The ultimate measure of success for any marketing effort is sales figures, purchases, or orders. Since the marketer or researcher tries to get an idea of sales potential before launching a campaign, or in a more isolated manner (attributing an increase in sales to a specific marketing act), purchase intent is used as a predictor of purchase.
Purchase intent should not be considered solely at face value. Although a few studies demonstrate that measures of purchase intent hold some predictive power (Jamieson & Bass, 1989; Stapel, 1971), the only widely-accepted inference is that consumers who report intentions to purchase a product possess higher actual buying rates than consumers who report that they have no intention of buying (Berkman, 1978).
Due to the elusiveness of the concept, many researchers tried to create scales that translate “purchase intent” into “purchase” in order to get a more accurate view of potential sales. Morrison (1979) created a 0 to 1 scale of purchase probability based on purchase intent, and described some systematic discrepancies in purchase intent: the 1.0 group (100% purchase intent) provides no information and should be ignored due to a suspected “yes sayer” effect; the 0.8 group is the more revealing group that should be regarded as potential buyers. These systematic discrepancies vary greatly between industries (Morrison, 1979), thus limit model’s generalizability.
In the online world, however, large amounts of field data are available and used to gauge purchase intent and make potential sales inferences. These inferences remain a challenging task since the online world is connected to the offline world, and many browsing sessions to an e-commerce website lead to an offline purchase that cannot be linked to the browsing session. Another problem in identifying the sales funnel can occur when a visitor performs an exploratory visit from her computer at work, yet completes the purchase using her mobile phone or home computer, leaving the vendor with no way to link the two visits. Maxymiser (2012) Internet Research Agency claims that in 2012, 82% of shoppers have abandoned their shopping baskets; on the other hand, McAfee (2009) Internet security giant claims that 65% of “shopping cart abandoners” returned within two days to complete the purchase. Identifying these returning customers is only possible if the online retailer employs a sophisticated system of visitor tracking through cookies, and only when all visits to the store are made using the same computer and without deleting cookies in-between visits. Thus, deciphering the exact path consumers go through on their way to an online or off-line purchase remains a challenge.