Recommending untapped M&A opportunities: A combined approach using principal component analysis and collaborative filtering
Introduction
Since its introduction in the 1990s, recommender systems (RS) have considerably gained in significance as an important instrument to suggest products or services tailored to user's preference (Park, Kim, Choi, & Kim, 2012). An important driver behind this development has been the simplicity in data collection and reporting activities. Users can easily provide a feedback on consumed items using a simple rating system (e.g. five-star rating system), while data on purchase and browsing histories can be monitored effortlessly. Amazon, for example, uses the feature “customers who bought this item also bought” to recommend online products for different users. Netflix, an online subscription-based streaming service for various media contents, generates a significant portion of sales volume mainly through its well-established recommendation engine. Hence, recommendation performances have considerable impact on the commercial success in terms of customer satisfaction, retention and revenue generation (Cho et al., 2005, Herlocker et al., 2004). In recent years, a diverse array of advanced recommendation models for various application domains and business contexts have been proposed (Carrer-Neto et al., 2012, Nilashi et al., 2014). A huge volume of literature pertaining to collaborative filtering (CF)-based approach for generating recommendation has been devoted to supporting decision making in real-world applications (Ignatov et al., 2016, Yoon et al., 2017). However, despite the widespread popularity of using RS as a way to extract meaningful business intelligence, they are mainly centered on topics related to online retail and e-commerce businesses, such as music (Sánchez-Moreno, Gil González, Muñoz Vicente, López Batista, & Moreno García, 2016), movie (Salter & Antonopoulos, 2006) and tourism (Nilashi, bin Ibrahim, Ithnin, & Sarmin, 2015). Applications of CF-based approaches in identifying new business or collaboration opportunities are rather scarce (Park & Yoon, 2017).
Among various forms of inter-organizational practices available to gain competitive edge (de Man & Duysters, 2005), M&A enable firms to increase shareholder wealth by building up complementary capabilities, which then again stimulate the creation of new knowledge and business values (Makri, Hitt, & Lane, 2010). However, despite the recurring waves of M&A activity, many of them did not lead to the expected outcomes both in terms of innovative and financial performances (Hitt, Hoskisson, Ireland, & Harrison, 1991). Many acquisitions are described as failures due to rushed due diligence process, poor cultural fit and lack of proper absorptive capacity (Cartwright and Schoenberg, 2006, Perry and Herd, 2004). Although previous studies have addressed the significance of assessing and valuing suitable acquisition targets, processes related to discovering promising opportunities at the front end of M&A process received limited attention (Marks and Mirvis, 2001, Morrison et al., 2008). Existing approaches tend to focus on the evaluation of potential target companies from technological perspective or are only concerned with the investigation of financial variables (McGee and Byington, 2016, Park et al., 2013). However, discovering novel M&A opportunities can be also pursued via generating market-based competitive intelligence, relying on the analysis of market characteristics (Peng & Liang, 2016). The most fundamental step in increasing the success of M&A is tracking down a manageable number of opportunities for further evaluation. In this context, recommending potential targets from an environment overloaded with information has become an important strategic issue. Especially, as evaluation of expert opinion might be subject to subjectivity and ambiguity, it is prudent to develop data-driven business intelligence to cope with this issue. Consequently, the development of a recommendation system that provides a high-quality decision support is increasingly regarded as a key factor in creating actionable business intelligence (Dao, Jeong, & Ahn, 2012).
Therefore, the paper aims to develop a framework for discovering promising M&A opportunity by combining PCA, which is used to transform a number of related variables into a smaller set of uncorrelated variables, with CF technique. CF relies on the use of retrospective data to make personalized recommendations for the target. Hence, the proposed framework of M&A opportunity discovery (MOD) can constitute a useful point of departure to assist the future M&A-related investment decisions and to achieve operational synergies. Moreover, small and medium-sized enterprises (SMEs) generally suffer from insufficient and biased understanding of the business landscape due to lack of capacity. In this respect, the proposed approach can increase the efficiency of their MOD process and provide decision-makers with expertise that goes beyond their knowledge level. The framework includes 1) collection and pre-processing of M&A data, 2) construction of a knowledge distribution matrix, 3) identification of key knowledge categories by applying PCA, 4) recommending suitable knowledge areas by applying CF, 5) sophistication of recommendation results. A case study approach using data on the biotechnology M&A market was used to demonstrate the applicability of the framework. The proposed approach can demonstrate the possibility of using a firm's market-oriented knowledge portfolio as another objective data source to develop a basis for future business intelligence system. This would be particularly helpful for companies with scarce resources or service-dominant offerings to review their M&A opportunities.
