them. Demonstrations initially display a product's most prominent features and customers experience for themselves their quality and usefulness. As time passes and the demonstration continues, less prominent features are noticed. For example, a short demonstration of a flashy sports car may introduce customers to the car's appearance and handling on the road, while a longer demonstration may call attention to the uncomfortable seats and small trunk.

Heiman reasoned that the optimum length for a demonstration should fall just after prominent, positive features are experienced and just before less prominent, negative features are encountered. But he also expected this optimum demonstration time to vary depending upon the knowledge a customer brought to the demonstration.

Customers with little product knowledge would require longer demonstrations to reach their optimum purchase probability. Customers with a great deal of product knowledge would quickly check off positive features they expected to find and would notice negative features much sooner.

Heiman conducted an experiment in Tel Aviv with an Italian luxury sports car, the Lancia Thema. He collected data on trial period lengths and purchase probability, and he found the same two-peak pattern noted above. But when he factored in customers' product knowledge, the two peaks were explained. The peak for knowledgeable customers was two hours. The peak for novice sports car buyers was twenty-four hours.

Heiman recommends that we divide our customers into two groups based on their product knowledge, and analyze their purchase probabilities and demonstration lengths separately. Armed with this insight, salespeople can ascertain a customer's product knowledge and then offer a demonstration period most likely to result in a sale. And Heiman advises us not to be surprised if we find this period to be much shorter than we customarily offer. That's what he found in Israel. He also believes his findings apply to all kinds of product demonstrations and to sampling of consumer goods.

Reference: Heiman, Amir and Eitan Muller (1996). Using Demonstration to Increase New Product Acceptance: Controlling Demonstration Time. Journal of Marketing Research, 33 (November 1996), 442-430.

Back to home page