T
HE IMPLICATION OF
MANAGEMENT
CONTROL SYSTEMS
IN DYNAMIC PRICING
Master thesis by Anita Laura van der Meulen
(S2210614)
MSc Business Administration Organizational & Management Control University of Groningen, Faculty of Economics and Business
Under the supervision of dr. A. Bellisario Co-assessed by dr. E.G. van de Mortel
2
ABSTRACT
3
TABLE OF CONTENTS
1. Introduction 4
2. Literature Review 6
2.1 Management Control Systems & Strategy 6
2.2 Management Control Systems & traditional pricing 7
2.3 Management Control Systems & Dynamic pricing 8
3. Method 11 3.1 Case selection 11 3.2 Data collection 14 3.3 Data analysis 15 4. Findings 17 4.1 ProductOnline 17 4.2 TelOnline 21 4.3 FashionOnline 24 5. Discussion 29 6. Conclusion 32 7. References 34 8. Appendices 37
8.1 Appendix A: Interview example questions 37
4
1. INTRODUCTION
As today’s retailers face a transparent and extremely competitive online market, the use of dynamic pricing has become their most important tool for managing revenues (Akter & Wamba, 2016; Chen & Chen, 2015; Hinterhuber & Liozu, 2014). The use of innovation in pricing is a powerful source of competitive advantage, as it enables companies to significantly outperform their competitors (Hinterhuber & Liozu, 2014). In response to the success stories of dynamic pricing in other industries, many companies have now taken up the challenge of building dynamic pricing capabilities in the hope of gaining a competitive advantage (Akter & Wamba, 2016; Chen & Chen, 2015). The key advantage that rises from these capabilities is the ability to regularly adapt price levels based on competing prices and data analysis, enabling organizations to better match supply with demand. This practice is commonly known as dynamic pricing (Hinterhuber & Liozu, 2014; Strauss, Klein, & Steinhardt, 2018).
Dynamic pricing has a relatively large impact on the profitability of companies, providing greater leverage than for example cost reductions (Hinterhuber, 2016). In contrast to the positive impact a beneficial pricing strategy has on revenue, a flawed pricing strategy can inhibit profitability (Kienzler & Kowalkowski, 2017). Pricing strategies are based on company objectives, and act as a road-map that provide guidance when making pricing decisions and trade-offs (Dolgui & Proth, 2010; Kumar, Anand, & Song, 2017; Kienzler & Kowalkowski, 2017). Good pricing strategies therefore, as any other functional strategy, guide the organization towards its objectives (Merchant & Stede, 2017, p. 11). This means that the proper execution of pricing strategies is of paramount importance, especially in today’s transparent and competitive online market (Akter & Wamba, 2016; Kienzler & Kowalkowski, 2017).
Achieving organizational goals, along this line, requires that the execution of a dynamic pricing strategy – as a strategy in its own right - is supported and monitored by management control systems (MCSs) (Ferreira & Otley, 2009; Merchant & Stede, 2017). MCSs support organizations in achieving their objectives through social and technical controls (mechanisms, processes, systems & networks) that support both strategy (current performance & future opportunities) and operations (performance & risk avoidance) (Ferreira & Otley, 2009; Tessier & Otley, 2012). To achieve organizational objectives, MCSs may be used to validate strategies (Campbell, Datar, Kulp, & Narayanan, 2015), to stimulate certain capabilities (Henri, 2006), or to discover threats and opportunities (Simons, 1995).
5 information (Ferreira & Otley, 2009). Moreover, MCSs may be needed to mitigate dynamic-pricing related risks (Simons, 1995), like deteriorating margins and large-scale pricing mistakes. These suggestions based on previous literature evoke curiosity about the MCSs that monitor and ensure proper implementation and functioning of these dynamic pricing strategies.
However, research on this intersection of MCSs and dynamic pricing is scarce, as the topic is only occasionally touched upon by authors who mostly focus on mathematical models, supply-demand functions and other more technical aspects of dynamic pricing (Chen & Chen, 2015; den Boer, 2015; Kienzler & Kowalkowski, 2017). One notable exception, however, is an article on the implementation of pricing innovations written by Hinterhuber & Liozu (2014). These authors touch upon the subject of management control (MC) as they describe that companies should adapt their organizational structure and processes to pricing innovations in general. Their findings indicate that the adoption of dynamic pricing indeed necessitates change in organizations. However, as their focus lies with organizing for pricing innovations, these changes are not considered from the perspective of MCSs. Therefore, Hinterhuber & Liozu’s findings only further raise the question of how MCSs are implicated in dynamic pricing. The lack of literature on this subject is striking if one considers the aforementioned criticality of pricing and the surging popularity of dynamic pricing strategies. The lack of literature on MCS’s implication in dynamic pricing therefore looks like an important oversight.
This study contributes to the advancement of knowledge on this particular subject by developing an understanding of current organizational MC practices set out for dynamic pricing. The research question that will be answered to shed light on this unknown domain reads as follows: ‘How are Management Control Systems implicated in Dynamic Pricing?’ To answer this exploratory question, a multiple-case study design was adopted using rich interview data from three different retail web shops that practice dynamic pricing (Edmondson & McManus, 2007). The findings of this study show that companies exercise diagnostic control over critical pricing variables to ensure goal achievement, while also pursuing dynamic pricing-related exploration. They also illustrate the importance of interactive control systems for dynamic pricing, as these ensure the spreading of insights from decentralized pricing experiments. Furthermore, tensions between exploration and goal achievement within dynamic pricing were revealed, as a high level of exploration of dynamic pricing strategies endangered goal achievement. Finally, this study uniquely finds that dynamic pricing software supports diagnostic control and experimentation, enabling the simultaneous pursuit of exploration and goal achievement. These findings bring a novel contribution to the current body of literature, as this study is the first to describe these phenomena in the context of dynamic pricing.
6
2. LITERATURE REVIEW
2.1 Management Control Systems & Strategy
Within this study, MCSs are viewed in a comprehensive way in order to explore the main features of MCSs within the specific context of an organization (Ferreira & Otley, 2009). Based on the works of Simons (1995), Ferreira & Otley (2009) and Tessier & Otley (2012), the following definition of the MCS is adopted for the purpose of this study: The complete MCS is comprised of all social and technical controls (mechanisms, processes, systems & networks) that increase the probability of achieving organizational objectives by supporting a range of managerial activities concerned with strategy (current performance & future opportunities) and operations (performance & risk avoidance)(Ferreira & Otley, 2009). This definition indicates that MCSs are involved in the whole strategic process, encompassing processes related to strategy formulation, strategy validation and its implementation through operations. Whereas a strategy can act as a guide map and directs an organization towards its objectives, MCSs also fulfil an important function, as they monitor whether the organization is still on the right path that leads towards the right goal (Merchant & Stede, 2017).
