Sönke Albers

PLS and Success Factor Studies in Marketing

While in consumer research the “Cronbach’s α – LISREL”-paradigm has emerged for a better separation of measurement errors and structural relationships, it is shown here that studies involving an evaluation of the effectiveness of marketing or organizational strategies based on structural relationships require the application of PLS. This is because we no longer distinguish between constructs and their reflecting measures but rather between abstract marketing policies (constructs) and their forming detailed marketing instruments (indicators). It is shown with the help of examples from literature that many studies of this type applying LISREL have been misspecified and would have better made use of the PLS approach. I also demonstrate the appropriate use of PLS in a study of success factors for e-businesses. I conclude with recommendations on the appropriate design of success factor studies, including the use of higher-order constructs and the validation of such studies.

Personal Selling Elasticities: A Meta-Analysis

This article presents a meta-analysis of prior econometric estimates of personal selling elasticity—that is, the ratio of the percentage change in an objective, ratio-scaled measure of sales output (e.g., dollar or unit purchases) to the corresponding percentage change in an objective, ratio-scaled measure of personal selling input (e.g., dollar expenditures). The authors conduct a meta-analysis of 506 personal selling elasticity estimates drawn from analyses of 88 empirical data sets across 75 previous articles. They find a mean estimate of current-period personal selling elasticity of .34. They also find that elasticity estimates are higher for early life-cycle-stage offerings, higher from studies set in Europe than from those set in the United States, and smaller in more recent years. In addition, elasticity estimates are affected significantly by analysts’ use of relative rather than absolute sales output measures, by cross-sectional rather than panel data, by omission of promotions, by lagged effects, by marketing interaction effects, and by the neglect of endogeneity in model estimation. The method bias–corrected mean personal selling elasticity is approximately .31. The authors discuss the implications of their results for sales managers and researchers.

A procedure for new product positioning in an attribute space

Abstract Planning a preferred position of a new product in an attribute space may be considered as a mixed integer non-linear programming problem. A procedure to solve this problem is described which draws on some special features of the problem. Optimality considerations of the procedure as well as computational experiences with it are reported.

Patient- or Physician-Oriented Marketing: What Drives Primary Demand for Prescription Drugs?

The authors analyze primary demand effects of marketing efforts directed at the physician (detailing and professional journal advertising) versus marketing efforts directed at the patient (direct-to-consumer advertising). The analysis covers 86 categories, or approximately 85% of the U.S. pharmaceutical market, during the 2001–2005 period. Primary demand effects are rather small, in contrast with the estimated sales effects for individual brands. By using a new brand-level method to estimate primary demand effects with aggregate data, the authors show that the small effects are due to intense competitive interactions during the observation period but not necessarily to low primary demand responsiveness. In contrast with previous studies, the authors also find that detailing is more effective in driving primary demand than direct-to-consumer advertising. A category sales model cannot provide such insights. In addition, a category sales model likely produces biased predictions about period-by-period changes in primary demand. The suggested brand-level method does not suffer from these limitations.

Methodische Probleme bei der Erfolgsfaktorenforschung — Messfehler, formative versus reflektive Indikatoren und die Wahl des Strukturgleichungs-Modells

ZusammenfassungDieser Beitrag ist ein methodisches Plädoyer zur Verwendung adäquater Methoden in der empirischen Erfolgsfaktorenforschung. Wir zeigen die methodischen Probleme bei der Anwendung der LISREL-Methodologie auf empirische Erfolgsfaktorenstudien auf. Die Trennung in latente Konstrukte und reflektive Indikatoren, um zwischen einem Mess- und Strukturmodell unterscheiden zu können, und die Eliminierung von intern nicht konsistenten Indikatoren ist falsch, wenn es sich wie bei Erfolgsfaktorenstudien häufig um formative Indikatoren handelt, aus denen sich die Konstrukte ergeben. Durch die Wahl der falschen Messstruktur werden wichtige Facetten des Konstrukts vernachlässigt, und man erhält inhaltlich andere Ergebnisse. Bleibt man in der Strukturgleichungs-Methodologie, so empfiehlt sich die Anwendung von PLS, da dort die Indikatoren sowohl formativ als auch reflektiv spezifiziert werden können und geringere Anforderungen an die Daten gestellt werden. Mit Hilfe eines Beispiels wird demonstriert, dass sich insbesondere durch die Eliminierung von Indikatoren und weniger durch die Schätzmethode stark unterschiedliche Ergebnisse ergeben. Hier wird allerdings argumentiert, dass man im Falle ausschließlich formativer Indikatoren zu Regressionsanalysen zurückkehren sollte, weil man Multikollinearität durch Indexbildung behandeln und zusätzlich Nichtlinearitäten und unbeobachtete Heterogenität berücksichtigen kann.SummaryWe demonstrate methodological problems with the application of the so called “Cronbach’s α — LISREL”-paradigm. Working with unobserved latent constructs operationalized through indicators allows for differentiating into a measurement and structural model. However, while the elimination of internally inconsistent indicators is correct for reflective indicators, this proves to be incorrect for the mostly formative indicators in success factor studies. This neglects important facets of the problem which lead to results that vary according to the included indicator set. In this context, the analysis of structural equation models with LISREL is not appropriate because it can handle only reflective indicators while PLS allows for formative indicators. However, in the latter case it is often better to apply ordinary least squares regression to the battery of indicators because multicollinearity has to be dealt with in both, PLS and OLS, while regression models can handle nonlinear relationships as well as unobserved heterogeneity.

