Why Serial Business Builders Outperform Everyone Else and How to Become One

There is a particular kind of organizational hubris that treats innovation as an event. A company convenes a task force, funds a skunkworks project, generates a press release, and then waits for the market to respond with gratitude. When the venture underperforms, the organization concludes, with some relief, that building new businesses is simply not its core competency. The experiment is quietly defunded, the team is reabsorbed, and the status quo reasserts itself.

New research from McKinsey & Company, presented by Senior Partner Jason Bello, offers a pointed rebuttal to this pattern. The central finding is straightforward but carries significant strategic weight: organizations that repeatedly build new ventures—specifically those running three or more simultaneously—dramatically outperform those that treat venture building as a one-off experiment. This is not a marginal performance gap. It is the difference between organizations that are building durable growth engines and those that are merely performing the theater of innovation.

The McKinsey report arrives at a moment when the economics of business building have shifted considerably. The average cost to reach breakeven on a new corporate venture fell from approximately $125 million in 2024 to $77 million in 2025. Bello attributes this compression to two forces: tighter capital discipline that is forcing teams to find customer impact within 12 to 18 months, and the accelerating integration of artificial intelligence into every stage of the building process. Both forces are structural, not cyclical. Leaders who are waiting for a more favorable environment to experiment with new ventures are, in practice, simply waiting to fall further behind.

The Muscle Memory of Building

The most persuasive argument in the McKinsey research is also its most intuitive: building businesses is a skill, and like all skills, it improves with repetition. Bello frames this as developing organizational muscle memory. The first time a company attempts milestone-based funding, quarterly OKR accountability cycles, or the discipline of killing a venture early when it misses customer metrics, the process feels foreign and uncomfortable. Bureaucratic immune systems activate. Annual budget cycles resist the logic of tranche-based investment. Senior leaders accustomed to measuring performance annually struggle with the quarterly cadence that venture building demands.

Serial builders have absorbed these tensions and metabolized them. They have built the governance structures, the talent pipelines, and the cultural permission to operate differently inside the larger organization. Critically, they have also developed comfort with portfolio-level risk management, understanding that not every venture will succeed, just as not every M&A transaction will generate the projected returns. This comfort with managed failure at the portfolio level is precisely what allows serial builders to kill underperforming ventures decisively rather than sustaining them out of institutional embarrassment.

This finding aligns closely with research published by Rita McGrath of Columbia Business School, who has argued extensively that the organizations best positioned for sustained growth are those that treat innovation as an ongoing process rather than a periodic event. McGrath’s work on transient competitive advantage suggests that in markets where product cycles are shortening, the ability to continuously launch new offerings is itself the competitive moat, not any single product or service. The McKinsey data provides a quantitative dimension to this argument: serial builders are not simply more innovative in spirit; they generate measurably superior returns on their innovation investments.

AI Is Not Just a Tool Here; It Is a Business Model

One of the more striking dimensions of the McKinsey research is the dual role that artificial intelligence now plays in corporate venture building. Bello draws a useful distinction between AI as a business and AI within the business of building. Both deserve serious attention from leadership teams.

The first category describes what an increasing number of established companies are attempting: building an AI-native version of their core business as a separate venture. Bello’s example of a bank constructing a fully agentic B2B banking platform alongside its traditional operations captures the logic precisely. The new entity does the same fundamental work as the parent, but with KYC processes, customer service functions, and fund flows all automated and agentic. The goal is not necessarily to replace the core business immediately, but to build the capability at speed, learn what works, and eventually reintegrate the new model into the broader organization.

This approach reflects a genuine strategic insight: incumbents who are unwilling to cannibalize their own business models will eventually find that someone else has done it for them. Amazon’s development of AWS alongside its retail operations is perhaps the most cited example of this logic in practice. More recently, JPMorgan Chase’s investment in its AI infrastructure and the creation of separate technology ventures within its corporate structure has demonstrated that even heavily regulated industries are not exempt from this imperative.

The second category, AI within the building process itself, is where the productivity gains are most immediately tangible. The compression of breakeven costs from $125 million to $77 million is substantially a story about AI-enabled speed and efficiency. Prototyping cycles that once took weeks now take days. Bello describes a shift from presenting clickable wireframes to clients to arriving with fully vibe-coded functional prototypes. Business case modeling, survey design, marketing material creation, A/B testing frameworks, and even early-stage functional roles that once required full-time hires can now be substantially automated. The cumulative effect is that the cost of being wrong early in the venture-building process has dropped significantly, which in turn lowers the barrier to experimentation.

A Harvard Business Review analysis by Marco Iansiti and Karim Lakhani on the AI-driven transformation of firm economics supports this observation. Their research on AI-native firms demonstrates that companies embedding AI at the process level, rather than simply layering it onto existing workflows, achieve substantially different unit economics. The McKinsey data on declining venture breakeven costs appears to reflect exactly this dynamic playing out in the corporate venture-building context.

The Portfolio Mindset as Organizational Discipline

Perhaps the most practically actionable element of the McKinsey research is its emphasis on the portfolio mindset. Bello is explicit: the organizations that succeed at venture building are not those that pursue the broadest possible range of bets, but those that take multiple shots on goal against a single strategic objective. The distinction matters. Spreading innovation investment across unrelated bets dilutes organizational focus and prevents the accumulation of domain expertise. Concentrating multiple experiments around a coherent strategic problem generates learning that compounds.

