M&A for digital capabilities routinely fail less because of technology than because of managerial mindset and weak integration; a new 'AI Strategy Compass' and case evidence show governance choices and GenAI-enabled IS practices determine whether acquisitions drive or derail digital transformation.

Main Finding

AI — and especially Generative AI (GenAI) — can materially improve how firms pursue digital transformation via mergers & acquisitions (M&A), but technological potential alone is insufficient. Success depends on (1) the strategic approach organizations adopt for AI (a governance-driven vs. momentum-driven path) and their resulting digital intelligence maturity, (2) managerial cognition and microfoundations (thinking patterns) that enable or disable dynamic capabilities during M&A, and (3) realistic alignment of AI methods and data with the practical constraints of each M&A deal phase. Properly applied, GenAI affordances can turn post-merger knowledge loss into knowledge leverage (e.g., streamlining IS integration), yet many deals still fail because of non-digital mindsets, poor governance, and mismatches between academic AI proposals and M&A data realities.

Key Points

• Chapter 2 – Literature review ("Automating the Deal"):

  • Systematic interdisciplinary review mapping AI capabilities and proposed models across M&A deal phases (strategy, search, valuation, due diligence, integration, evaluation).
  • Identifies concentration of research on certain phases/capabilities, uneven model maturity, and important blind spots (data availability, real-world applicability, and organizational constraints).
  • Highlights a gap between AI research (often optimistic models) and the messy, scarce, and heterogeneous data available in practice.

• Chapter 3 – "AI Strategy Compass":

  • Develops a practitioner-oriented decision framework distinguishing two competing AI strategy logics:
    • Governance-driven: deliberate, controlled, compliance- and architecture-first.
    • Momentum-driven: experimentation-first, business-value fast-tracking.
  • Presents a Strategic Decision Tree and a Digital Intelligence Maturity Framework to help firms choose and operationalize a path while avoiding common pitfalls.

• Chapter 4 – Cognitive microfoundations of failure ("Ruining M&A-Based Digital Transformation"):

  • Through an in-depth physical–digital M&A case, identifies detrimental managerial thinking patterns (pre- and post-deal) that erode sensing, seizing, and reconfiguring capabilities critical for digital transformation.
  • Proposes a microfoundational process model linking individual/group cognition to dynamic capability gaps and explains why many digitally-motivated acquisitions fail despite clear strategic rationale.

• Chapter 5 – GenAI and post-merger IS integration ("From Knowledge Loss to Knowledge Leverage"):

  • Uses two comparative cases (Rip-and-Replace vs. Retain-and-Revive integration strategies; BetaCo and GammaCo) to show how GenAI affordances (e.g., summarization, semantic search, code/data translation) can mitigate knowledge loss and speed integration.
  • Offers a framework for applying GenAI depending on integration strategy and legacy-system conditions, detailing affordances, knowledge-integration effects, and trade-offs.

• Cross-cutting messages:

  • Technological affordances must be married to governance, organizational learning, and managerial cognition.
  • Practical implementation requires realistic assessments of data quality/availability and tailored AI models rather than indiscriminate adoption.
  • There are actionable tools (AI Strategy Compass, maturity framework, GenAI-integration decision rules) for practitioners to increase odds of successful digital M&A.

Data & Methods

• Multi-method dissertation combining:
  • Systematic interdisciplinary literature review: human-led search augmented by AI-assisted selection/validation; mapping papers across deal phases, AI capabilities, and model maturity (heatmaps, allocation analyses).
  • Multiple qualitative case studies:
    • At least three distinct empirical settings:
      • A physical–digital acquisition used to trace cognitive microfoundations of failure (PhysicalCo / DigitalCo).
      • Two post-merger IS-integration cases labeled BetaCo and GammaCo (rip-and-replace vs retain-and-revive).
    • Data sources include semi-structured interviews with managers and integration teams (lists of interviewees provided in appendices), internal documents, archival materials, and newspaper coverage (systematic article search for contextual triangulation).
  • Qualitative coding and analysis techniques:
    • Iterative inductive–deductive coding (Gioia-style data structuring is explicitly referenced), process tracing, cross-case comparison, and theory-building from case evidence.
    • Validation via multiple iterations, triangulation across data sources, and practitioner-oriented design of frameworks (AI Strategy Compass).
• Analytical artifacts and outputs:
  • Strategic decision trees, maturity frameworks, microfoundational process models, and applied affordance-to-effect mappings for GenAI in IS integration.
  • Tables linking M&A challenges to proposed AI models, and tables assessing data availability and empirical testing across phases.

Implications for AI Economics

• Efficiency and transaction costs:

  • GenAI and other AI tools can reduce information frictions and due-diligence costs (faster document summarization, semantic search, automated red-flagging), potentially lowering M&A transaction costs and shortening deal timelines — with implications for deal volume and market liquidity.
  • However, gains are conditional on data readiness and governance; where data are poor or fragmented, AI may deliver limited or misleading value.

• Valuation and market pricing:

  • Better AI-supported search, screening, and valuation tools (when grounded in realistic data) can improve price discovery and reduce asymmetric information. But managerial cognition and integration risk remain central determinants of ex-post value capture, so AI improvements in prediction do not fully eliminate realized-value uncertainty.

• Reallocation and productivity:

  • Successful digital M&A enabled by AI can accelerate reallocation of capabilities and labor across firms and sectors (incumbents acquiring digital assets, scaling capabilities). This could raise aggregate productivity if integration succeeds, but high failure rates mean potential deadweight losses if organizational and cognitive barriers persist.

• Returns to skills and capital:

  • AI changes the complementarity structure between managerial skills, digital talent, and capital. Firms with stronger governance, data infrastructure, and digitally literate leadership capture disproportionate AI gains, reinforcing winner-takes-most dynamics and potentially increasing returns to digital organizational capital.

• Policy and market design:

  • Regulators and standard-setters should note the role of data standards, transparency, and governance in realizing AI’s economic benefits in M&A. Policies that improve data interoperability and disclosure in transactions would help translate AI affordances into economic value.
  • Investors (PE, VC) should integrate assessments of managerial mindsets and integration capabilities — not just technological assets — into acquisition valuation and post-deal monitoring to better predict returns.

• Research directions for AI economics:

  • Need for large-N quantitative studies linking AI use in deal phases to realized post-merger performance to estimate causal effects on returns, productivity, and market outcomes.
  • Modeling the interaction between AI-driven reductions in search/valuation frictions and endogenous managerial/integration risk would clarify net welfare and distributional consequences.

Limitations noted in the dissertation include qualitative focus and case-based generalizability; empirical claims about macroeconomic impacts remain to be tested with large-scale data. Overall, the work highlights that AI’s economic effects in M&A hinge less on raw model performance and more on organizational strategy, governance, managerial cognition, and data realities.