As organizations may follow different strategies and pursue different objectives over time, additional sets of control practices (MCSs) can be implemented to fulfil new needs (Gschwantner & Hiebl, 2016; Malmi & Brown, 2008). An illustrative example of how a MCS can support the execution of a new strategy is provided in an article by Campbell, Datar, Kulp, & Narayanan (2015). Their study showed how a convenience store was able to test and validate their new business strategy by adopting the balanced scorecard. The balanced scorecard is a MCS that provides performance measures that are in line with the different elements of a strategy. This MCS enabled the convenience store to monitor the execution of their strategy, and see whether implementation of its different elements actually led them to their desired goals. The balanced scorecard supports the achievement of organizational goals by bridging the gap between strategy and implementation (Kaplan & Norton, 1996). Depending on the organization’s needs, MCSs can assume a diagnostic role aimed at goal achievement by providing feedback and correcting deviations from standards (Bedford, 2015; Simons, 1995).
7 framework (1995), these different roles of MCSs are explained as ‘Levers of Control’. Furthermore, Simons recognizes not only interactive and diagnostic control levers, but argues that MCSs can also support strategy by providing boundaries and communicating the company’s values. These illustrative examples provide some insight into the ways in which the MCS can support strategy, depending on the needs of the organization. In the next paragraphs I discuss the ways in which MC is involved in traditional pricing.
2.2 Management Control Systems & traditional pricing
Several phenomena on the intersection of control and traditional pricing have been described in previous literature. Sandino (2007) demonstrated that even young, physical retail companies already have a functioning pricing system as part of their basic MCS. In her article, she explains that young retail firms select initial controls based on the perceived importance of these controls for strategy and structure. The importance of the pricing function is also emphasized by Dolgui & Proth (2010) and Hinterhuber (2016), as the pricing parameter has relatively high impact on the firm’s profitability. For this reason, a clear pricing strategy can benefit profitability greatly and it enables retailers to differentiate themselves against their competitors in the market (Kumar, Anand, & Song, 2017). Pricing strategies relate to the organizations goals and guide a company in selecting optimal prices based on different factors. Pricing strategies guide companies in making trade-offs, as they often have to choose between gaining market share, achieving higher margins and improving their image in the eyes of customers (Dolgui & Proth, 2010; Hinterhuber, 2004; Kumar, Anand, & Song, 2017). Properly implementing the right pricing strategy is crucial for retailers, especially in the competitive online environment of today (Kienzler & Kowalkowski, 2017). But discovering whether a strategy actually works is a challenge to retailers specifically, as previous research pointed out that their multiple level strategies make it difficult to identify the actual sources of profit (Kumar, Anand, & Song, 2017). For this reason, it might not be surprising that Sandino (2007) found pricing control systems to be present even in retailers’ most basic MCS. Fortunately, MCSs are able to support retailers in the validation and execution of strategies, as was demonstrated by the case of Campell et al. (2015). Their study illustrates the value of MC at a convenience store, as it enabled them to monitor critical performance variables, and assess the coherence between the different elements of their strategy.
8 supports the generation of information needed to support decision-making and organizational learning (Homburg, Jensen, & Hahn, 2012). As demonstrated by the examples in the previous paragraphs, the MCS is implicated in traditional pricing in several ways. Although literature does consider the implication of MCSs in traditional pricing and pricing strategies, the topic of dynamic pricing seems to be missing from this particular field of literature. Guided by the works of previous authors, dynamic pricing and the possible implications of MCSs within its practice are explored and discussed in the following paragraphs.
2.3 Management Control Systems & Dynamic pricing
The frequent adaption of prices based on competing prices and data analysis enables organizations to better match supply with demand. This practice is commonly known as dynamic pricing (Chen & Chen, 2015; Strauss, Klein, & Steinhardt, 2018). Dynamic pricing is commonly practiced by retailers that sell their goods online, often called web shops (den Boer, 2015). Digital technology enables these organizations to continuously optimize prices to factors of their own choice (den Boer, 2015). Companies can vary the extent of automation and select desired pricing responses based on information like weather forecasts, competitors’ prices or price elasticity (Borgesius & Poort, 2017; Brokesova, Deck, & Peliova, 2014; Chen & Chen, 2015; Kumar, Anand, & Song, 2017). As dynamic pricing is gaining popularity, companies do not necessarily develop their own pricing software anymore, as software companies now offer subscription-based dynamic pricing software (Kilroy, MacKenzie, & Manacek, 2015). The data used for dynamic pricing often comes from a web shops’ own online environment, and is complemented with other data sources (den Boer, 2015). In the current competitive online environment, the ability to use big data and software for the purpose of dynamic pricing is thought to provide key competitive advantages in many industries (Chen & Chen, 2015; den Boer, 2015).
9 Furthermore, previous studies have shown that technological innovations can greatly impact MCSs (Chou, Chou, & Tripuramallu, 2005; Schermann, Wiesche, & Krcmar, 2012). Innovative technology like the ERP system impacts a whole range of processes like planning, decision making and inventory control as large amounts of information suddenly become available throughout the whole organization (Chou, Chou, & Tripuramallu, 2005). Furthermore, if organizations have the capabilities to extract useful insights from all the data, this technology offers them the ability to enhance decision-making and control (Chou, Chou, & Tripuramallu, 2005; Raffoni, Visani, Bartolini, & Silvi, 2018). Similar to ERP technology, dynamic pricing as an innovation revolves around the use of information for a competitive advantage (Hinterhuber & Liozu, 2014). Data analysis specifically plays a large role in dynamic pricing, and may therefore influence decision-making processes by providing MCSs with valuable insights (Raffoni, Visani, Bartolini, & Silvi, 2018).
However, automating processes of decision making and the accumulation of even more data could provide organizations with difficulties and risks as well. The use of data often brings challenges like information overload, non-rational decision making and the complexity of analysing inconsistent data (Byrnes, Moffitt, & Warren Jr, 2015; Raffoni, Visani, Bartolini, & Silvi, 2018; Zhang, Yang, & Appelbaum, 2015). In addition, this great amount of data has to be managed as MCSs rely on essential information flows, systems and networks (Ferreira & Otley, 2009). Therefore, organizing data in a way that prevents information overload while facilitating organizational learning may be essential in obtaining the full advantages of dynamic pricing. In addition, MCSs may play a role in avoiding new risks (Simons, 1995), for example preventing large-scale deviations from target pricing that may emerge from automatic price adaptions. However, despite the aforementioned possible opportunities and challenges, thus far, it is not known how MCSs are implicated in the use of dynamic pricing.
As was illustrated in this review of literature, MCSs increase the probability of achieving the organization’s objectives by supporting the formulation of strategy, its validation and its implementation through operations (Ferreira & Otley, 2009; Campbell, Datar, Kulp, & Narayanan, 2015). For companies in retail, ensuring the proper execution of pricing strategies is critical for survival, especially in the increasingly competitive online environment (Kienzler & Kowalkowski, 2017). Recently, dynamic pricing has surged in popularity, making it the most important revenue management tool for online retailers (Chen & Chen, 2015). As data-based pricing strategies are now executed by specialized software, organizations may need to develop different capabilities, processes and structures to effectively operate this pricing innovation (Hinterhuber & Liozu, 2014).
