How firms bundle R&D and marketing capabilities for innovation: a systematic review and contingency model
Giacomo ZatiniPurpose
This study systematically reviews 25 years of research on the R&D–marketing strategy nexus (2000–2025) and proposes a contingency model explaining when and why specific capability configurations enhance innovation performance.
Design/methodology/approach
We identified 140 peer-reviewed studies through a PRISMA-compliant search of Scopus, filtered by the ABS Academic Journal Guide. Each study was coded using the Antecedents–Decisions–Outcomes/Theories–Contexts–Methods (ADO-TCM) framework. Bibliometric mapping and Latent Dirichlet Allocation (LDA) topic modelling complemented the deductive coding with inductive identification of thematic clusters. A co-word robustness check validated the LDA solution.
Findings
Marketing capability, network relationships and technology capability rank as the most frequently studied antecedents, with organisational structure and environmental dynamism close behind; market orientation, which earlier work placed at the centre, is coded in fewer than one study in ten. Marketing-strategy and product-development decisions dominate the corpus almost equally, while temporal sequencing of orchestration choices remains unstudied. Two in five studies invoke no explicit theory; evidence concentrates in the USA and East Asia and in manufacturing or multi-industry samples, with services almost absent. Integrating ADO–TCM patterns with LDA clusters, we derive three testable propositions: a turbulence–agility mechanism, a digital–complementarity mechanism and a legitimacy–consistency mechanism. We introduce the concept of integration intelligence, the meta-capability through which managers diagnose which orchestration configuration fits their current competitive context.
Originality/value
The paper reframes R&D–marketing coordination as capability orchestration rather than functional integration, fusing the Dynamic Capabilities View with Resource Orchestration Theory. The resulting contingency model moves beyond the question of whether integration improves performance toward specifying which configurations improve performance under which conditions, offering both a theoretical architecture and a research agenda organised around generative AI, platform ecosystems and sustainability transitions.