Marketing Response Modelling: Impact of Digital Marketing for a Luxury Car Brand

Abstract

The luxury car segment is the most vibrant segment in the luxury goods market, experiencing high growth in recent years in the emerging economies of China, India, and Brazil. In India, the luxury car segment is dominated by three major players, that is, Audi, Mercedes-Benz, and BMW, together accounting for 85 per cent of the total Indian luxury car segment.

The study proposes a marketing response model for luxury car brands, involving a linear model with all possible interaction effects. The model is applied in the case of a luxury car brand which had recently adopted digital marketing in addition to its traditional advertising media mix. The response in the form of customer queries at its showroom (situated in Bangalore, India) was taken as the dependent variable. The independent variables were the advertising expenditure in different media, viz. newspapers and magazines, display events, and digital media.

The results of the model provide a measure of the effectiveness of each of the media, the interaction between them, as well as the impact of digital marketing.

Keywords

Marketing response model luxury cars advertising media mix digital media interaction effects

Introduction

Traditionally, luxury brands are described as ‘goods that bring prestige, apart from any functional utility’ (Marinov & Marinova, 2011) and are often perceived as ‘emotional products’. Luxury brands have for several years been outperforming the rest of consumer goods (Trovato, 2014). It has been reported in numerous researches that there has been an increase in people’s need for appearances and materialism, which has, in turn, led to an increase in the consumption of luxury brands. There is an advantage in dealing with luxury brands that they can fund their own growth as luxury brands generate healthy cash flow for the business. Hirschman and Holbrook (1982) and Sheth, Newman, and Gross (1991) reported that certain goods and services have been known to possess emotional value in excess of their functional utility. Luxury cars fit into this description as they offer sensory pleasure, aesthetic beauty, or excitement. Advertisers have long been promoting the emotional responses expected from the use of luxury motor cars. For instance, BMW has been using ‘Sheer Driving Pleasure’ as its main slogan for many years (Vigneron & Johnson, 1999).

The growth of the automotive market has remained quite stable in the recent past and is expected to grow faster due to the rise in demand from the emerging economies of China, India, and Brazil. Some of the driving factors for the growth of luxury goods markets in these markets include rising disposable income, changing demographics, and economic growth. Also, the rise in tangible luxury offerings in vehicles, shifting consumer preferences from sedans to SUVs, and the growing trend of electric luxury vehicles across the region is fuelling the demand for luxury cars (Marketwatch, 2019). The luxury car market is anticipated to register a Compound Annual Growth Rate (CAGR) of about 5.83 per cent during the period 2019–2024. The emerging markets have experienced the highest rise in the demand for luxury cars, surpassing the demand in Europe. The demand in Europe declined for all the three major automakers, that is, Mercedes (2.3 per cent↓), BMW (0.3 per cent↓), and Audi (14 per cent↓) in 2018.The Asia-Pacific region is expected to witness the fastest growth rate during the period 2019–2024. China became the world’s largest automotive market, surpassing the US automotive market and maintaining its leading position since 2009. China is the new focus for automotive manufacturers in all segments; passenger and luxury that makes up nearly 30 per cent of global vehicle sales. In 2017, luxury cars in China accounted for more than 9 per cent of the overall passenger cars sales. 2018 has been a strong market year for India, Russia, and Southeast Asia. Table 1 highlights the 2018 car market position.

The luxury car segment is the most vibrant segment in the luxury goods market, with a very high revenue contribution. Some of the prominent luxury car brands currently available in the market include Mercedes S-Class, Range Rover, Rolls-Royce Phantom, Bentley Continental GT, Porsche Paamera, BMW 7 Series, Audi A8, Bentley Bentayga, Jaguar I-Pace, and Lexus LS (AutoexpressUK, 2019). Global sales of the major playersin 2018–2019 were as follows: Mercedes brand cars 2.31 million, BMW sold 2.12 million units, Audi sold 1.87millionunits.¹ The luxury car segment has undergone many changes in recent years. The German players, viz. BMW, Audi, and Mercedes-Benz, still dominate the luxury car segment with a combined share of around 80 per cent and are aiming at the luxury segment (Autonews, 2019). Luxury vehicles globally account for around 5–6 per cent of the overall market.

Table 1.

No. of Cars Registration 2018

Automotive Car Market	Country	Cars Registration (mn)
1.	China	23.2563
2.	United States	17.2152
3.	European Union	15.6245
4.	Japan	4.3912
5.	India	3.3947
6.	Brazil	2.4754
7.	Russia	1.8006

Source: Statista (2019).

The Indian automotive industry was revolutionized by the liberalization of the Indian economy in the 1990s that saw a reduction of automotive prices and import duties, and incentives for setting up of manufacturing and/or assembly plants and permitting 100 per cent Foreign Direct Investment for automotive Original Equipment Manufacturers (OEMs). Since then, India has emerged as an economic hub for automotive OEMs. It is an important link in almost all major automotive global supply chains, and many global organizations are looking at India as an assembling base. Also, India has become a leading automotive market, inducing players to enhance offerings in the market and increasing the number of dealerships across India.

