Some firms have successfully maintained their position as sales leaders for years, as evidenced by their presence on the annual Fortune 500 list. While sustained investments in both marketing and R&D are crucial to sales leadership maintenance, should firms invest more in marketing or R&D? Is the relative emphasis on marketing versus R&D for sales leadership maintenance contingent upon environmental and firm-related factors? We address these questions by building on resource-based and organizational adaptation theories. We present a conceptual model and hypotheses delineating the main effects of marketing and R&D capital, and the moderating effects of environmental dynamism, environmental munificence, and financial leverage on sales leadership maintenance. We use a left-truncated Cox proportional hazards survival model to test the hypotheses on an unbalanced panel dataset of 114 Fortune 500 manufacturing firms tracked over the period 1981 to 2016. We find that both marketing capital and R&D capital have a direct and positive effect on sales leadership maintenance. We also find that environmental dynamism and financial leverage interact with R&D capital, but not marketing capital, to enhance the probability of sales leadership maintenance; and investing more in R&D than in marketing enhances sales leadership maintenance in dynamic environments for highly leveraged firms. A sales leader firm that incrementally spends 1% of its five-year average sales revenue on each of marketing and R&D activities can improve its probability of sales leadership maintenance by 50%.
We acknowledge that other measures of marketing and R&D capabilities such as brand equity and number of patents may also influence sales leadership maintenance. However, the lack of continuous data on these for the period 1981 to 2016 precludes us from including these measures in our subsequent empirical analysis.
Incumbents may be just as adept as newcomers in introducing radical innovations (Sorescu, Chandy, and Prabhu 2003). However, this observation might be specific to an industry such as studied in Sorescu et al. (2003).
In the robustness check section, we use dispersion around a regression trend line as an alternative measure, consistent with Dess and Beard (1984) and Jindal and McAlister (2015). The results are substantively similar.