Marketing Spending Models - Innovation and Marketing in the Pharmaceutical Industry - page 567

level (Manchanda and Chintagunta 2004 ; Mizik and Jacobson 2004 ; Montoya et al.
2010 ) are smaller than those from studies using aggregate data. It should also be
noted that in some cases,the authors encountered negative ROIs pointing to an over-
spending on detailing (Chintagunta and Desiraju 2005 ; Manchanda and Chintagunta
2004 ; Mizik and Jacobson 2004 ; Wittink 2002 ).
A few studies considered the ROI of other physician-oriented marketing efforts
such as journal advertising or invitations to dinners and meetings (Neslin 2001 ;
Wittink 2002 ). These studies conclude that ROIs of these activities are highly posi-
tive and firms therefore should consider raising or reallocating their budgets to them.
In contrast to physician-oriented efforts, negative ROIs are found with respect to
DTC advertising unless the drug is classified as a recently introduced large brand
(revenues > US$ 500 million). Narayanan et al. ( 2004 ) consider such brands in their
sample and provide evidence that DTC advertising pays off at current spending
levels. In addition, the authors analyze the interaction with other communication
activities and price and find significant effects that alter ROIs for both DTC adver-
tising and detailing efforts.
ROI-analysis provides important insights into the effectiveness of marketing
activities in monetary terms. The analysis helps identify ineffective marketing poli-
cies that generate negative ROIs. Neslin ( 2001 ) and Wittink ( 2002 ) show that real-
locating resources from DTC advertising to physician-oriented activities can
substantially increase profits. However, an isolated ROI-analysis has its limitations.
First, it does not inform how large the budget for a marketing variable should be to
maximize total profits. Second, the growth potential of a product due to factors such
as life cycle is not reflected. Third, in models where interactions across marketing
instruments are specified, the ROI strongly depends on the level of investment of the
other instruments. The negative ROI of a marginal DTC advertising dollar, for
example, may turn into a positive ROI if the level of investment for other variables
is optimized. Finally, single product ROI-analysis does not consider different profit
improvement potentials across products or consumer groups. Optimal allocation
decisions may not be derived from ROI-analysis, but have to consider additional
parameters such as the scale of the revenue base and the profit contribution margin
(Fischer et al. 2011b ).

19.4 Setting Optimal Marketing Budgets

The key question in normative research on marketing spending models is how to
determine the profit maximizing marketing budget. Structurally, this involves two
further questions. What is the optimal total marketing budget? And how should that
budget be optimally allocated across allocation units that may be countries, prod-
ucts, customers, activities, time periods, and combinations of these. In theory, both
problems should be solved simultaneously yielding the optimal total budget that is
allocated in an optimal manner. Apart from the technical challenges associated with
a simultaneous solution, management practice usually solves these problems