The Indian luxury car segment began with the entry of Mercedes-Benz in 1995. Mercedes-Benz had a first-mover advantage in the segment, with practically no direct competition, and it remained at the top for more than ten years. BMW entered the Indian market in 2006, followed by Audi in 2007, inducing a realignment of market shares of the major brands.

The penetration rate of luxury cars in India is estimated to be 1.2 per cent (AutocarIndia: online). The luxury market in India is steadily growing in importance and has a great potential for sustained growth. Growing at a rate of 25 per cent per annum, and tripling its size in the last five years, the luxury car segment has become the fastest growing automotive segment. More than 40,000 luxury cars are being sold in India annually, and about 20 global luxury brands have entered the Indian market. In 2017, Mercedes sold 15,300 units, whereas BMW sold 9,800 units and Audi sold 7,876 units. The market for India is expected to register a CAGR of 23 per cent during 2019–2024 (Mordor Intelligence, 2019).

The major drivers of growth have been the increase in per capita income, the increase in aspiration levels, and the decrease in luxury tax rates (Ahuja, 2014). Also, the growth of high net-worth individual’s class in India has been amongst the highest in the world, attracting luxury brands to the Indian market. This ‘nouveau riche’ class has increasingly aspired for a luxurious lifestyle and are increasingly willing to spend extravagantly to sustain their lifestyle. The pre-owned segment of luxury cars is continually increasing owing to heavy depreciation. And this segment of luxury cars has been growing at an approximate rate of 35–40 per cent year-on-year (Mordor Intelligence, 2019).

Together, these three brands comprise around 85 per cent of the total Indian luxury car segment. 2018 figures for luxury cars remain 40,688 units, that is, 3 per cent more than 2017’s 38,989 units in all (AutocarIndia, 2019). Mercedes-Benz sold 15,538 cars (38 per cent), BMW with 11,105 (26 per cent) at the second spot, and Audi with 6,463 (16 per cent) cars at the third spot. Jaguar is closely following Audi with 4,596 cars (11 per cent) (Mordor Intelligence, 2019).

The sales response of a product depends on a variety of factors such as product features, pricing level, communication, and promotion efforts. Luxury cars demand in India is impacted by several factors including pricing levels owing to change in GST rates, liquidity pressure, demand for SUVs, brand power, and communication strategy involving advertisement, social, and digital media.

In the late 1940s, a marketing manager was termed a ‘mixer of ingredients’ by Borden (1964). The marketing mix includes product planning, pricing policy, branding, channels of distribution, personal selling, advertising, promotions, packaging, displays, servicing, physical handling, and fact finding and analysis (Cooper, 1983). The market forces that bear on the marketing mix include buyer behaviour, the trade’s behaviour, competition, and regulatory behaviour. This article focusses on few dependent and independent variables in the market response model. The independent variables involved in this study are advertising expenditure in newspaper and magazines, advertising expenditure in display events, and advertising expenditure in digital media. The dependent variable is the number of queries generated from the advertisement. An interaction effect is also studied in the model among these variables.

Literature Review

Consumer behaviour is an important study area for the marketers. The practice has evolved in the form of a market response model. Market response models are designed for the marketers to understand how consumers individually and collectively respond to marketing activities, and how competitors react to those responses (Hanssens, Leeflung, & Wittink, 2005). The pioneering studies in the econometric market response models were carried out from the 1950s, including some of the notable works by Dorfman and Steiner (1954), Bass (1961), Telser (1962a, 1962b), Kotler (1963), Rao and Lilien (1972), Lilien and Kotler (1983), and Parsons and Schultz (1990). Discussions on the response models appears in Lilien, Kotle, and Moorthy (1992), Leeflang, Wittink, Wedel, and Naert (2001), Hanssens, Parsons, and Schultz (2001), and Lilien and Rangaswamy (2003). Using market response models, marketers are enabled to capture the factors driving a market. Different variables are organized in the form of factors based on similarity and explanation capabilities of the variables. These variables focus on factors that affect performance called sales drivers. Response models are used in many industries. Hanssens et al. (2001) have presented a variety of static and dynamic response models to describe marketing response for stationary markets and econometric response models with single and/or multiple marketing time series for evolving markets.

The popularity of market response models has made them an accepted tool for marketing decision making across all industries. According to Hanssens, Parsons, and Schultz (2003), companies have relied on market response models to aid marketing decisions, including setting prices, allocating advertising expenditures, forecasting sales, and testing the effectiveness of alternative marketing plans. The market response analysis industry has progressed well beyond $125 million sector of marketing research industry. That essentially describes the scope and future of market response analysis as for marketers nothing interests more than knowing its customers to the core to make a sure and satisfying business with them. In other words, market response analysis is the key to sustainable business.

