The Most Dangerous Thing Your Company Did in 2025? Nothing.
How America's largest companies are choosing financial engineering over innovation—and what it means for your competitive future
You miss 100% of the shots you don’t take. — Wayne Gretzky
Every business school in America teaches the same lesson: good management means avoiding unnecessary risk. Prudent CEOs protect shareholder value. Responsible boards demand rigorous ROI analysis before approving major initiatives. Conservative capital allocation wins the day.
There’s just one problem: this conventional wisdom is killing American competitiveness.
The greatest risk facing most US corporations isn’t taking a big bet and failing—it’s not taking enough calculated risks. While business schools preach prudence, the market is teaching a brutal counter-lesson: companies that avoid strategic risk don’t survive platform shifts.
To be clear, this isn’t a call for recklessness. As we argue in Big Bet Leadership, the goal is to decouple the size of your ambition from the size of your risk. We pursue ‘high ambition’ concepts not through massive, blind capital outlays, but through a rigorous, experimental process that caps the downside while working to expand the upside.
When executed correctly, the downside still has an upside. Even if the experiment ‘fails’ to yield the envisioned optimistic outcome, you have acquired proprietary data, upskilled your team, or validated a strategic dead end—assets that your ‘prudent’ competitors lack. You win either way: you either capture the future, or you get smarter for the next attempt. When boards demand guaranteed ROI before action, they reject this learning loop, protecting against a small, capped loss while leaving the door wide open to the uncapped, existential risk of obsolescence.
This paradox—where the “safest” management decision becomes the most dangerous strategic error—is exactly what Howard Schultz identified:
Howard Schultz, reflecting on corporate decline, put it perfectly: “They (successful large companies) lose the ability to be on offense. And the worst thing that a company can do, like a sports team, is start playing defense because you’re afraid to fail. That is a disease. Not unlike another disease which has happened in Starbucks, which is hubris.”
That disease—the fear of failure that drives defensive posturing—has infected corporate America at scale. And at precisely the moment when AI and digital transformation demand bold strategic action, most US companies are retreating into financial caution. While a handful of technology giants place trillion-dollar bets on the future, the vast majority of corporate America is choosing the certainty of stock buybacks over the uncertainty of transformation.
The numbers tell a stark story. In 2024, S&P 500 companies spent a record-breaking $942.5 billion on share repurchases—an 18.5% increase from the prior year.[1] Meanwhile, R&D spending growth has decelerated to its slowest pace since the 2009 financial crisis, limping along at just 2.9% globally and a mere 1.7% in the US when you exclude the distortive effects of the “Magnificent Seven” tech giants.[2]
This isn’t just about missed opportunities. It’s about a self-reinforcing cycle that’s hollowing out America’s industrial competitiveness while creating a dangerous concentration of innovation capability in fewer and fewer hands.
The Myth of the “Fast Follower”
To be clear, there is a rational defense for doing nothing. It’s called Strategic Patience.
Defenders of the status quo—often sitting in the CFO’s office or the corporate board—will argue that the “first mover advantage” is a myth. They will point to Apple, a company that rarely invents a category but perfects it after pioneers have exhausted their resources. They will argue that in a high-interest-rate environment, the smartest move is to let competitors burn cash discovering “product-market fit,” and then swoop in as a “fast follower” once the technology commoditizes. In this view, doing nothing isn’t passivity; it’s discipline.
Here is why that logic is dead wrong for 2025.
The “Fast Follower” playbook only works when you are competing on features or hardware. It fails catastrophically when you are competing on learning loops and organizational capabilities.
The platform shifts occurring today—specifically in AI and algorithmic operations—are not static products you can simply buy off the shelf later. They are compounding systems and organizational capabilities and culture. The company that started their “Big Bet” in 2023 isn’t just two years ahead in technology; they are two years ahead in data acquisition, model tuning, and organizational muscle memory.
By the time the “prudent” company decides the water is safe to enter, they won’t just be behind on features; they will be structurally incapable of catching up. The gap between the leader and the laggard isn’t linear anymore—it’s exponential. In this environment, “waiting to see what happens” ensures only one thing: you will be the last to know you’ve already lost.
The Risk-Aversion Cycle: How We Got Here
The cycle works like this: Companies attempt digital or AI transformations (or other types of big bets). These transformations fail at alarming rates—between 70% and 90% depending on which study you cite. This high failure rate creates institutional fear of “big bets.” That fear drives companies toward safer, short-term moves like buybacks and purely incremental projects. Buybacks drain capital that could fund genuine innovation. Without innovation investment, companies stagnate. Stagnation creates competitive vulnerability. Vulnerability eventually forces rushed, poorly-led transformation attempts. And those transformations fail at... 70-90%.