An extended algorithm for optimal product positioning

Abstract Optimal product positioning in an attribute space has been formulated according to the axiom of choice as a mixed integer nonlinear programming problem. To solve it, Albers and Brockhoff have designed the special purpose algorithm PROPOSAS. It works under simplified assumptions: Euclidean metric, equally weighted dimensions of the attribute space and equal sales per customer. The following article shows that the basic ideas of PROPOSAS are flexible enough to be expanded to cover a weighted Minkowski-metric as well as different revenues from the customers. Furthermore, the calculation of a new upper bound is described which reduces CPU-time considerably.

Determinants of Performance in Customer Relationship Management — Assessing the Technology Usage-Performance Link

The management of customer relationships has become a top priority for companies in the last years. Despite this, little is known about the factors of successful CRM implementations and the role of information technology in this context. This study provides three models for the explanation of CRM performance separated according to the customer lifecycle phases initiation, maintenance, and retention. We analyze the relationship between CRM technology, CRM technology usage and CRM performance. In addition, we identify the drivers that affect implementation success for each phase the most.

Practice Prize Winner---Dynamic Marketing Budget Allocation Across Countries, Products, and Marketing Activities

Previous research on marketing budget decisions has shown that profit improvement from better allocation across products or regions is much higher than from improving the overall budget. However, despite its high managerial relevance, contributions by marketing scholars are rare. In this paper, we introduce an innovative and feasible solution to the dynamic marketing budget allocation problem for multiproduct, multicountry firms. Specifically, our decision support model allows determining near-optimal marketing budgets at the country--product--marketing--activity level in an Excel-supported environment each year. The model accounts for marketing dynamics and a products growth potential as well as for trade-offs with respect to marketing effectiveness and profit contribution. The model has been successfully implemented at Bayer, one the worlds largest pharmaceutical and chemical firms. The profit improvement potential is more than 50% and worth nearly €500 million in incremental discounted cash flows.

Optimization models for salesforce compensation

Abstract This paper reviews models which address the design of optimal salesforce compensation schemes. The discussion starts with agency theoretic models that are able to derive the optimal form for a compensation scheme under assumptions regarding the degree of information asymmetry and heterogeneity within the salesforce. For the four resulting extreme cases, models are presented that allow for the determination of optimal parameter values for specific compensation schemes such as fixed salary plus commission, bonus for quota achievement, and menus of plans. Finally, there follows an examination of approaches through which commission rates or bonuses can be optimally differentiated, such that the salespersons allocate their selling time in their own - and at the same time in the firms - best interest. All of these models are reviewed in particular with respect to their applicability in practice.

Misleading Rankings of Research in Business

Abstract The attempts by Schulze and colleagues and Ritzberger to develop a joint ranking of journals for economics and business research are critically evaluated. Their lists suggest that the quality of top business journals is substantially lower than that of many economics journals. If, however, the authors of these lists do not want to claim a general superiority of one discipline (economics) over another one (business), they should give a clear indication that these lists are only applicable for economists. This warning appears to be necessary because Fabel and colleagues derive a ranking of universities and departments with respect to research productivity in business from the business research discriminating list RbR_IMP by Schulze and colleagues. While Diamantopoulos and Wagner already show a lack of face validity of these results, this article explains that the reason for this lies not only in the downgrading and also biased weighting of the business journals across subfields, but even more importantly, in a remarkable incompleteness of the database.