Response models use appropriate factors (instruments) to study the sales response, that is, the intention or willingness to purchase or the actual sales volume. The commonly used marketing instruments in the market response models are price and advertising intensity/expenditure. The Sethi model (1983) is an advertising response model that describes how sales evolve over time in response to advertising. The rate of change in sales depends on three types of market effects: the gain due to advertising, the loss due to forgetting or competitive factor, and a random effect. Neslin (1990) used a marketing model to measure the effect of coupon promotions on market share, showing that coupons have a pronounced effect upon market share, but varying from brand to brand. Narasimhan, Ghosh, and Mendez (1993) proposed a dynamic marketing response model for durable goods, examining the influence of price and customer perception of product quality on sales. Little (1979) studied the effect of advertising on sales response and found an upward trend in response to sales soon after increased advertising and a relatively slower sales decay on withdrawal of advertising that can be attributed to customer satisfaction. Among other attributes that affect response are sales saturation that occurs at high advertising levels (no further scope for increased sales even after increasing advertising budget and frequency); a possible threshold-like effect at low levels; a change of effectiveness over time because of media and copy changes; and a loss of sales due to competitive advertising. Ailawadi, Lehmann, and Neslin (2001) suggested that deals and coupons increase market penetration for average brand but have a little impact on customer retention as measured by share-of-category requirements and category usage. Advertising too works primarily by increased penetration, but its effect is weaker than that of promotion, that is, promotion works better than advertising for market penetration in case of ordinary brand. Millward Brown International, a consulting firm, developed an integrated system for systematically analysing sales response to advertising expenditures. This model estimates the sales contributions instruments such as coupons, relative distribution, relative shelf price, feature ads, displays, temporary price reductions, pantry-loading, and advertising (Hanssens, Parsons, & Schultz, 2003). Park, Shenoy, and Salvendy (2008) found three groups of factors affecting the effectiveness of mobile advertising, viz. advertisement, audience, and environment. While management use of marketing models is on the rise, a relevant question is raised by Little (1970) as, ‘Why are so many models developed but not used?’ Hanssens et al. (2003) points out, ‘the successful implementation of models depends on data availability, the methodology used, and other characteristics’.

Further in the academic literature, interaction effects between marketing mix instruments has also been contributed by the researchers. These are relevant for marketers such as the interaction effect of advertising and price sensitivity. Models like SCAN*PRO by AC Nielsen and PROMOTION SCAN by IRI are examples of such standardized models. Cooper (1983) examined the interaction effect of the marketing expenditures for prescription drugs on the price-elasticity of demand in the short-term and the long-term. The study could not establish the direct relation with beneficial (information) and artificial (differentiation) effects of marketing based on traditional approaches. Other notable works on interaction effects in the response model include Kaul and Wittink (1995), who suggested that ‘price advertising increases price sensitivity and non-price advertising leads to lower price sensitivity’; Vakratsas and Ambler (1999), who argued that ‘advertising elasticities are higher for new than for established brands’; and Van Heerde, Leeflang, and Wittink (2001), who asserted that ‘supported price discounts are more effective than non-supported price discounts’. A relevant study for the car segment was that of Jayaraman, Arumugam, Kumar, and Kiumarsi (2018), who found that the demand response for non-national cars in Malaysia depends upon excellent after-sales service, better mileage, and value-added salient features.

Research Gap

Most of the literature on market response modelling considers advertising as an important input variable; however, the emergence of multiple forms of advertisements such as newspaper advertisement, magazine articles and advertisements, digital advertisements, display events, and internal branding require deeper research. Another area that is left open is the question of business implementation and strategic impact (Hanssens et al., 2005). A large part of empirical model-based research in marketing pertains to consumer products (less work on durable goods), and this area is an important source for the supply of models (Hanssens et al., 2003). Also, research on actual model use is scarce (Hanssens et al., 2003).

This study contributes to the literature by addressing this gap. The luxury car segment has not been touched upon adequately in the literature. Also, trending communication media such as social media and display events play an important role in customer response, however, research is lacking in this area.

In this article, the study of market response to advertising response is measured through a number of queries received in response to different advertising media. Queries are generated through online and offline and wired mode. Advertising entails a broad spectrum of activities like public relations, digital advertising, social media, print advertising, point of purchase, and so on. The instruments in the model focus on media expenditure as an input for the model.

Branding

A brand is defined as ‘a name, term, sign, symbol, or design, or combination of them which is intended to identify the goods and services of one seller or group of sellers and to differentiate them from those of competitors’ (Kotler, 1991). A prestige brand is a brand that has the highest quality and performance within a product category (Dubois & Czella, 2002). Branding is the process that aims to establish a brand. Branding is one of the most vital aspects of any company, large or small, retail or business-to-business. Branding gives company a unique image, distinguishing it from its competitors. The brand image is established in the hearts and minds of the customers. It is the aggregate of a company’s encounters and discernments with customers and prospects, some of which the company can directly impact and some of which it cannot. It is critical for companies to invest resources into building their image. After all, a company’s image is the source of a guarantee for customers, and it is the foundation for their advertising communication.

An effective brand strategy gives company a competitive edge in increasingly competitive markets. The goals for an effective brand strategy are to deliver the brand image clearly, confirm credibility, connect emotionally with target prospects, motivate the buyer, and build customer loyalty. Luxury markets are all the more relevant for branding strategies, as a luxury brand is used by customers as a symbol of status. Further, different variants of the brands may be targeted towards different groups of people. Hence, a combination of different branding media is often employed to reach different customer groups. Some of the different methods through which the branding takes place include newspaper advertisements, magazine articles and advertisements, digital marketing, display events, and internal branding. The ultimate aim of branding is to generate maximum number of queries through a combination of these different modes (Weidmann & Hennings, 2012).