10 Current literature does not consider MCSs in relation to dynamic pricing, leaving it unknown in what ways control is implicated in dynamic pricing’s specific challenges and benefits. Despite its recent surge in popularity and the adoption of dynamic pricing within many organizations, the implication of MCSs in dynamic pricing remains unexplored. Therefore, although in theory it is unknown how MCSs are implicated in dynamic pricing, exploring MCSs at companies that have taken dynamic pricing systems into practice may shed light on the matter. This study therefore contributes to what is known so far by exploring and answering the following question:
11
3. METHOD
In this study I adopt a qualitative approach, which is appropriate in the light of limited prior theory and research in the field of dynamic pricing and MCSs (Eisenhardt, 1989; Gioia, Corley, & Hamilton, 2012; Glaser & Strauss, 1967). A multiple-case study design is used to enable exploration of unique patterns within cases before stepping beyond these impressions and uncover commonalities between cases (Eisenhardt, 1989). I collected data from the case companies by conducting in-depth interviews and complemented these with several observations and documents. Consistent with the other methodological aspects, I then explored the data for themes and patterns in a qualitative data analysis (Edmondson & McManus, 2007). Finally, my qualitative data analysis was guided by using the systematic inductive approach adopted from Gioia, Corley & Hamilton (2012), which provided me with a backbone for a ‘rigorous’ analysis. Collection and analysis of data was done in the Dutch language, as this enabled me to stay close to the actual words used by the interviewees. The choices and considerations that I made in this study with regard to these subjects are discussed more elaborately in the following sections.
3.1 Case selection
Dynamic pricing is commonly practiced by companies that sell their goods online (den Boer, 2015). Therefore, the setting of this research is the E-commerce market that consists of companies that sell goods to customers over the internet. Data from four different companies is included in the present study (table 1). To enable both the exploration of unique patterns and common themes, this study focusses on the findings from three cases; FashionOnline, TelOnline and ProductOnline1. These three companies make use of dynamic pricing and offer their products online to their customers. The selection of these three cases was guided by the maximum variation strategy in order to uncover both unique variations of phenomena as well as central themes. In line with maximum variation strategy, the selected cases differ on dimensions that might be of interest to the implication of MCSs in dynamic pricing. As Malmi & Brown (2008) described, different MCSs are implemented for different purposes at different times. In addition, MCSs reach different levels of sophistication and complexity depending on their age and growth developments (Sandino, 2007). As these characteristics might give rise to unique variations of phenomena, I chose these to guide my selection of cases. For this reason the three case companies that were selected are approximately the same age but differ in their experience with dynamic pricing and their growth developments. FashionOnline’s dynamic pricing system is still quite young as they adopted pricing software for their 15.000 products three months ago. In contrast, TelOnline was one of the early dynamic pricing adopters 7 years ago, and current pricing processes for its 500 products have been stable
12 for quite some time. Finally, though ProductOnline has also been using its own dynamic pricing software for the past 7 years, the company’s pricing system is going through a turbulent phase of development. Since ProductOnline’s assortment has grown over 150.000 items, the organization faces pricing challenges that necessitate change and a possible transfer to subscription based software. By including these three companies and their varying characteristics, I hope to discover both unique phenomena and common themes.
TABLE 1:
Case company characteristics
Company Products DP Experience,
current developments1 Data collection
ProductOnline >150.000
Broad range of products
7 years of experience Outsourcing to gain control
Exploratory meeting
Observations of dynamic pricing system Semi-structured interviews:
1. E-commerce manager: 67 minutes 2. Controller: 70 minutes
TelOnline >500
Mobile phones
7 years of experience Stable & in control
Exploratory meeting
Observations of dynamic pricing system Semi-structured interviews:
1. Postpaid-performance specialist: 47 minutes 2. Senior purchaser: 58 minutes
FashionOnline >15.000 Fashion items
3 months of experience Start-up phase
Exploratory meeting
Observations of dynamic pricing system Semi-structured interviews:
1. E-commerce manager: 58 minutes 2. Content specialist: 32 minutes 3. Purchasing coordinator: 45 minutes
DynamicCo DP software -
Observations of dynamic pricing system Semi-structured interview:
1. Product manager: 91 minutes
1 Number of years’ experience with dynamic pricing based on data analysis and/or software. Current developments refers to developments within dynamic pricing
14 In order to provide a broader perspective and view the context beyond the three main cases, I included a fourth company in the interviews (Table 1). This originally Dutch company is called DynamicCo and they specialize in developing dynamic pricing software for their clients. Gaining insight into pricing software considerations and its workings is valuable because it is through their software that users are provided with additional pricing skills. Through a plug-and-play construction, the user is suddenly able to monitor prices, use data for market insights and automatically adapt prices without needing advanced IT skills. As companies like ProductOnline consider outsourcing their pricing tasks to a company like DynamicCo, it might be that some of the operational control logic is crystallized into its software. For this reason it is valuable to discover in what way a pricing software company like DynamicCo is involved in the dynamic pricing systems of its clients. Therefore, an interview with the company was scheduled to provide more insight.
3.2 Data collection
Several data sources are used in the course of this study. Primarily, I collected data by conducting interviews with employees involved in dynamic pricing. Furthermore, I observed the workings of the companies’ pricing systems and used information from company documents and company websites.
After initial contact was made with each company, I arranged an informal exploratory meeting to discover whether the company’s pricing practices could be relevant for this study. Although these meetings were informal, I made notes afterwards and these provided me with some preliminary insights into the companies’ activities which supported the development of interview topics. The actual interviews were preceded by two semi-structured pilot interviews with both a MC student and an outsider who was not involved in either control or pricing. These pilot interviews enabled me to discover whether topics and interview questions were clear and unambiguous.
One day before the scheduled interview I sent an e-mail with a reminder of the appointment, agreements on anonymity and interview topics. Each interview I started with a small topic introduction before continuing the in-depth exploration of the topics. According to Baarda, Goede, & Teunissen (2005), the level of structure in an interview depends on the level of prior knowledge on the topic. The use of semi-structured interviews fits the purpose of this study, as there is limited prior theory on the intersection of MCSs and dynamic pricing. In addition, semi-structured interviews unfold in a conversational manner, enabling interviewees to tell their own stories and provide information that they deem relevant (Baarda, Goede, & Teunissen, 2005). In advance I based broad interview themes on Ferreira & Otley’s performance management questions, as these “provide a powerful means of relatively
quickly outlining the main features of a PMS in a comprehensive manner, and the ways in which it is used in the context of a specific organization” (Ferreira & Otley, 2009, p. 266). These themes and some
15 been discussed. This way of interviewing provided me with rich information while I also maintained a focus on relevant themes (Baarda, Goede, & Teunissen, 2005).