From an expert systems point of view, the proposed framework is useful for supporting the M&A decision making processes of experts by a systematic approach. Using this knowledge base, the experts can expedite their initial screening and assessment for building a consensus among the stakeholders based on the degree of knowledge relatedness. This suits the zeitgeist of gleaning insights from Big Data analytics, as it can empower the subjective perception of experts.
The paper proceeds by reviewing the theoretical background on the overall research setting and applied methods in Section 2. The data and the proposed empirical strategy are discussed in Section 3, while Section 4 deals with its practical application to identify untapped M&A opportunities. Conclusions and outlook for future research are provided in Section 5.
Section snippets
M&A
In today's rapidly evolving business environment, firms are forced to continuously acquire new capabilities that meet the requirements and expectations of empowered customers and obtain external support to create marketable products (Lopez Flores, Belaud, Le Lann, & Negny, 2015). As well-established organizational instruments to respond to changes in the industry, M&A provide a means to quickly enrich the product or research pipeline by moving the affected firms outside their conventional
Methods
The analysis framework proposed in this paper involves the following analysis steps and is depicted as follows (Fig. 1):
Data
A case study approach was adopted to demonstrate the practical applicability of the proposed analysis framework. This allows more interpretative investigations to be conducted and can provide background information on the selected target firm's knowledge portfolio. M&A data were collected from the SDC (Securities Data Corporation) Platinum database, which is one of the most comprehensive sources of information on global corporate transactions such as mergers, acquisitions, strategic alliances
Conclusion
Although M&A are frequently used to strengthen the position of companies in an ever-changing business environment, the failure rate of M&A transaction is relatively high. Previous studies have emphasized the importance of assessing and valuing suitable acquisition target (Marks and Mirvis, 2001, Morrison et al., 2008), yet processes for discovering promising opportunities remain scarce in literature, with existing approaches tending to focus either on technological or financial aspects (McGee
Credit author statement
Both LJA and CHS conceived the idea and wrote the article. They also jointly contributed to acquisition, analysis and interpretation of the data. JL provided guidance and participated in editing of the final manuscript. All authors read and approved the final manuscript.
Declaration of interest
None.
Acknowledgments
The authors would like to acknowledge the kind support of the Helmholtz-Institute Münster for providing access to the database.
References (62)
- et al.
Mergers and acquisitions in Western European health care: Exploring the role of financial services organizations
Health Policy
(2012) - et al.
Why good things don't happen: The micro-foundations of routines in the M&A process
Journal of Business Research
(2015) - et al.
A hybrid content-based and item-based collaborative filtering approach to recommend TV programs enhanced with singular value decomposition
Information Sciences
(2010) - et al.
An intelligent fuzzy-based recommendation system for consumer electronic products
Expert Systems with Applications
(2007) - et al.
Social knowledge-based recommender system. Application to the movies domain
Expert Systems with Applications
(2012) - et al.
Mining changes in customer buying behavior for collaborative recommendations
Expert Systems with Applications
(2005) - et al.
Which app? A recommender system of applications in markets: Implementation of the service for monitoring users’ interaction
Expert Systems with Applications
(2012) - et al.
A novel recommendation model of location-based advertising: Context-aware collaborative filtering using GA approach
Expert Systems with Applications
(2012) - et al.
The competitive advantage potential of vertical integration
Omega
(1996) - et al.
Online recommender system for radio station hosting based on information fusion and adaptive tag-aware profiling
Expert Systems with Applications
(2016)