Market response models are also used in litigation field. Brands suffer from brand gossip, appalling events, and unwarranted competition that may have a negative effect on customer response for a brand. The well-publicized ‘sudden acceleration’ rumour around the Audi 5000 in the US had a strongly negative sales impact, not only on sales of this product but also on the entire Audi brand (Sullivan, 1990).

Newspaper Advertisement

Advertisements are the key instruments that influence the buying behaviour of the consumers and the purchase intent. Advertising spending represents a significant proportion of firms’ marketing budgets. Among buyer behaviour models, advertisements are used to draw customer attention, that is, creating awareness among customers (AIDA model) that further may culminate into response. Advertisements use different appeals such as humour, fear, adventure, emotional, romance, scarcity, sex, social, youth, rational, bandwagon, and endorsement to grab attention and to persuade people to act. Aristotle identified three main appeals of communication known as the rhetorical triangle: ethos (credibility), pathos (emotion), and logos (logic). Advertisement appeals are the material that leads to the arousal of psychological motive for buying. The use of different appeals for the same or different varieties arises from the existential difference in human psychological and physiological needs. Advertisements are used to target the mass audience scattered across geographies. The use of advertisements in a firm’s communication strategy has a strong effect on the consumer. According to Pope (2003), the repetition of an advertisement leaves an impression in the mind of the customers, which will help them to remember the product and leads to repeat purchases. The Gear model (2009) of advertising in this regard explains how well a given advertisement achieves the ultimate objective of all advertising; to increase the ‘purchase intent’ towards the advertised brand.

The ‘death’ of the newspaper industry has not occurred as predicted. In actuality, it has picked up steam. Nielsen Research reports that nearly 70 per cent of the population read newspaper regularly. However, revenue rise has been low in the publishing industry. It is beyond doubt that newspaper advertisements are a very effective way of branding because of their wide reach, especially among the affluent, educated, and professional customer segments. It is the most trusted medium, whereby the brand’s message can be fine-tuned to the best. Newspaper advertisements are usually used for informational rather than transformational branding. The brand identity is solidified by the newspaper advertisements. Also, competitor branding strategies can be monitored by examining what benefits/features the competitors are offering to their customers through newspaper advertisements.

Magazine Advertisement

Magazine advertisement offers an advantage over regular postal mail or daily newspapers due to their capacity to show higher quality pictures, giving a clearer picture of the product, and leaving a better impression in the minds of the potential customers. Further, magazines often focus on a particular demographic, so that the advertisement can be aimed at a specific target group. Also, magazines have a longer shelf-life than the daily newspapers due to their higher printing quality. In fact, some readers collect magazines as a hobby, and some magazines are brands in themselves. The stronger the reader’s connection with the magazine as a brand is, the higher is the level of support that the advertisement gets from the magazine’s identity. Some of the prominent specialized automotive magazines in which the luxury car advertisements and articles are published include AutoBild India, AutocarIndia, and Car India.

Digital Media

Brands need to have a direct dialogue with their prospective customers to begin a profitable business interaction with them. Having realized the power of social media, many luxury brands are leveraging the platform to build a strong connection with their customers through 24x7 online conversations. In particular, luxury brands actively make use of the social media for advertising and marketing (Kim & Ko, 2012).

Digital media is an enabler for business communication. The main attribute of digital media is that the content does not depend on product or business designer alone. It could be user or non-user generated, free of temporal and regional boundaries, has infinitesimal opportunity for reproduction and can be stored with minimal or virtually no cost, providing opportunity for consumer networking, can be expanded from any media source to any other technology platforms that has a digital interface (Mulhern, 2009).

Digital media supports branding efforts by engagement through digital media, mainly internet and mobile phones, having a very wide reach. It is also the most cost-effective advertising channel. Some of the techniques used include search engine optimization, search engine marketing, social media marketing, social media optimization, e-commerce marketing, and e-mail direct marketing. These techniques are becoming more common with the advancement in technology. Social media has emerged as the most powerful channel to interact with consumers, leading automotive manufacturers to try to create a base across all platforms: Twitter, Facebook, Instagram, and YouTube. There are also several car websites that are used by luxury car manufacturers to promote their brands, including CarDekho, Car Wale, India-Top Gear, Overdrive, and Drivespark.

Display Events

Display events provide an effective channel to reach the customers within a target market. By educating the customers, they serve to build trust for the brand. Customers are able to know comprehensively about the brand, product display, features, price, and so on, thereby building brand awareness effectively. This kind of branding generally takes place in public places including residential blocks, shopping malls, and IT parks, depending on the target customer segment.

Internal Branding

Internal branding is the process that empowers employees to propagate the company brand to potential customers. Internal branding focusses on inculcating the company’s culture and identity among the employees, inducing them to be more actively engaged with the company mission, recruitment practices, induction and training programmes, rewards and recognition, and innovation promotion schemes. Employees who are mentally and emotionally connected with the company are the best propagators of the company brand. Internal branding is often done inside the showroom, showing the unique things that employees feel about the company. ‘Employees devoid of brand knowledge are unable to transform the brand vision into the brand reality’ (Berry, 2000). King and Grace (2008) highlight that employees need more information than just technical competency in order to deliver the brand promise to the customer that eventually interests customers to the brand.