The cycle repeats.
Bain & Company’s 2024 research found that 88% of business transformations fail to achieve their original ambitions—only 12% meet or exceed their targets.[3] BCG consistently reports that 70% of digital transformation efforts fall short.[4] Most damning of all, Gartner found that 95% of corporate AI projects in 2025 failed to demonstrate any profit-and-loss impact, with most stalling in perpetual “pilot purgatory.”[5]
These aren’t new problems. These exact failure rates have persisted for over a decade, which tells us something critical: the problem isn’t the technology. It’s the leadership capability to execute transformational change.
The Magnificent Seven vs. The S&P 493
The bifurcation of American business has created two distinct economies operating under the same stock index.
The “Magnificent Seven”—Apple, Microsoft, Google, Amazon, Meta, Nvidia, and Tesla—operate with aggressive R&D intensity, long time horizons, and often founder-led cultures that can absorb short-term losses for long-term dominance. These companies captured the vast majority of market cap growth over the past five years, with US tech firms representing 18 of the top 24 global technology companies by valuation.[6]
Then there’s everyone else: the S&P 493. These companies face stagnant earnings growth, constrained R&D budgets, and quarterly earnings pressure that makes multi-year transformation bets nearly impossible. J.P. Morgan’s analysis shows this group being “throttled lower” by macroeconomic headwinds while the tech giants soar.[7]
Consider the capital allocation choices:
Apple, despite its innovation pedigree, spent $100 billion on buybacks in the 12 months ending September 2024[8]
General Motors cut R&D spending by 7.1% in 2024[9]
Stellantis reduced R&D by 8.4% during the same period[10]
The Financials sector hoarded cash instead of investing, citing regulatory uncertainty[11]
These aren’t companies being reckless. They’re being rational within a system that punishes long-term risk-taking and rewards short-term financial engineering. They’re playing defense because they’re afraid to fail—exactly the disease Schultz warned about.
Why Transformations Keep Failing
The transformation failure crisis has clear root causes, and none of them are technological:
1. The Digital Veneer Problem
Companies buy expensive software—Salesforce, SAP, Databricks, etc—without redesigning the underlying workflows. McKinsey’s research shows that organizations focusing on cultural change see 5.3x higher success rates than those focused solely on technology.[12] But culture change is hard, slow, and can’t be capitalized on a balance sheet. It also requires the executive team to change perspective— good luck with that!
When you digitize a broken 20-step approval process without simplifying it, you just get a faster way to do something that shouldn’t be done at all.
2. Talent Overload and Burnout
Bain identified a critical pattern: companies rely on too shallow a talent pool. They tap their “star players” for transformation initiatives on top of their day jobs. These leaders burn out. Initiatives lose focus. The research shows successful transformers dedicate at least 50% of key leaders’ time to change efforts rather than treating transformation as extracurricular.[13]
3. Integration Nightmares and Technical Debt
Salesforce research indicates that 95% of IT leaders report integration issues as a primary impediment to transformation.[14] Legacy systems create “technical debt” that suffocates new initiatives.
Volkswagen’s Cariad software unit provides a cautionary tale. Their attempt to build a unified operating system for all VW Group brands simultaneously resulted in two-year delays, spiraling costs, and the eventual firing of the entire leadership team. The failure stemmed from trying to integrate disparate legacy architectures from Audi, Porsche, and VW while simultaneously building new capabilities.[15]
The Cost of Inaction
Corporate history is littered with companies that optimized themselves into irrelevance:
Kodak invented the digital camera but suppressed it to protect film margins
Blockbuster dismissed Netflix to preserve late fee revenue
Nokia dominated mobile phones but missed the smartphone revolution
In the 2020-2025 period, we’re watching the modern equivalent unfold:
Boeing provides perhaps the most sobering example of how defensive management destroys value. After the 737 MAX disasters killed 346 people, investigations revealed a company that had spent decades prioritizing shareholder returns over engineering excellence. Between 2013 and 2019, Boeing spent $43 billion on stock buybacks—more than the entire cost to develop the 737 MAX program.
