The session was led by Michelle Gonzalez, corporate vice president and global head of M12, Microsoft’s venture arm, alongside Nicolas Sauvage, president of TDK Ventures. Both are experienced CVC leaders with a track record of linking venture investing to corporate strategy and innovation outcomes.
The following points summarise the key themes for corporate venture investors that emerged from their discussion.
- CVC as a structured discovery engine, not a passive investment arm
- Leading CVCs are increasingly positioning themselves as proactive “discovery machines” rather than reactive investors.
- This involves systematically identifying emerging technologies before they reach internal product roadmaps, which tend to operate on shorter time horizons.
- Thesis-driven and ‘deep exploration’ approaches are becoming best practice
- M12 emphasises multi-year, thesis-led investing (e.g., AI systems pre-ChatGPT), enabling early conviction in underexplored areas.
- TDK Ventures formalises this as “deep explorations”, with the explicit goal of investing up to four years before markets become obvious.
- These processes generate three outcomes: invest, park (monitor), or learn without investing—each still delivering strategic value.
- Information advantage is a core differentiator
- CVCs aggregate “mosaic” insights from multiple startup interactions, producing proprietary intelligence for corporate leadership.
- This insight loop — startup → CVC → corporate — can materially influence long-term strategy, particularly in fast-moving domains such as AI.
- Translating insight into corporate impact requires disciplined communication
- Short, executive-friendly formats (e.g., two-page briefs) outperform lengthy reports in driving engagement at C-suite level.
- Curated interactions (e.g., bringing startups directly to executives) can catalyse strategic shifts, especially when timed ahead of inflection points.
- CVC can shape corporate strategy, but timelines are long
- Examples cited include influencing data centre strategy (AI systems, token economics) and enabling new product lines (e.g., advanced manufacturing collaborations).
- In deep tech sectors, the feedback loop from investment to corporate impact may take several years.
- ‘Mothership advantage’ remains the defining competitive edge
- Corporates offer unique assets — technical expertise, customer access and problem definition — that traditional VCs cannot replicate.
- The most effective CVCs actively productise this advantage (e.g., structured portfolio support teams, concierge-style access to internal resources).
- Winning competitive deals requires a clear value proposition
- CVCs must “sell” themselves to top startups and co-investors, particularly in oversubscribed rounds.
- Successful approaches include tailored value propositions (“art of the possible”) and demonstrable case studies of portfolio support.
- Entrepreneur-centric operating models improve effectiveness
- TDK Ventures explicitly defines entrepreneurs — not the parent company — as the primary customer.
- This clarity enables consistent behaviour and the use of metrics such as NPS to evaluate value delivered to founders.
- CVC is shifting from ‘nice-to-have’ to strategic necessity
- Rapid technological change — particularly in AI — is compressing innovation cycles and raising the cost of being late.
- Panel consensus: large corporates (especially those with >$1bn revenue and sufficient cash flow) can no longer afford to operate without a CVC capability.
Implications for CVC leaders
- Build structured, thesis-led discovery processes rather than relying on deal flow alone.
- Invest in internal knowledge translation — insight only matters if it changes decisions.
- Systematise and evidence corporate value-add to remain competitive in top-tier deals.
- Align incentives clearly: ambiguity over whether the “customer” is the startup or the corporate weakens execution.
- Treat CVC as a core strategic function in an era of accelerated disruption, not a discretionary innovation budget.
This summary was generated by AI and lightly edited by GCV staff.