Methodology

The objectives of the study were:

to evaluate the effectiveness of different advertising media on customer response for a luxury product

to evaluate interaction effects between the media, and

to evaluate the impact of the introduction of digital marketing on the marketing response to traditional media, viz. newspapers & magazines and display events.

Sampling Plan

The data for the study pertains to the advertising expenditures of a major luxury car brand in Bangalore, India, and the queries generated in the period January 2013–June 2015, was collected from the central marketing office for the brand. The different modes of advertising included newspaper, magazines, display events, and digital media. The dependent variable was the number of queries generated for the cars (at its showroom in Bangalore, India) as a result of the advertising efforts. The data is presented for reference in the Appendix.

Model Proposed

The design of a marketing response model involves variables, functional relations among variables, and data. Variables represent the building blocks of a response model, that is variables are the representation of input and output data. In the model, the independent variables are advertising expenditure in the newspapers and magazines, advertising expenditure in display events, advertising expenditure in digital media. Relations are the connections among variables. Functional form refers to the nature of a relation. Finally, data are the actual realization of variables. Taken together, these four elements provide the basis for building a response model (Hanssens et al., 2003).

For example, Hanssens et al. (2003) presented a model used for measuring the impact of positive public relations and awards in Dell’s consumer segment. The business model of Dell is pull-based, that is, lead generation through advertisement. Customers pay attention to ads placed in magazines, TV, newspapers, and other media vehicles and call in on toll-free 1800 telephone numbers to make a purchase. The dependent variable for their model was the call volume, that is, the number of customers calling to enquire about the product, which is highly correlated with unit sales. The independent variables in the model include quality of product offer, brand awareness, advertising spend, reach of the marketing vehicles, and disposable income of the target market.

The current study proposes a marketing response model for luxury car brands, involving a linear model with all possible interaction effects. The independent variables involved in this study are advertising expenditure in newspapers and magazines, advertising expenditure in display events, advertising expenditure in digital media. Interaction effect is also studied in the model among these variables. The response variable is the number of queries generated from advertisement.

The model is represented as follows:

\begin{matrix} y_{t} = a + b D_{t} + c_{1} N M_{t} + c_{2} D E_{t} + c_{12} N M_{t} . D E_{t} + d_{1} N M_{t} . D_{t} + d_{2} D E_{t} . D_{t} \\ + d_{3} D M_{t} . D_{t} + d_{12} N M_{t} D E_{t} D_{t} + d_{13} N M_{t} D M_{t} D_{t} + d_{23} D E_{t} D M_{t} D_{t} \\ + d_{123} N M_{t} D E_{t} D M_{t} D_{t}, \end{matrix}

where the dependent variable y_t represents the marketing response variable, that is, the number of queries in response to the advertising efforts through different media, NM_t represents the advertising expenditure in newspapers and magazines, DE_t represents the advertising expenditure in display events, DM_t represents the advertising expenditure in digital media, and D_t represents the dummy variable for the introduction of advertising in digital media.

Dummy variables (or indicator variables), act as ‘off/on switches,’ describing which level of a qualitative variable is currently in effect (Hanssens et al., 2003):

D i = {\begin{matrix} 1 i f t h e c h a r a c t e r i s t i c i s p r e s e n t \\ 0 o t h e r w i s e \end{matrix}

When two or more events occur within the same period, it is expected that there will be some effect of the presence of one on the other, that is, an interaction effect. Dummy variables are used to capture their joint influence. For example, Duckworth (1997) captured the joint effect of TV advertising and a ‘four for one’ sales promotion on UK sales of Murphy’s Irish Stout using dummy variables for each of these effects. As another example, Franke and Wilcox (1987) used dummy variables to study seasonality in the effect of advertising on alcoholic beverage consumption; one dummy variable captured the marked increase in beer consumption in the second and third quarters of each year, and the other dummy variable represented a fourth-quarter peak in wine consumption. Similarly, Naik (1999) studied seasonality in sales of Dockers khaki pants using two dummy variables: one for the sharp increase in retail sales in December, and the other for the drop in sales in January.

Table 2.

Tests of Between-Subjects Effects

Source	Model I	Model II
Source	F Stat	F Stat
Corrected Model	2.272*	2.830*
Intercept	3.690*	3.306*
Newspapers and Magazines	1.764	6.253*
Display Events	2.025	5.059*
Newspapers and Magazines*Display Events	1.248	4.406*
D	1.215	4.438*
D* Newspapers and Magazines	0.639	3.464*
D* Display Events	0.289
D*Digital Marketing	6.872**	9.494**
D* Newspapers and Magazines*Display Events	0.082
D* Newspapers and Magazines*Digital Marketing	6.199*	8.429**
D* Display Events*Digital Marketing	4.018*	8.460**
D* Newspapers and MagazinesDisplay EventsDigital Marketing	3.870*	7.148**
R²	58.1%	56.0%

Source: SPSS (ANCOVA) result for marketing data.