3.3 Data analysis
The data analysis approach in this study is based on descriptions of qualitative analysis given by Glaser & Strauss (1967), Strauss (1998), and more recently Gioia, Corley, & Hamilton (2012). In the first phase of the analyses, I transcribed all audio files into coding software and I commenced open coding by searching for relevant text fragments. Open codes were based on interviewees’ own words and done within each case, to prevent the emerging themes from being contaminated by other ideas from previous literature or other interviews (Gioia, Corley, & Hamilton, 2012; Glaser & Strauss, 1967). An example is this statement taken from one of the interviews: “What I am saying, those experiments, we
should record those. We should learn from those experiments”. As the interviewee talks about
experimenting and learning, I coded these sentences with the words “Recording Experiments” and “Learning from Experiments”. As shown with this example, the first level of coding is quite basic as I stayed close to the words of the interviewees. Similar chunks of raw data received the same code, resulting in several categories. This first phase resulted in a great amount of 1st order codes, which I formed and adapted in an iterative way during the process.
The second step of the analysis is the creation of the more general second-order codes (Gioia, Corley, & Hamilton, 2012). By developing these second-order codes, I reduced the number of codes by creating more general descriptors which encompassed collections of first-order codes. This step can be demonstrated by looking at two different first-order quotes. This first quote “If you don’t have the article
in inventory, your price has to be lower than when you do have the product in inventory” was coded
"Inventory influences Price”. Another quote was coded “Competition influences Price” and said “We
look what our competition does and we adjust our prices to match theirs.” These first-order quotes I
could combine into one second-order code named “Establishing Prices” as they both describe how companies establish their prices. Collections of such quotes relating to similar topics are also visible in the data structure in Appendix B.
16 (Simons, 1995). As illustrated by this example, this third level of coding is more abstract and concerned with the theoretical realm (Gioia, Corley, & Hamilton, 2012). In this way, by adopting a holistic perspective and taking an overview of all the second-order codes, I identified several underlying themes. In constructing the findings I also considered my observations of the case company’s pricing systems, and these observations were useful in deepening my understanding of dynamic pricing at the case companies. The data structure in Appendix B shows the results from these coding phases and a selection of quotes.
17
4. FINDINGS
In this chapter I create an encompassing and holistic image of MCSs within the practice of dynamic pricing at the three case companies. This chapter illustrates how the case companies pursue both the achievement of their pre-defined pricing goals while also exploring new possibilities and pricing strategies. A summarizing overview of these themes within the case companies is provided at the end of the findings chapter in table 2. The individual cases and the overview at the end of the chapter are crafted around the main themes that emerged from the data structure (Appendix B). First, each case description starts with a short introduction, followed by findings within the theme of monitoring critical performance variables for reliable goal achievement. Subsequently, the theme of exploration within dynamic pricing and the role of the interactive control system is considered. Findings from each case are concluded with case-specific challenges and a short summary.
4.1 ProductOnline
ProductOnline is a Netherlands-based retail company that offers a broad range of different products to its customers over the internet. Since its founding over ten years ago, ProductOnline has experienced rapid growth in both revenue, sales and employees. To manage setting prices for its fast-growing number of products, ProductOnline’s IT department developed its own dynamic pricing system seven years ago in 2011. At the moment, ProductOnline is doing a trial with different pricing software provided by a software company specialized in dynamic pricing systems. However, currently the majority of their 150.000 products is still handled by their own pricing system, which automatically adapts prices on a daily basis. Automated pricing decisions are based on so-called ‘business rules’, which can be seen as rules of thumb that guide the system in making decisions. An example of such a rule is given by ProductOnline’s e-commerce manager:
‘The rule is Be the cheapest in the market and the goal is Increase the sales
with 30%. The system itself then decides which product prices to adapt, based
on products’ sales-levels. […] It’s a bit of a black box, you don’t know exactly what happens.’
18 With over 150.000 products there have been some instances of pricing going awry, for example because of human mistakes in the documentation of the cost price. ProductOnline exercises management by exception to limit the impact of these mistakes. An example is the monthly check performed by the controller, as he scans lists of product prices for outliers and exceptions and so keeps the risk of errors within an acceptable range. This same method of scanning for outliers is also applied to the product inventory. Excess inventory is not considered a great risk, but low turnovers can harm margins because of storage costs. These and similar controls have been implemented to ensure that attention is given where needed, increasing the probability of spotting critical flaws.
In addition to management by exception, some extra checks and controls have been built into the pricing system and therefore happen automatically. As the controller explains:
‘There are several safety margins. If a price change is too big it falls into a check and it has to be approved. Then I have to approve it, or the system asks: “Do you really want to do this?” Because if you create a rule, it can influence thousands of products. It can mean an adjustment of -50% on thousands of products. So it is important to perform extra checks.’
This quote illustrates the possible risk and impact a business rule can have in relation to pricing. In addition, it shows the necessity of controlling this risk by building in additional business rules that act as automatic checks, like a rule asking for manual authorisation once a certain threshold is reached.
Another area of attention is the monitoring of relative margins on products. Estimating margins is difficult due to rapidly changing circumstances, therefore the controller monitors the company’s course by formulating monthly margin objectives based on the company’s growth ambitions. The e-commerce manager explains why the relation between dynamic pricing, margin and strategy is so important:
‘Margins are pressured on the internet: Because of the transparency it is very easy to become another euro cheaper. […] And it is always volume versus margin. It seems easy, if you want to grow you just lower your price. But for many companies this leads to bankruptcy. […] So there should be a good connection between your strategy and where you want to go as a company. It is not necessary to always be the cheapest, but you do need a dynamic price and you have to score on the things that matter.’
In this quote the e-commerce manager emphasizes the importance of making sure that profit margins are in line with the company’s strategies and growth ambitions, as sales volume often goes at the expense of sales margins. The controller confirms this, he emphasizes the importance of focussing attention on this area by monitoring and frequently discussing margins.
19 strategies come in, as they are connected to their overall strategy and fundamental to achieving satisfactory margins. Experimentation with different business rules on product- and category level therefore plays a big role in discovering good pricing strategies at ProductOnline. Predicting the outcome of a changed business rule on forehand is complex, as sales are also influenced by things like seasonal holidays, delivery times, weather and revenue growth. The controller emphasizes the importance of the category managers in this process:
‘They are the experts on the articles they manage. So they basically have to run their own shop, and especially there they have to experiment and share knowledge.’