According to a 2020 report by the House Committee on Transportation and Infrastructure, Boeing’s culture had shifted from “safety first” to “profits over safety,” with engineers pressured to cut costs and rush timelines to compete with Airbus. The company’s risk aversion paradoxically created catastrophic risk: afraid to make the expensive bet on an all-new aircraft design, Boeing tried to incrementally upgrade a 1960s airframe, leading to fatal design flaws, a 20-month grounding, over $20 billion in losses, and permanent damage to its reputation as the gold standard of aerospace engineering.[16]
Traditional banks move cautiously on AI due to regulatory fears about “black box” algorithms, while fintech disruptors like Klarna adopt an “AI-first” strategy. Klarna replaced 700 customer service contractors with AI agents handling two-thirds of inquiries, driving a $40 million profit improvement. This allowed them to offer more competitive rates, contributing to 43% growth in US Gross Merchandise Volume in Q3 2025—significantly outpacing traditional credit card growth.[17]
The lesson is clear: in a platform shift, speed is a risk mitigant. Moving slowly doesn’t reduce risk—it creates internal tension to create “predictable” outcomes versus experiments, defined ROI vs. calculated risks, incrementalism instead of reinvention.
Anatomy of a Successful Big Bet
Not all companies are paralyzed. The most successful transformations of the past five years share common DNA:
Nvidia: The Zero-Billion Dollar Market
Jensen Huang’s bet on CUDA—a software platform allowing GPUs to be used for general-purpose computing—is perhaps the most successful “big bet” in business history. In the mid-2000s, Huang invested billions into a market that didn’t exist. Wall Street punished the stock. The investment depressed margins for over a decade.
But Huang was betting on first principles: that computing demand would outstrip CPU capacity and that parallel processing would become essential. When the AI boom materialized with AlexNet in 2012, Transformers in 2017, and ChatGPT in 2022, Nvidia was the only company with both the hardware and software stack ready. By 2024, they controlled 80-95% of the AI chip market.[18]
The lesson: Big bets require horizon planning—betting on a future 10 years out—and the stomach to ignore short-term critics. This is offense, not defense.
Microsoft: The Structured Bet
Rather than building large language models entirely in-house (which might have failed due to bureaucracy), Satya Nadella invested $13 billion in a capped-profit partnership with OpenAI. This structure gave Microsoft exclusive access to models for Azure and Copilot, having a major equity upside potential, while offloading the “lumpy” training costs to OpenAI.[19]
Even in a downside scenario where OpenAI falters, Microsoft retains IP rights and customer relationships. It’s a masterclass in risk transfer—a “hedged big bet” that balances upside potential with downside protection.
Why Leadership Incentives Drive the Wrong Behavior
The structure of executive compensation is a major driver of risk aversion:
Value-based equity grants are increasingly common—executives receive a fixed dollar amount of stock rather than a fixed number of shares. A 2025 study found this structure actively discourages innovation because if the stock price rises, executives get fewer shares, capping their upside. This creates a perverse incentive: executives prefer stock price stability over explosive growth, leading to reduced R&D investment.[20]
EPS-linked bonuses directly incentivize buybacks over capital expenditures. If a CEO’s bonus depends on hitting a $5.00 earnings-per-share target, buying back 5% of shares is a far safer path than building a new factory that won’t be profitable for three years.
CEO tenure dynamics create a mismatch between leadership timelines and transformation payoff periods. The average CEO tenure is 5-7 years. Most transformations require 7-10 years to fully realize value. Research shows “rookie CEOs” in their first 1-3 years take more risks to establish credibility, but become increasingly conservative as tenure lengthens and they accumulate unvested equity to protect.[21]
The result? Professional managers optimizing for personal wealth preservation rather than company transformation. They’re playing defense—protecting what they have rather than pursuing what they could build.
Breaking the Cycle: The Big Bet Leadership Playbook
So how do you break the risk-aversion cycle? How do you shift from defense to offense without being reckless?
The answer isn’t to make reckless bets or to avoid buybacks entirely. It’s to develop a portfolio approach to strategic risk—and more importantly, to build the leadership capability to execute high-ambition transformations successfully.
The framework that’s emerging from successful transformers follows a 70/20/10 portfolio model:
70% core optimization: Low-risk improvements to existing operations that fund the portfolio
20% strategic plays: Medium-risk adjacent market moves that extend current advantages
10% transformational moonshots: High-risk, high-reward bets on future platforms[25]
[note: this is a typical allocation, not a recommendation. What is a recommendation is to have a clear-eyed portfolio model]
But here’s what the failure statistics tell us: having a portfolio isn’t enough. You need the executive capability to lead these bets to successful outcomes.