The analysis was performed in two phases. Model I included all possible interaction terms (i.e., full factorial model), while Model II was performed stepwise (i.e., only significant terms).

Findings

The results of the dummy variable regression model are presented in Tables 2 and 3. Table 2 presents the tests of between-subjects effects (ANCOVA), and Table 3 presents the parameter estimates for the model.

Table 3.

Parameter Estimates

Parameter	Model I		Model II
Parameter	Coeff	t Stat	Coeff	t Stat
Intercept	−1,171.859	−1.734*	−691.719	−2.085*
Newspapers and Magazines	0.002	1.328	0.001	2.501*
Display Events	0.003	1.423	0.003	2.249*
Newspapers and Magazines*Display Events	−0.000	−1.117	−0.000	−2.099*
[D = 0]	854.544	1.102	422.040	2.107*
[D = 1]	0(a)		0(a)
D* Newspapers and Magazines	0.002	0.799	0.001	1.861*
D* Display Events	0.002	0.538
D*Digital Marketing	0.016	2.621**	0.012	3.081**
D* Newspapers and Magazines*Display Events	−0.000	−0.286
D* Newspapers and Magazines*Digital Marketing	−0.000	−2.490*	−0.000	−2.903**
D* Display Events*Digital Marketing	−0.000	−2.005*	−0.000	−2.909**
D* Newspapers and MagazinesDisplay EventsDigital Marketing	0.000	1.967*	0.000	2.674**
Dependent Variable: Enquiries Generated

Source: SPSS (Parameter Estimates) for marketing data.

Table 4.

Response Function

	Coefficient
Intercept	−691.719088314252000
Newspapers and Magazines	0.002446791394742
Display Events	0.002836715314605
Digital Marketing	0.012149328292351
Newspapers and Magazines * Display Events	−0.000000006898140
Newspapers and Magazines * Digital Marketing	−0.000000028066941
Display Events * Digital Marketing	−0.000000032459338
Newspapers and Magazines * Display Events* Digital Marketing	0.000000000000076

Source: The author.

The results of Model I were significant, explaining 58.1 per cent of the total variation in the enquiries generated; however, the only significant variables in the model were advertising expenditure on digital marketing and its interactions with advertising expenditure on newspapers and magazines and advertising expenditure on display events.

The results of Model II were significant, explaining 56.8 per cent of the total variation in the enquiries generated. All three media were found to have significant positive contributions in the model, with significant negative interactions between each pair of media, and significant positive interaction between all the three media. Also, there was a significant improvement in the contributions of newspapers and magazines after the introduction of digital marketing. The response function obtained from the panel regression results above are summarized in Table 4, up to 15 decimal places.

No. of Queries = −691.719 + 0.00245 * Newspapers & Magazines + 0.00284 * Display Events + 0.01215 * Digital Marketing + (−0.0000000069) * Newspaper & Magazines * Digital Marketing + (−0.00000003245) * Display Events * Digital Marketing + 0.0000000000000076 * Newspapers & Magazines * Display Event * Digital Marketing

The coefficient of advertising expenditure on newspapers and magazines in the model was 0.00245, implying that an increase of ₹1000 in advertising expenditure on newspapers and magazines would result in an average of 2.45 additional queries, other advertising expenditures being held constant and ignoring interaction effects. The coefficient of advertising expenditure on display events in the model was 0.00284, implying that an increase of ₹1000 in advertising expenditure on display events would result in an average of 2.84 additional queries, other advertising expenditures being held constant and ignoring interaction effects. The coefficient of advertising expenditure on digital marketing in the model was 0.01215, implying that an increase of ₹1000 in advertising expenditure on newspapers and magazines would result in an average of 12.15 additional queries, other advertising expenditures being held constant and ignoring interaction effects.

Thus, the model indicates that advertising expenditure in digital marketing has the best response, followed by advertising expenditure in display events and in newspapers and magazines. However, there is a negative interaction between advertising expenditure in display events and in digital marketing and between advertising expenditure in newspaper and magazines and in digital marketing. The negative interaction between advertising expenditure in newspaper and magazines and in display events was less in magnitude than the former two interactions.

Discussion

The study proposes a marketing response model for luxury car brands, involving a linear model with all possible interaction effects. The results of the model provide a measure of the effectiveness of each of the media, the interaction between them, as well as the impact of digital marketing.

The results of the study show that all the three media have significant positive contributions. The medium with best response is digital marketing, with almost five times the response of newspapers and magazines and of display events. This highlights the power and cost-effectiveness of digital media.

The results of the study also show the impact of introducing digital marketing on other media. The response of newspaper and magazine advertisements has significantly increased with the introduction of digital marketing; thus, digital media can be used to reinforce the response to promotional campaigns in newspapers and magazines. On the other hand, the response of the display events was not significantly impacted by the introduction of digital marketing.

The results of the study also indicate significant negative interactions between each pair of media and significant positive interaction between all the three media. This implies that though the contribution of each of the media is positive to start with, as expenditure in the other media increase, the contribution of the media become negative after a certain point. Thus, a ‘saturation point’ is reached.