This is a point that deserves more direct challenge than Bello offers in the conversation. The portfolio mindset carries a real risk of becoming an organizational comfort blanket. If every venture failure can be reframed as a portfolio-level learning event, organizations may use the framework to avoid confronting the deeper question of whether their venture-building capability is actually improving. A portfolio of three ventures that all fail for the same underlying reason—insufficient customer insight, organizational resistance to the new model, or chronic underfunding—is not a portfolio strategy. It is repeated error with better branding.

Effective portfolio management in the venture-building context requires genuine post-mortems that generate transferable learning, not just narrative reframing. Bello’s example of a client that discovered its target market was smaller than projected, pivoted to a narrower build, and ultimately succeeded is instructive precisely because the failure led to a revised hypothesis that was then tested. That is the correct mechanism. Organizations that treat portfolio diversification as a substitute for rigorous learning will find that their third or fourth venture fails for the same reasons as their first.

Research published in the Strategic Management Journal by scholars studying corporate venture capital programs reinforces this caution. Organizations that built systematic knowledge management processes around their venture portfolios—capturing lessons from both successes and failures in transferable form—significantly outperformed those that treated each venture as a standalone experiment. The accumulation of institutional knowledge, not merely the accumulation of experience, is what generates the serial builder advantage.

Structural Isolation and Cultural Permission

Bello’s prescription for creating the conditions in which new ventures can succeed inside large organizations is both pragmatic and consistent with the broader organizational behavior literature. New ventures need to be structurally isolated from day-to-day management pressures, funded through dedicated tranches rather than absorbed into existing budget cycles, and given direct lines of reporting to C-level leadership. Without these structural protections, ventures will be perpetually raided for talent, subordinated to quarterly earnings pressures, and eventually dissolved when the next planning cycle demands budget reallocation.

The cultural dimension is equally critical and considerably harder to engineer. Bello’s characterization of the required leadership posture—fact-based, psychologically safe, comfortable with early failure, and genuinely present at the product level—describes a set of behaviors that run counter to the default operating mode of most large organizations. Senior leaders in established firms are typically rewarded for protecting margin, managing risk, and delivering predictable results. Venture building demands tolerance for ambiguity, appetite for managed risk, and willingness to publicly acknowledge when initial assumptions were wrong.

Amy Edmondson’s foundational research on psychological safety at Harvard Business School provides the academic scaffolding for Bello’s cultural prescriptions. Edmondson’s work demonstrates that teams operating in psychologically safe environments are more likely to surface early warning signals, share negative findings, and course-correct before small problems become expensive failures. In the venture-building context, where the ability to pivot quickly is a core competitive advantage, psychological safety is not a soft cultural aspiration. It is a hard operational requirement.

The challenge for most organizations is that psychological safety at the venture team level is insufficient if it is not matched by tolerance at the leadership level. A team that feels comfortable surfacing bad news to its immediate manager but knows that the senior leadership team will punish failure regardless will eventually learn to manage upward rather than manage the venture. Bello’s insistence that the culture shift must come from the top is not merely good advice. It is a structural necessity.

The Case for Building Now, Not Later

For leaders inclined to defer venture-building investment during periods of economic uncertainty, the McKinsey research presents a direct counterargument. Periods of disruption are precisely when established organizations are most vulnerable to displacement by attackers who are willing to build AI-native versions of the incumbents’ own business models. The cost of not building is not zero. It is the accumulated competitive disadvantage of watching new entrants institutionalize operating models that are structurally cheaper, faster, and more responsive than anything the incumbent currently operates.

Moreover, the declining cost of venture building, driven by AI-enabled acceleration, means that the capital required to get meaningful early validation has never been lower. Organizations can now obtain 30 to 50 percent of the informational value of a full venture build within the first 12 to 18 months at a fraction of historical costs. The risk calculus has genuinely shifted. What once required a $125 million commitment to reach breakeven now requires $77 million, and that number is likely to continue declining as AI tooling matures.

The organizations that will look back on the current period as a moment of strategic clarity are those that treated this window as an opportunity to build the venture-building muscle while their competitors were paralyzed by uncertainty. Serial builders do not wait for perfect conditions. They build the capability in imperfect ones and arrive at the next competitive cycle with institutional knowledge that cannot be easily replicated.

What Leaders Should Do Differently Starting Now

The practical implications of the McKinsey research reduce to a small number of high-leverage actions.

• Reframe venture building as a repeatable capability rather than a periodic initiative. Invest in the governance structures, funding mechanisms, and talent development pipelines that allow the organization to run multiple ventures simultaneously.
• Anchor each portfolio of ventures around a coherent strategic problem rather than distributing bets across unrelated opportunities. Focused experimentation generates compounding learning; unfocused experimentation generates noise.
• Integrate AI into the venture-building process at every stage, from ideation and market validation through prototype development and business case modeling. The organizations that are already doing this are building a speed advantage that will be difficult to close for those who wait.
• Create structural isolation and direct C-level sponsorship for every active venture. Without these protections, the organizational immune system will ultimately prevail.
• Build the cultural infrastructure for fact-based learning. Establish norms that distinguish between competent failure (well-executed work that revealed a market assumption was wrong) and incompetent failure (breakdowns in basic execution discipline). Celebrate the former as organizational learning and address the latter with the same rigor you would apply to any other operational underperformance.

The serial builder advantage is real, it is measurable, and it is accumulating right now in the organizations that have committed to building as a discipline rather than an event. The question for every leadership team is not whether the business environment is favorable enough to justify the investment. The question is how many ventures behind they are willing to fall before they start.