This quote illustrates that the category managers are expected to operate independently on their own products, but also simultaneously share their knowledge with others. Experimenting and learning are supported by the way in which control is used, as category managers are stimulated to experiment with business rules and mistakes are accepted and even seen as part of a valuable learning process. As the e-commerce manager explains:
‘There are guidelines, like you never go 5% below the market price because just one euro is already sufficient. [..] But if you want to experiment and exceed this number, you can. Then they set a rule like that, and at the end of the month they show and share: “This is what I have done, and this is the
outcome.” And everyone learns from that, so you can often start such an
experiment.’
As shown by this quote, category managers experiment by setting business rules and learn by looking at the outcomes of these rules. In addition, this quote illustrates a MC balance between boundaries and freedom, as category managers follow pre-defined guidelines but are also provided with room to experiment. However, providing the freedom to experiment while maintaining oversight and control also creates some challenges. For example, the e-commerce manager notes that:
‘Some projects only needed a few days but were eventually left being activated for months, they had just fallen into oblivion.’
20 on subscription-based software, as this system offers structure, more insight into its workings and restricts freedom somewhat.
The e-commerce manager is happy about the fact that currently more written rules and procedures are being developed, as he thinks the system is growing more mature. However, he also notes that:
‘You want to stay a young and explosive organization. […] You don’t want to become a bureaucracy, that you want something and have to stop at six different desks.’
In this quote, the other side of developing established procedures is addressed, as it expresses the need for balance between flexibility and formalization. Where this balance lies is not yet clear, but the controller believes that they could benefit from more formalized procedures in relation to the pricing experiments. He states:
‘There is a large risk that we experiment but that we don’t learn. […] How do you make sure that this knowledge reaches other people? How do you make sure we actually use this knowledge as a company? […] We have so many opportunities. We have a lot of data. We have young and qualified people. This organization can change super-fast. […]. It would be great if we can find a format for it. And somewhat formalize it, so it will actually continue to happen.’
As the controller expresses in this fragment, there are not many formalized processes that ensure new-found knowledge to be shared and used. Both he and the e-commerce manager believe that much more would be gained if ProductOnline would standardize and formalize its information-flow processes more, organizing people and information in a way that supports organizational learning.
21
4.2 TelOnline
TelOnline is a Dutch telecom company that was founded over ten years ago and nowadays employs over 200 people. The company sells over 400 types of devices, all of which are sold as postpaid phones or as hardware-only deal. To support their low-price strategy, TelOnline developed its own dynamic pricing system seven years ago in 2011. Their software enabled them to automatically follow competitor pricing and so guarantee their customers the lowest prices in accordance with their strategy. TelOnline suspects they were the first company in the Dutch telecom industry to implement automated pricing. Nowadays however, their sales margins are pressured as many competitors use similar pricing systems and automatically lower prices in response to competition, resulting in a snowball effect. TelOnline’s pricing system automatically updates the prices of hardware-only deals based on competition every day. The price changes and strategies are monitored by a small purchasing team which consists of three people. The senior purchaser describes dynamic pricing at TelOnline:
‘Every day we look for each device what our competition does, and then change our prices accordingly. This happens largely automatically. We spider our competitors by using pricing websites. [...] But some of it is done manually because we don’t want to follow every competitor on every device’
This quote explains what dynamic pricing means at TelOnline, as their price adaptions are based mainly on competitors in the market. The postpaid department, however, manually adapts prices of postpaid phones as they need to take into account other factors like operator deals and discounts. Therefore, pricing of postpaid phone deals is too complicated for the pricing software to handle.
Since their first implementation of the pricing system, TelOnline has developed a multitude of controls and checks in response to their experiences. As the senior purchaser notes:
‘There are so many controls in there, and that’s because at some point in time each of these things went wrong.’
One of these controls the senior purchaser refers to limits the system’s pricing options to a predefined range, which prevents under-pricing but also excessive overpricing of a product when the system finds no competing prices. Another check was added to the system in response to an unnoticed system failure, which they only discovered after a few days without price adaptions. In response to this experience the system was adapted to send a notification every time the prices are adapted. Furthermore, major changes in price settings are documented so that employees can check each other’s adjustments. Finally, TelOnline’s customer service department provides an extra check, as customer service employees will often spot pricing that deviates from targeted prices.
22 ‘We check how often we sold it in the past week, the past two weeks and the
past four weeks. Then we can see the run-rate and whether sales are increasing. […] Then we look at our inventory to see whether it is one or two weeks. Next we look at our competition: “Where are we in relation to the market? And can
we do something about our margin?” Based on this information we decide.
And we do this every morning for nearly five hundred models.’
As this quote illustrates, this ritual of monitoring prices, inventory and margin that the purchasing team executes each morning is quite extensive. The senior purchaser emphasizes that these variables are critical and enable the company to exercise dynamic pricing in the way they currently do. TelOnline’s pricing system responds quickly to price-drops in the market, but such price drops initiated by competitors result in risking insufficient margins, excess inventory and write-offs. Forced write-offs on inventory due to changes in the market are a major risk for TelOnline, which necessitates a tight control on the turnover of inventory. Furthermore, TelOnline tries to make more accurate sales forecasts through their daily process of evaluating real-time market information. This results in an inventory turnover of only eight days, thus mitigating the risk of being left with overprized inventory and therefore enabling TelOnline to dynamically move its prices along with the market. According to the senior purchaser this differentiates them from the rest of their competitors, as TelOnline is relatively fanatical in monitoring these variables.
In addition to continuously responding to the market, it is also important for TelOnline to learn in what ways the market responds to their pricing. For example, TelOnline experiments to discover how much they should to lower their prices to seduce customers that would normally be loyal to competitors. The senior purchaser describes some of the advantages and challenges they meet in the process:
‘A few guys from the IT department explore all data of the past years. […] We discovered that one brand has more elasticity than the other. We already had the gut-feeling that this was true, and now it was really supported by a test. […] But the outcome you received back then does not mean that it is true for this moment as well. […] That makes it extremely difficult. But in the end we do think about it, and that is the most important thing.’
23 insight and experience is particularly important to TelOnline. Moreover, supporting this process by exploring and sharing historical data as described above may be crucial to gaining insight.
The senior purchaser and the postpaid performance specialist are both quite content with how information is shared within their organization. Additional formalization of structures and procedures that enable knowledge sharing is not needed, as the postpaid performance specialist describes:
‘Of course everyone has their own stuff they are working on. The purchasing team probably also has stuff that works well. There are short lines of communication between both our teams, we consult each other a lot. So we know from each other what we are working on, and which things are successful. Additionally, the organisation is not quite big enough for many people from different teams to be working on the same subjects. I think it works very well, everyone has their own specialisation and also knows what their colleagues are working on.’
This quote indicates that, in the case of TelOnline, there is sufficient face-to-face communication between the functional teams. The head of the purchasing team agrees that gained knowledge reaches different teams and can thus be used throughout the organization. It seems that the way in which TelOnline has currently organized its communication of knowledge and insight works quite well for them.
