Building AI Capabilities for Business Value Creation: A Case Study on Cross-Functional Teams at Swisscom

Topic

This thesis investigates how organizations can overcome major challenges in building AI capabilities by focusing on cross-functional collaboration. It specifically explores collaboration not within isolated teams, but across business units, functional areas, and hierarchical levels. This firm-level perspective reflects the real-world complexity of AI implementation in large organizations, where coordination, integration, and strategic alignment takes place across diverse processes and responsibilities.

Relevance

Artificial Intelligence (AI) has become a cornerstone of digital transformation. However, many organizations struggle to move AI initiatives beyond pilot projects and achieve scalable impact. This is rarely due to technological immaturity alone. Rather, it often stems from fragmented organizational structures, a lack of clear governance, and ineffective collaboration between technical and business functions. Understanding how to build AI capabilities in a structured and cross-functional way is therefore crucial for translating AI potential into sustainable business value.

Findings

The study developed an AI Capability Model that describes how organizations can build AI capabilities through three interconnected dimensions.

• Centralized capabilities ensure consistency, scalability, and strategic governance of AI across the organization.
• Decentralized capabilities enable flexibility, innovation, and context-specific adoption of AI in different teams and units.
• Transversal capabilities connect all parts of the organization to foster alignment, collaboration, and shared learning.

The AI Capability Model shows how collaboration between different parts of an organization can be effectively supported to generate long-term business value.

Implications for Practitioners

• AI capability building requires a structured interplay between central, decentral, and transversal teams. Organizations should ensure that infrastructure, roles, and collaboration mechanisms are aligned to enable both top-down guidance and bottom-up innovation.
• A successful AI transformation depends on an AI strategy that is deeply integrated into the organization’s governance, investment logic, and operational routines. Leaders must ensure that strategic goals are translated into actionable priorities at all levels.
• To move beyond isolated use cases, organizations must design AI capabilities that are scalable across the organization yet adaptable to local contexts. This includes empowering teams to act autonomously while leveraging shared standards and infrastructure.

Methods

The study is based on a qualitative and exploratory single-case study, carried out in collaboration with Swisscom. A total of 14 semi-structured interviews were conducted with experts from different business units and functions. The analysis of the interview data took place in two steps. First, a qualitative content analysis was conducted following the method of Mayring and Fenzl (2022). This step used predefined categories from the Organizational Framework of AI and Business Value by Enholm et al. (2022) to guide the analysis. In the second step, inductive coding based on the approach by Gioia et al. (2014) was used to identify new themes that emerged from the data.