Chapter 2

Research Design

Building on the aforementioned definition of ‘H’ and ‘T’ products, the research question is enounced in this chapter. How can e-tailers drive profitable growth through smart pricing? Then, how should e-tailers respectively price Head products (whose prices consumers tend to compare and remember) and Tail ones (whose prices are often neglected and with a few discoverable alternatives)?

In other words, as smart pricing means ‘starting with the customer’, the dependent variable studied is the purchase likelihood in responce to a price variation, which is supposed to be influenced by customer Price Awareness ( H p 1 ), and the presence (or absence) of Comparison Effect ( H p 2 ). As different degrees of Price Awareness (PA) and the Comparison Effec (CE) define Head (H) and Tail (T) products (as described in the previous section), they are hypothesized to be key proxies to identify the different sensitivity to price in each case. This findings could be eventually used to craft the most appropriate pricing approach to adopt over hits (H) and niches (T).

This section better identifies the study purposes. In the next parapraphs, the research question is addressed, and the related objectives are enounced. A better understanding of the study contributions and the research model is provided. In particular, the concepts of Price Awareness (PA), Comparison Effect (CE), and customer response to price (PS) are investigated, drawing fully from existing literature.

This premise will ease the explanation of:

1. the relationships between the independent variables and the dependent one explored by the existing literature so far;
2. the possible analogical extension of these relations to this thesis model ( H p 1 ,H p 2 ).

2.1 Research Question

This study precisely aims at building a bridge between Head and Tail Theory and Pricing Strategy, posing the following questions: how can e-tailers drive profitable growth through smart pricing? Then, how should e-tailers respectively price Head products (whose prices consumers tend to compare and remember) and Tail ones (whose prices are often neglected and with a few discoverable alternatives)? This question has been further framed in two hypotesis, in order to translate concepts into measurable variables.

The finishing line that this thesis wishes to cross is the identification of a balance between customer satisfaction and profitability preservation (see 4.4 Managerial Implications). Customer satisfaction – eventually translating into retention – has to be maintained meeting price expectations. Profitability preservation depends on how accurately items driving price perception are identified, and eventually attractively priced. To the research hypotesis, price sensitive behavior is more likely to take place when evaluating the purchase of a hit (Head product), while the contrary is expected for niches (Tail products). In other words, this study also wishes to make the extra-mile to translate these different approaches into a virtuous cycle, where:

1. attractively priced Head items bring sales,
2. positive price perception generates customer satisfaction,
3. customer satisfaction evolves into retention,
4. customer retention generates additional sales on Tail.

Please note that, naturally, this cycle can even start from the Tail.

As anticipated in the paragraph 1.3 How long is the Long Tail?, consumers may be attracted to an e-tailer because it is able to offer niche, hard-to-find products. The demand for niche products could have a spillover effect on the demand for hits, and vice-versa.

This research is indeed supported by a statistical analysis (Logistic regression), aiming at:

Figure 2.1: Translate Pricing into Long-term value: strike the balance between profitable growth and customer expectations

• diagnosing price-sensitive and non-price sensitive behaviours in different experimental scenarios, designed to fit the Head and Tail purchase cases;
• accordingly identify the most congruent pricing approach for each product-role,
• unlocking the aforementioned benefits (Figure 2.1).

Before moving to the analysis, a better understanding of the study contributions and the research model is needed. In particular, the concepts of Price Awareness (PA), Comparison Effect (CE), and Sensitivity to price (PS) are investigated, drawing fully from existing literature.

Contribution to existing background

As mentioned in section 1.3, in the context of the Long Tail Theory, the previous research mostly focused on the sale concentration changes. The Internet channel exhibits a significantly less concentrated sales distribution when compared with traditional channels (Brynjolfsson et al., 2011)[14]. Netessine and Tan (2009)[94] suggest that an accurate detection of the Tail thickness can better help e-tailers to evaluate how worth it is to carry its low-selling selection. Lee (2010)[47] analyses Head and Tail roles to develop ad-hoc marketing strategies.

However, none of the previous studies struggle to find a pricing pattern to adopt.

In the context of the pricing literature, research has generally focused on routine decisions in response to price format (e.g., Dhar & Hoch, 1996)[22], or price framing (e.g., Lichtenstein & Bearden, 1989)[51]. Although still in an embryonic phase, Rao (1984)[80] was the first talking about ‘smart pricing’. Rao’s study generated a growing awareness of the need to increase focus on customer-related factors such as customer satisfaction (Oliver 1999)[71], customer service (Parasuraman and Grewal 2000)[73], customer loyalty (Kumar and Shah 2004[45]; Reichheld 2001[81]), and quality as perceived by the customer (Boulding et al. 1993[11]; Rust, Moorman, and Dickson 2002[83]). However, to my knowledge, none of these studies properly bridges the Long-Tail Theory with pricing literature.

Surely, none of them drives related considerations about the relation of Price Sensitivity with H- and T- product features (Price Awareness, Comparison Effect). What’s more, none of the previous studies is based on a comprehensive data set collected through a realistic simulation, which could simultaneously capture the customer behavior naturalness and guarantee an optimum sample size in terms of statistical representation of the results.

It is worth to restate that the context the study refers to is the online environment, characterized by peculiar trends, other than those prevailing in the physical, largely studied brick-and-mortar retailing (as suggested in 1.2 Long Tail Theory at market level). Even though the recent literature counts of a not negligible number of studies about consumer habits and the Internet, none of them look at the topic from the managerial perspective of striking the balance between customer expectations and profitability preservation.