Implications

The marketing response model developed in the current study is an aggregate-level model that uses market-level data, such as brand advertising expenditure and brand sales or market share (refer Vakratsas & Ambler, 1999).

Academically, many marketing models are developed in the past few decades, but only a small fraction of these models are actually used in the market where it can help the decision-makers, that is, managers and senior executives. The marketing model developed in this study will help the marketers to strategize effectively for the use of communication budget. With much advertising expenditure wasted in ineffective campaigns (Abraham & Lodish, 1990; Lodish et al., 1995a), advertisers should be able to formulate effective strategy based on the market response model. The Indian automobile industry is currently facing stress; major automobile companies see sales decline in 2019 due to downturn in economy, liquidity constraints, and so on. The cascading effect is on selling strategies. Response models like this ensures effective use of budget and generating better response. A more robust model is desired incorporating factors like competition, category effect, long-term effect along with interaction effects and sales effect used in this model. Implementation of this model is easy when sales effect is to be analysed on advertising and promotion budget as marketers also prefer simple standardized models (Hanssens et al., 2003).

Limitations

There are several limitations inherent in the study. The study period taken for the study is relatively small, that is, only 18 months. Also, explanatory power of the model is low, only 56.8 per cent, suggesting that other factors play a role. In particular, the model has not considered any lagged impacts. Also, the negative intercept term in the model is not as expected that says response will be nil or negative if no advertisement is done. Practically, however, even if there is no advertising in any of the media, there would still be some response due to the brand. Thus, a different form of the model may be appropriate.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Marketing Data

Automobile office, Bangalore (2013–15).

Year	Month	Newspaper and Magazines	Display Events	Digital Marketing	Enquiries Generated
2013	Jan	200,000	150,000	0	243
	Feb	230,000	150,000	0	272
	Mar	350,000	180,000	0	331
	Apr	270,000	160,000	0	194
	May	170,000	175,000	0	306
	Jun	320,000	190,000	0	311
	Jul	265,000	170,000	0	318
	Aug	290,000	157,000	0	323
	Sep	325,000	166,000	0	318
	Oct	180,000	198,000	0	297
	Nov	330,000	145,000	0	326
	Dec	355,000	150,000	0	318
2014	Jan	370,000	165,000	50,000	284
	Feb	280,000	210,000	50,000	287
	Mar	420,000	260,000	65,000	356
	Apr	290,000	218,000	68,000	311
	May	310,000	195,000	56,000	309
	Jun	325,000	290,000	75,000	329
	Jul	430,000	266,000	75,000	331
	Aug	415,000	310,000	72,000	351
	Sep	290,000	215,000	85,000	362
	Oct	355,000	240,000	80,000	315
	Nov	275,000	305,000	90,000	378
	Dec	395,000	256,000	96,000	366
2015	Jan	448,000	660,000	100,000	295
	Feb	237,000	491,000	116,000	288
	Mar	780,000	426,000	120,000	375
	Apr	468,000	1580,000	120,000	341
	May	660,000	480,000	120,000	325
	Jun	637,500	226,000	82,200	349

Note

References

Ahuja

(2014). Luxury cars: A new definition of necessity in India. International Journal of Commerce, Business and Management, 3(2), 260–68.

Ailawadi

K. L.

, Lehmann

D. R.

, & Neslin

S. A.

(2001). Market response to a major policy change in the marketing mix: Learning from Procter & Gamble’s value pricing strategy. Journal of Marketing, 65(1), 44–61.

Abraham

M. M.

, & Lodish

L. M.

(1990). Getting the most out of advertising and promotion. Harvard Business Review, 68(3), 50–60.

Bass

F. M.

, Buzzell

R. D.

, Greene

M. R.

, Lazer

, Pessimier

E. A.

, Shawver

, Schuchman

, & Wilson

G. W.

(1961). Mathematical Models and Methods in Marketing. Homewood, IL: Richard D. Irwin.

Berry

L. L.

(2000). Cultivating service brand equity. Academy of Marketing Science, 28(1), 128–137.

Borden

(1964). The concept of marketing mix. Journal of Advertising Research, 4, 2–7.

Cooper

L. G.

(1983). A review of multidimensional scaling in marketing research. Applied Psychological Measurement, 7(4), 427–450.

Dorfman

, & Steiner

P. O.

(1954). Optimal advertising and optimal quality. The American Economic Review, 44(5), 826–836.

Dubois

, & Czellar

(2002). Prestige Brands or Luxury Brands? An Exploratory Inquiry on Consumer Perceptions. Archive ouverte UNIGE.

10.

Franke

G. R.

, & Wilcox

(1987). Alcoholic beverage advertising and consumption in the United States, 1964–198. Journal of Advertising, 16(3), 22–30.

11.

Hanssens

D. M.

, Parsons

L. J.

, & Schultz

R. L.

(2001). Market response models: Econometric and time series analysis (2nd ed.). Boston, MA: Kluwer Academic Publishers.

12.

Hanssens

D. M.

, Parsons

L. J.

, & Schultz

R. L.

(2003). Market response models: Econometric and time series analysis (p. 12). The Netherlands: Springer Science & Business Media.

13.