This is where many companies fail. They approve the budget. They launch the initiative. But they lack the discipline to:
Separate reversible from irreversible decisions (Amazon’s “one-way vs. two-way doors”)[26]
Design transformation-specific governance that protects long-term initiatives from quarterly budget cuts
Build psychological safety where “smart failure” generates learning rather than punishment
Dedicate genuine leadership capacity rather than overloading star players
Drive cultural change alongside technology deployment
Maintain conviction through the inevitable J-curve productivity dip
These capabilities aren’t intuitive. They’re learned. And they’re teachable.
What This Means for Your Organization
If you’re a C-suite executive or board member, here are the hard questions you should be asking:
What’s our actual risk posture? Calculate what I call the Corporate Vitality Ratio: R&D + Capex vs. buybacks + dividends. If shareholder returns significantly exceed growth investments or the ratio is trending that direction, you’re in defensive mode.
What’s our transformation track record? Be brutally honest. How many of your last five major initiatives actually delivered their promised value? If the answer is less than 30%, you have a leadership capability gap, not a strategy problem.
Who owns long-term risk? In most organizations, no single executive has a mandate to protect 10-year bets from short-term pressures. Without this protection, transformation budgets become the first casualty of any earnings miss.
How do we size bets? Are you applying the same approval rigor to a $10 million AI pilot as you do to a $1 billion factory? Type 2 decisions (reversible, low-consequence) need fast, distributed authority. Type 1 decisions (irreversible, existential) need heavyweight deliberation.
What’s our AI posture? The NACD recommends boards adopt an explicit “AI Posture” statement defining risk appetite, then review it annually.[27] Most boards haven’t done this, creating a vacuum where tactical pilots proliferate without strategic coherence. But work to make this posture helpful and actionable, not “empty calories".
Are we playing offense or defense? To use Schultz’s framework: are you making decisions because you’re excited about what you could build, or because you’re afraid of what you might lose?
The Playbook You Need
The difference between the 12% of transformations that succeed and the 88% that fail isn’t luck. It’s not even strategy. It’s execution discipline in the face of uncertainty.
You need a systematic approach to:
Making the big bet decision correctly
Structuring transformation governance to survive organizational antibodies
Building the coalition of support that sustains multi-year change
Navigating the inevitable setbacks without losing conviction
Scaling pilots into production systems
Measuring progress when traditional ROI metrics don’t apply
This is exactly what Big Bet Leadership: Your Transformation Playbook for Winning in the Hyper-Digital Era provides — a battle-tested system for leading high-stakes transformations successfully, drawn from the successes and failures of companies ranging from Amazon, T-Mobile to the small & mid-sized companies.
You can’t avoid making big bets anymore. In an AI-driven era, the only choice is between making smart, well-led calculated risks—or making desperate, poorly-executed ones later when you’ve run out of time.
The risk-aversion cycle can be broken. But it requires playing offense backed by the executive competence to execute transformations that actually work.
Which will you choose?
Onward,
John
PS — I’d be grateful if you shared this post with your network both in substack and on LinkedIn.
John Rossman is a former Amazon executive who launched and scaled Amazon Marketplace from startup to a platform generating hundreds of billions in annual sales. He’s the author of four books on digital transformation and strategic leadership, including Big Bet Leadership, works with companies as a strategic advisor on high-stakes transformation initiatives, and delivers invigorating keynotes to challenge conventional thinking on leadership, innovation and operations.
References
[1] S&P Global. “S&P 500 Q4 2024 Buybacks Increase 7.4% and 2024 Expenditure Sets New Record by Increasing 18.5%.” March 2025. https://www.spglobal.com/spdji/en/documents/index-news-and-announcements/20250319-sp-500-buybacks.pdf
[2] World Intellectual Property Organization. “Global Innovation Index 2025 - Global Innovation Tracker.” 2025. https://www.wipo.int/web-publications/global-innovation-index-2025/en/global-innovation-tracker.html
[3] Bain & Company. “88% of business transformations fail to achieve their original ambitions; those that succeed avoid overloading top talent.” 2024. https://www.bain.com/about/media-center/press-releases/2024/88-of-business-transformations-fail-to-achieve-their-original-ambitions-those-that-succeed-avoid-overloading-top-talent/
[4] Boston Consulting Group. “Why 70% of Digital Transformations Fail.” 2024. Multiple reports 2021-2024.