The process of gaining insight is also supported by the way control is used, as there are no negative consequences to disappointing results or a failed experiment. The senior purchaser explains this:
‘There are no negative consequences. We experimented and we asked the manager for his opinion […] In the end, it has al cost us money. Some things have earned us more money as it did have a desirable effect, and then we thought to ourselves “Why didn’t we try this sooner?” […] And in some areas we sacrificed a lot of margin, in those cases it did not lead to the desired outcome at all, and it actually cost us money. All in all, it means we are constantly searching.’
And, as the postpaid performance specialist describes:
‘Sometimes you try something, and it just doesn’t work out. […] In that case we remember it, so we can do it differently next time.’
24 manager considers whether he wants us to implement it or not.’ He emphasizes that although the team is trusted to have the experience and insight to make decisions, they discuss their strategies on a weekly basis with management. This combination of using bottom-up insights and experience while also regulating through top-down approval provides a balance between the use of new ideas and maintaining managerial control.
To summarize the previous paragraphs, TelOnline seems to consciously combine tight control with the possibility to explore. The first paragraphs showed how TelOnline uses a set of more diagnostic procedures to keep the critical performance variables of its strategy on track, like price levels and inventory turnover. Making accurate sales forecasts is difficult due to changing market conditions, but TelOnline compensates for this fact by daily monitoring procedures and using real-time market insight for decision making. In addition, TelOnline feels the need to gain more insight into the ways in which the market responds to their pricing strategies. Therefore, room is provided for experiments in a regulated manner, which supports the discovery of new strategies and insights. By balancing managerial consent with the use of bottom-up insight and ideas, TelOnline is able to experiment without losing oversight. Sharing useful insights throughout the organization is not an issue, as small teams have their own specialized tasks but also collaborate and consult each other.
4.3 FashionOnline
Fashion Online sells accessories and other fashion items over the internet, and currently offers its customers more than 16000 products to choose from. The originally Dutch company was founded over ten years ago and nowadays employs more than 50 people. Recently, FashionOnline realized that their market was more dynamic than initially thought, as they discovered that their competitors change product prices quite frequently. As a result of this realization, FashionOnline decided to implement a dynamic pricing system, because they suspected that not responding to these price changes was hurting their sales. Now, five months later, they are in the process of experimenting with a dynamic pricing system and discovering whether automated dynamic pricing brings benefits.
25 In addition to the aforementioned adjustments, multiple control procedures have been implemented to control some of the risks related to dynamic pricing. An example is the e-commerce manager who exercises management by exception by checking pricing lists on a weekly basis in the hope of spotting prices that deviate outside of the normal range. As the e-commerce manager explains:
‘I run through the highest numbers, the lowest numbers, the most expensive products and the cheapest products that were sold the day before. And I hope to prevent mistakes by doing that.’
This is an example of management by exception, as the e-commerce manager hopes to spot errors among the pricing extremes of the list. Before the pricing lists are imported into the system, this same check is performed one more time by the content specialist who is responsible for the website. Although these checks mitigate the risk of errors, prices are changed in the middle of the week as an extra precaution so any mistakes that have slipped through can still be corrected before the weekend.
As the pricing system most often follows competitors’ pricing downwards, controlling for the risk of inventory excesses is not yet a priority at FashionOnline. However, the purchase coordinator recognizes that some additional inventory controls need to be implemented if FashionOnline is to continue its dynamic pricing activities. She explains how their inventory levels are influenced by dynamic pricing:
‘We gave a nice discount on seasonal product X. And it sold way faster than before. Therefore we adjusted our ordering amount to that level as well, and we purchased inventory based on these high sales. Until we discovered that at the bottom-line this discount was a bit too outrageous. And then you want to raise the price again, but the amount of inventory was already purchased.’
As this quote illustrates, FashionOnline risks excess inventory if they choose to raise their prices after a price drop. However, mitigation of this risk is not yet a priority, as the purchase coordinator states that such a price-rise situation almost never occurs. For this reason, back orders are still done in a reactive way, by ordering a monthly supply once inventory levels drop below fifty percent of the monthly sales.
In addition to managing inventory levels and tracking pricing errors, FashionOnline also faces the challenge of monitoring the sales margins that are pressured by the price reductions. Although these margins are critical to the company’s survival, the purchase coordinator worries that they are not properly monitored. She states that:
‘We need more reporting. What are the differences between lowering prices and not lowering prices? What does it do with our sales and margins? What does it mean for our back orders and handling in inventory?’
26 measure because indirect costs like back orders have to be estimated. More concerns were raised from other departments as both the E-commerce manager, the purchase coordinator and the content specialist wonder about possibly negative consequences of dynamic pricing for their profitability. As the E-commerce describes his doubts:
‘The most important question is: “If I lower the price of a product with 12
euros because my competition does the same, then do I even sell any more products at all? And how do I sell in a way that, with net margin included, it will actually yield more money?” I can’t make that business case, because I
don’t know how much more or less I will sell. It is the most important question, and I can’t answer it.’
This quote illustrates an important dilemma, as the e-commerce manager expresses that he is unsure about the actual sales benefits that dynamic pricing brings. He states that he does not know how sensitive his customers are to a price drop of 12 euros, indicating that he wonders about the price elasticity in his market. FashionOnline’s e-commerce manager and content specialist point out two challenges that make this question difficult to answer.
Firstly, interfering factors like changes in weather and upcoming commercial holidays influence sales levels and make it difficult to discover whether a sales rise is actually caused by their own price alterations. Secondly, FashionOnline is not able to discover how price sensitive their customers are as the technical limitations of their pricing system prevent them from experimenting with different price levels. FashionOnline’s pricing system either follows the lowest price in the market or leaves the price unaltered. As the content specialist explains:
‘There is no in between. If we can’t follow the price of the lowest retailer, then nothing happens and we don’t follow the second lowest either. Then the whole product is left out of consideration. […] But I think that maybe if we come in second, we would also earn money. Maybe customers would consider us a more reliable party, and despite us being 10 euros more expensive they’d still want to buy from us.’
28
TABLE 2:
Themes that emerged from the data structure
Themes ProductOnline TelOnline FashionOnline Goal achievement: Monitoring of critical performance variables
Diagnostic control practices:
Management by exception
Built-in checks: acceptable price range
Measures focus attention on margins & inventory Trial system provides performance overviews
Diagnostic control practices:
Management by exception
Built-in checks: acceptable price range Daily inventory & margin monitoring Top-down authorization of experiments
Diagnostic control practices:
Management by exception Manual checks
Centralized pricing decisions
Main challenges:
Feedback on outcomes of price change
Main challenges:
Feedback on outcomes of price change
Main challenges:
Feedback on outcomes of price change Reliable margin monitoring
Exploration: Pursuing emergent strategies, innovation and organizational learning
Interactive control practices:
Decentralized experiments & learning Software allows insight in pricing in industry Software enables experiments with business rules Any result is a learning opportunity
Decentralized pricing decisions
Interactive control practices:
Partial decentralized experiments & learning Software allows insight in pricing in industry Software enables experiments with elasticity Any result is a learning opportunity
Knowledge-sharing routines & rituals
Interactive control practices:
Current experiment is dynamic pricing Software allows insight in pricing in industry Group e-mail contains all price changes
Main challenges:
Finding format for regulated experimenting Finding format for sharing & using knowledge
Main challenges:
Using test results to estimate outcomes
Main challenges:
Restraints of pricing system limit experiments Balancing goal achievement with exploration: Achievement of defined goals and Exploring new opportunities
Balance: Leaning towards exploration
Decentralized experiments prioritized over control Experimenting within boundaries of defined strategy Top-down authorization of exceptional changes
Balance: middle ground of exploration & control
Bottom-up ideas with top-down authorization
Balance: Leaning towards goal achievement
Centralized pricing decisions and limited opportunities for exploration.