Hanssens

D. M.

, Leeflang

P. S.

, & Wittink

D. R.

(2005). Market response models and marketing practice. Applied Stochastic Models in Business and Industry, 21(4–5), 423–434.

14.

Jayaraman

, Arumugam

, Kumar

K. M.

, & Kiumarsi

(2018). Factors influencing the purchase decision of non-national cars in Malaysia: An empirical study. Global Business and Management Research, 10(1), 150–162. Retrieved from https://search.proquest.com/docview/2131782901?accountid=164287

15.

Kaul

, & Wittink

D. R.

(1995). Empirical generalizations about the impact of advertising on price sensitivity and price. Marketing Science, 14(3), G151–G160.

16.

Kim

A. J.

, & Ko

(2012). Impacts of luxury fashion brand’s social media marketing on customer relationship and purchase intention. Journal of Global Fashion Marketing, 1(3), 164–171.

17.

King

, & Grace

(2008). Internal branding: Exploring the employee’s perspective. Journal of Brand Management, 15(5), 358–372. doi: 10.1057/palgrave.bm.2550136

18.

Kotler

P. H.

(1963). The use of mathematical models in marketing. The Journal of Marketing, 27(4), 31–41.

19.

Kotler

P. H.

(1991). Marketing management (7th ed.). Englewood Cliffs, NJ: Prentice-Hall, Inc.

20.

Leeflang

P. S. H.

, Wittink

D. R.

, Wedel

, & Naert

P. A.

(2000). Building models for marketing decisions. Dordrecht/Boston, MA: Kluwer Academic Publishers.

21.

Lilien

G. L.

, & Kotler

P. H.

(1983). Marketing models. Englewood Cliffs, NJ: Prentice-Hall, Inc.

22.

Lilien

G. L.

, Kotler

P. H.

, & Moorthy

K. S.

(1992). Marketing models. Englewood Cliffs, NJ: Prentice-Hall, Inc.

23.

Little

J. D. C.

(1970). Models and managers: The concept of a decision calculus. Management Science, 16(8), B466–B485.

24.

Little

J. D. C.

(1979). Aggregate advertising models: The state of the art. Operations Research, 27(4), 629–667. https://doi.org/10.1287/opre.27.4.629

25.

Lodish

L. M.

, Abraham

, Kalmenson

, Livelsberger

, Lubetkin

, Richardson

, & Stevens

M. E.

(1995). How TV advertising works: A meta-analysis of 389 real world split cable TV advertising experiments. Journal of Marketing Research, 32(2), 125–139.

26.

Marinov

, & Marinova

(Eds.). (2011). The changing nature of doing business in transition economies. The Netherlands: Springer.

27.

Mulhern

(2009). Integrated marketing communications: From media channels to digital connectivity. Journal of Marketing Communications, 15(2–3), 85–101.

28.

Narasimhan

, Ghosh

, & Mendez

(1993). A dynamic model of product quality and pricing decisions on sales response. Decision Sciences, 24(5), 893–908.

29.

Neslin

S. A.

(1990). A market response model for coupon promotions. Marketing Science, 9(2), 125–145.

30.

Park

, Shenoy

, & Salvendy

(2008). Effective advertising on mobile phones: A literature review and presentation of results from 53 case studies. Behaviour and Information Technology, 27(5), 355–373.

31.

Parsons

L. J.

, & Schultz

(1990). Marketing models and econometric research. New York, NY: North Holland.

32.

Pope

(2003). Making sense of advertisements. History matters: The U.S. survey course on the Web. Retrieved from http://historymatters.gmu.edu/mse/ads/amadv.html.

33.

Rao

A. G.

, & Lilien

(1972). A system of promotional models. Management Science, 19(2), 152–160.

34.

Sethi

S. P.

(1983). Deterministic and stochastic optimization of a dynamic advertising model. Optimal Control Applications and Methods, 4(2), 179–184.

35.

Sullivan

(1990). Measuring image spillovers in umbrella-branded products. The Journal of Business, 63(3), 309–329.

36.

Telser

L. G.

(1962a). The demand for branded goods as estimated from consumer panel data. The Review of Economics and Statistics, 44(3), 300–324.

37.

Tesler

L. G.

(1962b). Advertising and cigarettes, Journal of Political Economy, 70, 471–499.

38.

Trovato

(2014). Features: Consumption trends—luxury brands—luxury loses some lustre amid emerging market woes. Professional wealth management. Retrieved from https://search.proquest.com/docview/1509011426?accountid=164287

39.

Van Heerde

H. J.

, Leeflang

P. S. H.

, & Wittink

D. R.

(2001). Semiparametric analysis to estimate the deal effect curve. Journal of Marketing Research, 38(2), 197–215.

40.

Vakratsas

, & Ambler

(1999). How advertising works: What do we really know? Journal of Marketing, 63(1), 26–43.

41.

Vigneron

, & Johnson

L. W.

(1999). A review and a conceptual framework of prestige-seeking consumer behaviour. Academy of marketing science review. Retrieved from https://search.proquest.com/docview/200920758?accountid=164287

42.

Weidmann

K.-P.

, & Hennings

(2012). Luxury marketing: A challenge for theory and practice. The Netherlands: Springer.