[5] ComplexDiscovery. “Why 95% of Corporate AI Projects Fail: Lessons from MIT’s 2025 Study.” 2025. https://complexdiscovery.com/why-95-of-corporate-ai-projects-fail-lessons-from-mits-2025-study/
[6] EY. “US corporations dominate global stock markets – Europe and Asia lag far behind.” January 2025. https://www.ey.com/en_ch/newsroom/2025/01/us-corporations-dominate-global-stock-markets-europe-and-asia-lag-far-behind
[7] J.P. Morgan Asset Management. “Magnificent 7, sure… but increase exposure to S&P 493 too.” Investment Executive, 2024. https://www.investmentexecutive.com/soundbites/equities/magnificent-7-sure-but-increase-exposure-to-sp-493-too/
[8] S&P Global. “S&P 500 Q3 2024 Buybacks Decrease 4.0% from Q2 2024.” October 2024. https://www.prnewswire.com/news-releases/sp-500-q3-2024-buybacks-decrease-4-0-from-q2-2024--as-12-month-expenditure-increases-4-7-from-previous-year-302331086.html
[9] World Intellectual Property Organization. “Global Innovation Index 2025.” 2025.
[10] Ibid.
[11] S&P Global. “S&P 500 Q4 2024 Buybacks.” March 2025.
[12] Mavim Blog. “Why 70% of Digital Transformations Fail: Insights and Solutions.” 2025. https://blog.mavim.com/why-70-of-digital-transformations-fail-insights-and-solutions
[13] Bain & Company. “88% of business transformations fail.” 2024.
[14] Integrate.io. “Data Transformation Challenge Statistics — 50 Statistics Every Technology Leader Should Know in 2025.” 2025. https://www.integrate.io/blog/data-transformation-challenge-statistics/
[15] DigitalDefynd. “15 Digital Transformation Failure Examples [2025].” 2025. https://digitaldefynd.com/IQ/digital-transformation-failure-examples/
[16] U.S. House Committee on Transportation and Infrastructure. Final Committee Report: The Design, Development & Certification of the Boeing 737 MAX. September 2020. https://transportation.house.gov/download/final-737-max-report
[17] Investing News Network. “Klarna Delivers Record-breaking Q3 as AI-Powered Digital Bank: $903 Million in Revenue and 4 Million Card Sign-ups in 4 Months.” 2025. https://investingnews.com/klarna-delivers-record-breaking-q3-as-ai-powered-digital-bank-903-million-in-revenue-and-4-million-card-sign-ups-in-4-months/
[18] Entrepreneur. “How Nvidia CEO Jensen Huang Transformed a Graphics Card Company Into an AI Giant.” 2025. https://www.entrepreneur.com/business-news/how-nvidia-pivoted-from-graphics-card-maker-to-ai-chip-giant/477530
[19] Official Microsoft Blog. “The next chapter of the Microsoft–OpenAI partnership.” October 2025. https://blogs.microsoft.com/blog/2025/10/28/the-next-chapter-of-the-microsoft-openai-partnership/
[20] Virginia Tech News. “A common CEO pay strategy is stalling innovation, a new study reveals why.” April 2025. https://news.vt.edu/articles/2025/04/pamplin-common-ceo-strategy-stalling-innovation.html
[21] EconStor. “The impact of rookie CEOs on corporate risk-taking.” 2025. https://ideas.repec.org/a/eee/finana/v107y2025ics105752192500746x.html
[22] McKinsey. “Seizing the agentic AI advantage.” 2025. https://www.mckinsey.com/capabilities/quantumblack/our-insights/seizing-the-agentic-ai-advantage
[23] MLQ.ai. “The GenAI Divide: State of AI in Business 2025 Report.” 2025. https://mlq.ai/media/quarterly_decks/v0.1_State_of_AI_in_Business_2025_Report.pdf
[24] Allvue Systems. “Venture Capital Trends 2025: Outlook & Insights.” 2025. https://www.allvuesystems.com/resources/top-trends-in-venture-capital/
[25] International Institute for Analytics. “Strategic Bets and Portfolio Thinking for AI Use Cases.” 2025. https://iianalytics.com/community/blog/strategic-bets-and-portfolio-thinking-for-ai-use-cases
[26] Startup Archive. “Jeff Bezos explains the difference between one-way door decisions and two-way door decisions.” https://www.startuparchive.org/p/jeff-bezos-explains-the-difference-between-one-way-door-decisions-and-two-way-door-decisions
[27] National Association of Corporate Directors. “Artificial Intelligence Governance.” 2024-2025. https://www.nacdonline.org/all-governance/governance-resources/trending-oversight-topics/artificial-intelligence/