Main challenges:
Decentralized experiments at expense of oversight
Main challenges:
Interviewees are content with current balance
Main challenges:
5. DISCUSSION
In this section, the case findings are considered together in light of previous literature and enriched with insights I gained from interviewing pricing software developer ‘DynamicCo’. The discussion is crafted around the three themes that emerged from the findings, as presented in table 2. First, critical performance variables of dynamic pricing are discussed in relation to diagnostic control and goal achievement. Subsequently, exploration within dynamic pricing and the role of the interactive control system is considered. Finally, the balance and tension between these different aspects is discussed.
The first implication of MCSs in dynamic pricing emerging from the findings is the diagnostic control of the critical performance variables related to pricing. The variables related to pricing that were deemed critical by the three case companies are price values, product margins and inventory levels. Goal achievement is supported as diagnostic control systems monitor critical performance variables and draw managerial attention to problematic areas of performance (Bedford, 2015; Simons, 1995). These case findings of performance monitoring within dynamic pricing are in line with Hinterhuber & Liozu (2014), as these authors briefly mention similar diagnostic control aspects in their study on organizing for pricing innovations. However, my findings also showed that the pricing systems of TelOnline and ProductOnline support control by mitigating the risk of price deviations through automatic pricing controls in the form of business rules. This particular function was also observed in the pricing software offered by DynamicCo. Whereas previous studies focused on modelling demand functions and did not elaborately consider the use of business rules in dynamic pricing (Chen & Chen, 2015), this study views business rules from a MC perspective and shows that a number of these business rules serve control purposes. Although previous research showed that specific MC software is used for diagnostic control practices (Schermann, Wiesche, & Krcmar, 2012), this study uniquely finds that the business rules in dynamic pricing software are also used for diagnostic control practices.
30 pricing strategy. These findings provide an important and new contribution, as they show that dynamic pricing software is used diagnostically for performance monitoring and interactively for exploration.
Despite the use of data analysis and dynamic pricing software for the evaluation of dynamic pricing strategies, all case companies report that relationships between their price changes and the resulting revenues are still difficult to assess (Bauer & Jannach, 2018). The current study complements previous literature by showing that, as consequence of these difficulties, pricing strategies are still mainly based on personal experience and insight. This finding is further affirmed by DynamicCo, as they explain that even elaborate predictive models are not reliable due to constantly changing conditions. According to DynamicCo, the power of their software lies not in predicting pricing strategy outcomes, but in its ability to provide real-time insight and shorten the feedback loop between pricing actions and results.
To ProductOnline and TelOnline, building personal experience and insight are critical for implementing good pricing strategies and therefore also of great importance in achieving profitable margins within the transparent online environment.To achieve the exploration purposes of learning and developing insight, both TelOnline and ProductOnline use their pricing software to engage in a fair amount of experimenting with dynamic pricing strategies. In line with the goal of exploration, employees are stimulated to learn from mistakes and control is not used for the punishment of failed experiments (Jørgensen & Messner, 2009). The findings also showed that pricing software is used at these companies to explore new pricing strategies by experimenting with different business rules. Moreover, pricing software capabilities proved to be critical in enabling these experiments, as FashionOnline reported that their limited pricing software prevented them from engaging in pricing experiments. According to DynamicCo, dynamic pricing software also contributes to experimentation by providing employees with the ability to ‘play’ with prices on a local level, without needing any help from IT-specialists. Although Hinterhuber & Liozu (2014) have reported that some leading-edge companies treat dynamic pricing as organizational learning, previous literature has not considered the role of pricing software itself in this exploration behaviour. Therefore, this study expands what is known about learning and building insight from dynamic pricing, as its findings reveal that pricing software is an important enabler of decentralized pricing experiments.
31 ProductOnline also has a center-led pricing function but has decentralized decisions of day-to-day pricing and experimenting to a greater extent (Hinterhuber & Liozu, 2014). Despite the learning benefits of decentralized experimenting, ProductOnline’s organizational learning is less efficient because locally gained knowledge is not sufficiently spread throughout the organization. Their case confirms that decentralized learning may happen at the expense of sharing and using knowledge (Fang, Lee, & Schilling, 2010). This side-effect of decentralized learning has not been reported before in the context of dynamic pricing. In addition, ProductOnline misses some of the formalized communication patterns associated with the interactive control system. This lack of sharing knowledge therefore also illustrates the importance of interactive control systems, as knowledge sharing is needed in order to fully achieve the benefits of decentralized pricing experiments (Bedford, 2015; Bisbe & Otley, 2004). The importance of implementing interactive control systems specifically in the organization of dynamic pricing has not been considered by previous research. This finding is an important contribution to previous recommendations by Hinterhuber & Liozu (2014), as it shows that simply combining a center-led pricing function with decentralized execution of pricing tasks is not enough if a company wishes to share and exploit locally gained knowledge. Rather, interactive control systems are needed to ensure that knowledge is shared and the benefits of exploration activities are realized.
32
6. CONCLUSION
In this study I investigated the implication of MCSs in dynamic pricing at three different web shops in the Netherlands. This multiple case study approach was adopted to enable exploration of MC practices at companies that have adopted a dynamic pricing system. A holistic perspective on MCSs was gained as the interviews that were conducted were guided by themes adopted from Ferreira & Otley’s extended framework for analysis (2009). Furthermore, I conducted a rigorous qualitative data analysis in which I was guided by Gioia, Corley & Hamilton’s systematic inductive approach (2012). Furthermore, I enriched my findings by gaining additional insight into the possibilities of dynamic pricing through observing the workings of different dynamic pricing software systems and conducting an interview with a developer of dynamic pricing software.
