Do Targeted Discount Offers Serve as Advertising? Evidence from 70 Field Experiments

Abstract

T he prevalence and widespread usage of email has given businesses a direct and cost-effective way of providing consumers with targeted discount offers. While these discounts are expected to increase the demand for the promoted products, are they effective in increasing revenues? Do they have effects beyond acting as price reductions? We study these questions using individual-level data from 70 randomized experiments run by a large online ticket resale platform. We estimate the redemption rates of the offers and also measure the broader impact of emailed promotions by comparing purchases by individuals who received the experimental promotions with purchases by those who did not receive the offers because of the experimental randomization. We find that the offers cause the average expenditure to increase significantly, by $3.03 (a 37.2% increase) during the promotion window. However, the redemption rate of these offers is low. Importantly, 90% of these gains are not through redemption of the offers. The individuals who spent more on the platform in the past are more responsive to the offers, and the effect of the offers is significantly higher among individuals who did not transact on the platform in the year before the offer was given. Interestingly, the offer causes carryover to the week after the promotion expires; we find that spending increases by $1.55 in the week after the offer expires. Additionally, we find evidence for cross category spillovers to nonpromoted products: offers not applicable to a ticket genre cause an increase in spending in that genre. We conclude that emailed offers can serve as a form of "advertising" for the firm's products, in addition to being tools for price discrimination.

Experimental Designs and Estimation for Online Display Advertising Attribution in Marketplaces

“…This strategic effect significantly moves the net campaign effect (lift D=1 Camp = −18 88% −9 15% 2 62% ). Similar spillovers across product campaigns have been detected before by Sahni et al (2015) in the context of email coupon promotions. Note that the user selection effect of this CPM campaign significantly contributes to the negative presence effect.…”

The Economic Consequences of Partisanship in a Polarized Era

“…Of these 44 individuals, 37 informed us that they were ultimately interested in completing the transaction (2.2% of the sample and 84.1% of those who initially replied to the offer). Although these percentages seem small, they are actually quite high given the 0.1% transaction rates typically found in email solicitations per the digital marketing literature (Sahni, Zou, and Chintagunta ).…”

Changing perspectives: Using graph metrics to predict purchase probabilities

“…However, studies on online couponing are scarce in the literature and we are not aware of research considering its cost setting (e.g. Sahni et al, 2016).…”

“…Companies use conditional incentives regularly in the form of discounts and the most salient applications have attracted much research, e.g. customer retention (Ascarza & Hardie, 2013;Backiel et al, 2016) or coupon targeting Sahni et al, 2016). Because the treatment cost is conditional on the customer action, the uncertainty about the customer action translates into uncertainty about the realization of the cost of the incentive.…”