
Here is a statistic that should concern every executive who has ever funded a hackathon, launched an innovation lab, or greenlit a pilot project: 96% of innovation projects fail to make a return on investment, according to research from Deloitte's Doblin unit. That is not a rounding error. That is a system-level failure rate that would be unacceptable in virtually any other area of corporate spending.
And yet, the appetite for innovation has never been higher. McKinsey reports that 84% of executives view innovation as crucial to their growth strategy. The problem is not ambition. It is execution. Only 6% of those same executives are satisfied with their organization's innovation performance.
The gap between ambition and execution is where billions of dollars go to die. But it does not have to be this way. Companies that shift from one-off innovation projects to repeatable venture building capabilities — often delivered through a corporate venture studio — are generating dramatically better returns. The data is now clear enough that the question is no longer whether to build ventures, but whether you are building them the right way.
Most large organizations have experimented with innovation in some form. They have run design sprints, hosted hackathons, set up innovation labs, and funded intrapreneurship programs. These efforts create visibility and excitement. They generate internal press releases and conference presentations. But they rarely produce scalable businesses.
Deloitte's research puts it bluntly: 91% of innovation departments confirm that the phase of visibility-first innovation is ending. The corporate world is waking up to the fact that innovation theater — activities that look like innovation but do not produce commercial outcomes — is an expensive distraction.
The pattern is familiar. A company runs a hackathon. A few promising ideas emerge. The winning team gets a small budget and a vague mandate. Six months later, the project has stalled. The team members have been pulled back to their day jobs. The idea sits in a slide deck, never to be revisited. As we explored in Why Most Innovation Systems Stall, the structural problems are predictable: no dedicated resources, no clear governance, no path from prototype to market.
The core issue is that one-off projects lack the infrastructure to turn ideas into businesses. They are designed to generate ideas, not to build ventures. And building a venture requires a fundamentally different set of capabilities: customer discovery, product development, go-to-market strategy, and the operational discipline to iterate quickly based on market feedback.
McKinsey's 2025 research on corporate venture building reveals a striking pattern. Companies that build ventures repeatedly — serial builders — generate 1.9x revenue per dollar invested, compared to just 1.3x for companies that attempt one-off builds. That is not a marginal difference. It is the difference between a venture building program that pays for itself and one that does not.
The advantage compounds over time. McKinsey's 2024 data shows that expert builders achieve 12x the revenue of novices over a five-year period. The implication is clear: venture building is a skill that improves with practice. Organizations that treat it as a one-time experiment never develop the muscle memory required to succeed consistently.
There is also a cost advantage. McKinsey's 2025 study found that the weighted average investment to break even dropped from $125 million to $77 million for serial builders. Repeatability does not just improve outcomes — it reduces the capital required to achieve them. Serial builders learn what works, reuse proven frameworks, and avoid the expensive mistakes that plague first-time efforts.
McKinsey's venture factory model identifies three building blocks that separate repeatable programs from ad hoc efforts:
This is also the logic behind the ambidextrous organization, a concept developed by Charles O'Reilly and Michael Tushman at Harvard Business School. Their research shows that the most successful companies simultaneously exploit existing businesses and explore new opportunities — but only when they create structurally separate units with their own processes, metrics, and culture. Bolting innovation onto an existing business unit almost never works.
Alphabet's X division is perhaps the most visible example of a venture factory in action. X does not run one-off experiments. It operates standardized resources, tools, and processes for spinning out new ventures. The results speak for themselves: Waymo, Wing, Verily, and Chronicle all emerged from this repeatable system. Each venture benefited from shared infrastructure while maintaining the autonomy to operate like a startup.
Procter & Gamble Ventures offers a different but equally instructive model. P&G built a systematic pipeline for creating new brands in categories where the company does not currently compete. Zevo, an insect control brand, and AeroFlexx, a flexible packaging venture, both emerged from this process. The key insight is that P&G did not ask its existing brand teams to innovate on the side. It created a dedicated engine with its own mandate and resources.
In the corporate venture studio world, Highline Beta has helped multiple organizations make this transition from ad hoc innovation to repeatable venture building. The approach centers on building the systems, processes, and capabilities that allow a company to launch ventures consistently — not just once.
When ZX Ventures — the innovation and investment arm of AB InBev — wanted to move beyond one-off experiments, they partnered with Highline Beta to create a repeatable program for venture building. The goal was not to launch a single venture. It was to build a capability that could be deployed again and again across different markets and opportunity areas. The result was a structured program with clear stage gates, reusable playbooks, and a pipeline that could evaluate multiple opportunities simultaneously.
Westfield Insurance took the concept even further. Through its 1848 Ventures initiative, Westfield worked with Highline Beta to build a dedicated corporate venture studio. This was not a side project or a temporary program. It was a permanent organizational capability with its own team, budget, and mandate to build new ventures that could leverage Westfield's core strengths in insurance and financial services.
What both examples share is a deliberate shift from project-based thinking to capability-based thinking. Instead of asking "what venture should we build?", these organizations asked "how do we build the engine that produces ventures?" That reframing is the critical difference between companies that generate isolated wins and companies that generate sustained growth through venture building.
One reason ad hoc innovation persists is that organizations struggle to measure the performance of venture building programs. When every project is unique, there is no baseline for comparison. When there is no baseline, executives default to vanity metrics: number of ideas generated, employee engagement scores, press mentions.
Repeatable programs solve this by creating consistent metrics across a portfolio of ventures. As we outlined in our Innovation Measurement Framework, effective measurement requires tracking leading indicators (customer discovery velocity, experiment throughput, validation milestones) alongside lagging indicators (revenue, market share, strategic value created).
When you build ventures repeatably, you accumulate data that makes each subsequent decision smarter. You learn which types of opportunities are most promising for your organization. You learn which validation methods are most predictive of market success. You learn how to allocate capital more efficiently across a portfolio. None of this is possible when every innovation effort starts from scratch.
A venture studio is not the only way to build ventures repeatably, but it is increasingly the preferred vehicle for corporations that want to move fast. Studios bring several structural advantages that are difficult to replicate through internal teams alone:
Ed Essey's experience building Microsoft's innovation engine — discussed on the Highline Beta podcast — illustrates the scale of what is possible when you systematize innovation. Essey built a system capable of processing 20,000 ideas, not by evaluating each one individually, but by creating a scalable pipeline with clear criteria and stage-gated processes. The system, not the individual heroics, produced the results.
The shift from one-off innovation projects to a repeatable venture building engine requires three commitments:
First, commit to building capability, not just ventures. The goal of your first venture building effort should not only be a successful venture. It should be a documented, reusable process that makes the second venture easier and faster to build. Every experiment, every customer interview, every pivot should feed back into a playbook that the organization can use again.
Second, separate the engine from the core business. The ambidextrous organization model is clear on this point: exploration and exploitation require different structures, incentives, and cultures. Trying to run a venture building program inside an existing business unit is like trying to train for a marathon while running a sprint — the demands are fundamentally different.
Third, measure what matters. Move beyond vanity metrics to outcome-based measurement. Track how quickly you move from hypothesis to validated learning. Track capital efficiency across your portfolio. Track the rate at which ventures achieve product-market fit. These metrics will tell you whether your engine is improving over time.
The data is unambiguous. Companies that build ventures once and hope for the best are playing a losing game. Companies that build ventures repeatably — through dedicated structures like a corporate venture studio — are generating outsized returns with decreasing capital requirements. The question is not whether your organization should innovate. It is whether you are willing to invest in the engine that makes innovation systematic, measurable, and repeatable.