The findings of this study show that the case company’s MCSs are implicated in dynamic pricing in several ways. Firstly, diagnostic control supported goal achievement as critical pricing performance variables were constantly monitored through different processes. Pricing software supports this diagnostic function by providing real-time insight into critical performance variables and through business rules that act as automatic checks. Two case companies considered exploration within dynamic pricing to be critical to building insight. Pricing software supported exploration, by providing real-time insight and by enabling decentralized experimenting. Interactive control systems were shown to be very important for dynamic pricing exploration, as they proved critical in spreading insights gained from decentralized pricing-strategy experimentation. Furthermore, one of the case companies experienced some tensions between the level of experimenting with business rules and the achievement of organizational goals. Finally, my findings indicate that dynamic pricing software has the potential to enable companies to simultaneously pursue exploration and goal achievement, by enabling local experiments, strategy validation, oversight, insight development and performance monitoring. From these findings I am able to answer the research question, as MCSs were implicated in dynamic pricing in several ways as they support web shops both in their pursuit of goal achievement and exploration. Furthermore, this study provided preliminary insights into some of the opportunities that may be gained from the use dynamic pricing systems.
33 pricing literature. Therefore, future research may contribute more knowledge by adopting a more detailed level in a broader range of contexts.
Future research directions may focus on the use of business rules in dynamic pricing, as Chen & Chen (2015) already indicated that the literature field lacks elaborate studies on this subject. As business rules underlie pricing strategy implementation in the companies I studied, the subject was touched upon multiple times. The use of a pricing system with business rules seems important in enabling experimentation while providing boundaries in support of goal achievement, which may provide a further avenue for research. Additional research could more thoroughly explore if and how this form of information technology may help balance goal achievement with exploration activities. Such research can further strengthen previous findings on the value of information technology in supporting the simultaneous pursuit of exploration and goal achievement (Schermann, Wiesche, & Krcmar, 2012).
34
7. REFERENCES
Akter, S., & Wamba, S. F. (2016). Big data analytics in E-commerce: a systematic review and agenda for future research. Electron Markets, 26, 173–194.
Baarda, B., Goede, M., & Teunissen, J. (2005). Basisboek kwalitatief onderzoek : Handleiding voor het
opzetten en uitvoeren van kwalitatief onderzoek (Vol. Chapter 6). Groningen: Stenfert Kroese.
Bauer, J., & Jannach, D. (2018). Optimal pricing in e-commerce based on sparse and noisy data.
Decision Support Systems, 106, 53-63.
Bedford, D. S. (2015). Management control systems across different modes of innovation: Implications for firm performance. Management Accounting Research, 28, 12-30.
Bisbe, J., & Otley, D. (2004). The effects of the interactive use of management control systems on product innovation. Accounting, Organizations and Society, 29, 709-737.
Borgesius, F. Z., & Poort, J. (2017). Online Price Discrimination and EU Data Privacy Law. J Consum
Policy, 40, 347-366.
Brokesova, Z., Deck, C., & Peliova, J. (2014). Experimenting with purchase history based price discrimination. International Journal of Industrial Organization, 37, 229–237.
Byrnes, P., Moffitt, K. C., & Warren Jr, J. D. (2015). How Big Data Will Change Accounting.
Accounting Horizons, 29(2), pp. 397–407.
Campbell, D., Datar, S. M., Kulp, S. L., & Narayanan, V. G. (2015). Testing Strategy with Multiple Performance Measures: Evidence from a Balanced Scorecard at Store24. Journal of
Management Accounting Research, 27(2), 39–65.
Chen, M., & Chen, Z.-L. (2015, May). Recent Developments in Dynamic Pricing Research: Multiple Products, Competition, and Limited Demand Information. Production and Operations
Management, 24(5), 704-731.
Chou, A. Y., Chou, D. C., & Tripuramallu, H. B. (2005). BI and ERP integration. Information
Management & Computer Security, 13(5), 340-349.
Corbin, J., & Strauss, A. (1998). Basics of qualitative research : Techniques and procedures for
developing grounded theory. Thousand Oaks: Sage Publications.
den Boer, A. V. (2015). Dynamic pricing and learning: Historical origins, current research, and new directions. Surveys in Operations Research and Management Science, 20, 1-18.
Dolgui, A., & Proth, J.-M. (2010). Pricing strategies and models. Annual Reviews in Control, 34, 101-110.
Edmondson, A. C., & McManus, S. E. (2007). Methodological fit in management field research.
Academy of Management Review, 4(32), 1155–1179.
Eisenhardt, K. M. (1989). Building Theories from Case Study Research. The Academy of Management
35 Fang, C., Lee, J., & Schilling, M. A. (2010). Balancing Exploration and Exploitation Through Structural Design: The Isolation of Subgroups and Organizational Learning. Organization Science, 21(3), 625-642.
Ferreira, A., & Otley, D. (2009). The design and use of performance management systems: An extended framework for analysis. Management Accounting Research(20), 263-282.
Gioia, D. A., Corley, K. G., & Hamilton, A. L. (2012). Seeking Qualitative Rigor in Inductive Research: Notes on the Gioia Methodology. Organizational Research Methods, 16(1), 15-31.
Glaser, B. G., & Strauss, A. L. (1967). The Discovery of Grounded Theory. New Brunswick; London: AldineTransaction.
Gschwantner, S., & Hiebl, M. R. (2016). Management control systems and organizational ambidexterity. Journal of Management Control, 27, 371-404.
Hansen, A.-K., Joseph, K., & Krafft, M. (2008). Price delegation in sales organizations: an empirical investigation. Official Open Access Journal of VHB, 94-104.
Henri, J.-F. (2006). Management control systems and strategy: A resource-based perspective.
Accounting, Organizationas & Society, 31, 5229-558.
Hinterhuber, A. (2004). Towards value-based pricing—An integrative framework for decision making.
Industrial Marketing Management(33), 765– 778.
Hinterhuber, A. (2016). Towards value-based pricing—An integrative framework for decision making.
Business Horizons, 59, 71-83.
Hinterhuber, A., & Liozu, S. M. (2014). Is innovation in pricing your next source of competitive advantage? Business Horizons(57), 413-423.
Homburg, C., Jensen, O., & Hahn, A. (2012). How to organize pricing? Vertical delegation and horizontal dispersion of pricing authority. Journal of Marketing, 76, 49-69.
Jørgensen, B., & Messner, M. (2009). Management control in new product development: The dynamics of managing flexibility and efficiency. Journal of Management Accounting Research, 21, 99-124.
Kaplan, R., & Norton, D. (1996). Using the balanced scorecard as a strategic management system.
Harvard Business Review.
Kienzler, M., & Kowalkowski, C. (2017). Pricing strategy: A review of 22 years of marketing research.
Journal of Business Research, 78, 101-110.
Kilroy, T., MacKenzie, I., & Manacek, A. (2015, April). Pricing in retail: Setting strategy. (McKinsey&Company, Ed.) Retrieved from mckinsey.com: https://www.mckinsey.com/industries/retail/our-insights/pricing-in-retail-setting-strategy Kumar, V., Anand, A., & Song, H. (2017). Future of Retailer Profitability: An Organizing Framework.
Journal of Retailing, 93, 96-119.