In detail, this study broadens and enriches the existing literature background because:

• it closes the gap between the stream of research of “Long-Tail Theory” and the established pricing literature;
• it embraces a cross-category view, resulting in a widely fitting approach for online retailers;
• it uses an ‘ad hoc’ data collection setting designed to maximize the realism of an online purchase experience (see 3 Methodology).

Before translating the above-mentioned point into structured hypothesis, the variables of interest – respectively, Price Sensitivity, Price Awareness and Comparison Effect – are debriefed. For each of them, a theoretical background is provided.

2.1.1 Customer Sensitivity to Price

According to previous discussion, Price Sensitivity (PS) has been identified as one of the key factors affecting company’s pricing choices as well as its ultimate profitability. It refers to the extent to which individuals perceive and respond to changes or differences in prices for products or services (Monroe, 1973)[65]. Price Sensitivity can also be interpreted as the reaction of the consumers to what they perceive about the cost within which they will buy a particular product or service. Each customer will have a certain price acceptability range and different customers have different limits in their perceptions of what price is within their ranges.

Generally, economists are familiar with the concept of price elasticity. Price Elasticity of demand is the is the percentage change in the quantity demanded of a good or service divided by the percentage change in the price (Marshall, 1890)[58]. If the changes in price have a proportionately greater impact on demand for a product, then it is known as elastic demand. On the other hand, inelastic demand narrates the situation where changes in price have a negligible effect on the demand.

Price Sensitivity has also been approached as willingness-to-buy in response to price changes (Gabor and Granger, 1966)[30], with the focus shifting from the change in quantity purchased (as a reaction to price changes), to the measurement of an ideal price to attribute to a product (especially when introduced for the first time in a certain market). Then, price sensitivity can be tested subjecting the respondents to different proposed price increases for the same item. The respondent will then accept or not the increased price, according to its sensitivity. According to Gabor and Granger (1966)[30], the results can be eventually used to produce a demand chart (where x-axis are the prices and y axis the percentage of people willing to pay that price) and a revenue curve(where y-axis is the percentage of optimal revenue and x-axis is still price). For the purpose of my study, the binary response provided by the customer in the experimental simulation (accepting or not the price increase) will be used to design a model predicting the purchase decision likelihood in response to a set of variables.

The marketing studies using Gabor and Granger (1966) approach are several. Goldsmith and Newell (1997)[31] measured price sensitivity in relation to customers’ behavioural features. They found shopping innovators to be less price sensitive than laggards. However, it is generally true that customer’s evaluation of the value of a good or service is based on their perceptions that what they receive and what they expected of having it (Monroe 1990[66]; Zeithaml 1998[74]).

This means that, when making pricing decision, adopting the customer perspective as starting point allow to go beyond than just the product cost. Managing price perception, not just pricing structure and actual price points, thus has become a critical capability for e-tailers.

Given the availability of information for both customers and competitors, prices are typically lower online (Shankar et al., 1999)[84]. In spite of the dominance of very low prices online, even price sensitive customers do not always purchase at the lowest price on every single item (Smith and Brynjolfsson, 2001)[88]. In 1999, Lynch and Ariely[56] had already conducted experiments in a simulated online wine store to test this hypothesis. They found that price sensitivity was lower under experimental conditions where buyers had information on both price and product quality as compared to conditions where they had information only on price. Researchers also found that – when evaluating partitioned prices – consumers focus more on the price of the component that constitutes a smaller proportion of the total price (Chakravarti et al. 2002)[19]. Also, customers become less sensitive for hedonic products or services consumed in hedonic occasion (Wakefield and Inman, 2003)[102]. Loyal customers are price insensitive to the price changes while non-loyal customers are sensitive in making decision about a brand (Yoon and Tran, 2011)[104]. A customer’s price sensitivity can be significantly explained by a customer’s price perception. Perception is the process by which people translate sensory impressions into a coherent and unified view of the world around them (Munnukka et al. 2014)[68]. How customers perceive the price is as important as the price itself. Even if customers fail to notice specific price moves in isolation, companies should make sure customers have a good sense of how the firm’s prices compare to those of competitors. According to a survey conducted by Bain & Company and ROI Consultancy Services (2016) on a sample of 2,200 consumers in Atlanta and Washington, DC, retailers can get either more or less credit for their pricing than actual shelf prices would suggest. The study involved eight retail chains carrying groceries. Already back in 1984, Kahneman and Tversky[40] had shown that people perceive less positive utility with a gain of – say – 10€ than negative utility with a loss of the same amount. This general conclusion about loss aversion has many times over the years been shown to have a powerful impact on decisions made in diverse situations, also in retailing.

In their overview of empirical reference price research, Kalayanram and Winer (1995)[41] also found support for demonstrating that consumers’ reaction is stronger to price increases than decreases. This notion is also known as ‘asymmetric consumer price response’. Consumers are thought to interpret price variations as gains or losses relative to their internal reference price.

Table 2.1 collects some key finding previously enounced in this paragraph body, before giving way to some literature review about the relation between customers’ responses to price fluctuations and related pricing approaches.

In order to provide sound managerial advices, buyers’ responses to prices have received a great deal of scholarly attention. As previously enounced, research has typically focused on routine decisions in response to changes in price (Bucklin, Gupta, & Han, 1995)[15], price format (Dhar & Hoch, 1996)[22], or price framing (Lichtenstein & Bearden, 1989)[51]. Liechty, Ramaswamy, and Cohen (2001)[53] examine price sensitivity relative to light alternative information services offered on web-based menus. Steenburgh and Avery (2011)[90] demonstrate that, generally, the higher the absolute value of the price, the more sensitive customers are to price changes. However, none of these studies investigates across the relation of price sensitivity and H/T roles.

Table 2.1: Customer sensitivity to price according to some research (1890 – today)