Bibliographische Detailangaben
Beteiligte: Anderson, Eric T., Simester, Duncan I.
In: Marketing Science, 20, 2001, 2, S. 121-142
veröffentlicht:
Institute for Operations Research and the Management Sciences
Medientyp: Artikel, E-Artikel

Nicht angemeldet

weitere Informationen
Umfang: 121-142
ISSN: 0732-2399
1526-548X
veröffentlicht in: Marketing Science
Sprache: Englisch
Kollektion: sid-55-col-jstoras4
sid-55-col-jstorbusiness1archive
sid-55-col-jstorbusiness
JSTOR Arts & Sciences IV Archive
JSTOR Business I Archive
JSTOR Business & Economics
Inhaltsangabe

<p>We analyze data from a variety of sources, including historical data from a women's clothing catalog, a field study in that catalog, survey responses to catalog stimuli, and grocery store data for frozen juice, toothpaste, and tuna. The analysis yields three conclusions. First, sale signs are less effective at increasing demand when more items have them. Second, total category sales are maximized when some but not all products have sale signs. Third, placing a sale sign on a product reduces the perceived likelihood that the product will be available at a lower price in the future, but the effect is smaller when more products have sale signs. By ruling out alternative hypotheses, the findings suggest that moderation of the sale sign effect is in part due to reduced credibility when they are used on more products. The credibility argument is motivated in part by a recent paper that presented an equilibrium model predicting that customers who lack knowledge of market prices rely on point-of-purchase sale signs to help evaluate posted prices (Anderson and Simester 1998). The model predicts that sale signs increase demand but that the increase is smaller when more products have them. This moderating effect regulates how many sale signs stores use and makes customer reliance on these cues an equilibrium strategy. To evaluate whether the number of sale signs moderates their effectiveness we compare demand for items with sale signs when varying the number of sale signs on other products. For this comparison we use two datasets describing demand for products in a women's clothing catalog. The first dataset describes customer orders for the same set of items across three sequential issues of the catalog. The second dataset is a field test conducted by mailing different versions of the catalog to randomly selected customer samples. Reassuringly, although the datasets do not share the same limitations, the findings are consistent and indicate that sale signs are less effective when more products have them. In addition to the credibility explanation there are at least two alternative explanations for this result. First, using a sale sign to make a product more attractive may lead to substitution of demand from other sale items within a store, and so demand may vary even if there is no change in the credibility of the sale signs. To discriminate between the substitution and credibility explanations, we evaluate whether total category demand is maximized when some but not all products in the category have sale signs. The credibility explanation implies that adding another sale sign will eventually decrease total category sales when many items already have sale signs. In contrast, if adding a sale sign to an item leads solely to substitution from other products within (or outside) the category we will not observe a decrease in total category demand. We test the total category sales prediction using grocery store scan data describing total category demand for frozen juice, toothpaste, and canned tuna. The findings reveal that in all three categories there is a significant reduction in aggregate demand. A second alternative explanation is that sale signs focus customer attention on products with these cues. Distinguishing between the attention and credibility explanations is a difficult task. They both imply that sale signs deliver less information when there are too many sale signs. The credibility argument predicts that this occurs because the signs are noticed but not believed. The attention explanation predicts that the signs are less likely to be noticed when attention is diluted by a large number of sale signs. To discriminate between the credibility explanation and this attention effect we use survey measures to evaluate a third hypothesis. This hypothesis predicts both that placing a sale sign on a product reduces the perceived likelihood that the product will be available at a lower price in the next period, and that this effect is smaller when more items have them. In an attempt to control for the attention effect we focus subjects' attention on a set of focal items. Under these conditions it is unlikely that subjects overlook sale signs on these items. More important, the likelihood of overlooking sale signs on the focal items is unlikely to depend on how many other items also have sale signs. The results confirm both that the presence of a sale sign reduced subjects' expectations that an item would be available at a lower price in the future and that this effect is smaller when more items have sale signs.</p>