Every executive desires game-breaking notions; scant few make them into lasting results. Seldom it is a question of imagination, but rather the discipline to find a balance between upside and downside and to move fast without breaking the business. It’s worth nothing, though, that novelty is only superficially seductive: research from MIT Sloan Management Review and BCG finds that a mere one in 10 companies has achieved substantial economic benefits through AI to date, a reminder that more and newer isn’t necessarily better—when novelty is disconnected from execution, it becomes an expensive hobby.
Five diverse industry leaders including retail, elite motorsport, professional services, higher education and talent solutions share how they bridge the gap between innovation and real-world business results, all while managing their exposure to risk. Their guidance coalesces around a single principle: Put up strong guardrails, then give teams permission to run.
Anchor growth first, then try guardrails
David Walmsley, the chief digital and technology officer at Pandora, prefers investments that are top-line accretive. Back-end optimization is important, he says, but he prods leaders to pair execution with ambition — reflecting company values that prioritize both delivery and dreaming.
His modus operandi here is pragmatic field testing: prove by doing, but not recklessly. That involves time-boxed pilots, staged funding, and pre-agreed “kill criteria” so sunk cost fallacy doesn’t kick in; methods like innovation accounting, as popularized (if not originated) by lean startup practitioners, enable teams to track learning velocity, not just vanity metrics.
It’s a portfolio game. A good mix — incremental improvements plus horizon-two bets and a small number of transformative plays — keeps risk diversified. Many executives have used the 70/20/10 rule of thumb and adjusted up or down based on market conditions.
Don’t treat technology as decoration; it’s a performance lever
In Formula 1, reliability is gold and milliseconds determine podiums. Steve Riley, who heads IT operations and service management for the Mercedes-AMG PETRONAS F1 Team, thinks of innovation in a different way – it is a means to an end of enhanced performance rather than the novelty itself. When the car has to run, “shiny” is a liability; when the opportunity is there, incremental tech gains compound into competitive advantage.
That lens moves IT from back office to factory floor. Linked autoclaves, remote guidance and digitally choreographed manufacturing amplify the blast radius of any change. To mitigate risk, teams rely on the practices of site reliability engineering – clear service-level objectives, “shadow mode” rollouts, blue–green deployments – and they experiment with digital twins to simulate conditions before touching production.
The lesson for all sectors: That reliability and reinvention are complements. The more stable your core services, the more latitude you earn to take a measured swing elsewhere.
Create a safe-to-operate core to unleash bold bets
Kirsty Roth, the chief operations and technology officer at Thomson Reuters, warns that lasting innovation sits on the coattails of mature risk management. Security, privacy and A.I. governance overhauls gave them the confidence to accelerate. Frameworks such as ISO 27001, NIST’s A.I. Risk Management Framework, and good data stewardship hone the organization’s “risk appetite” and elucidate where to push.
She also highlights some very explicit parameters: what needs to be zero-defect, what can be sandboxed, and what garners expedited approval with compensating controls. Clarity minimizes the cultural drag of fear. When people understand the rules, the over-analyze stops, and the shipping begins.
Program metrics imitate this approach. The Project Management Institute, for example, has for years recorded that significant capital investment is left on the table through bad execution; constraining governance and also preserving pace immediately lifts the yield of innovation.
Reframe risk as learning, not loss
The London Interdisciplinary School’s CEO and co-founder Ed Fidoe helps leaders see innovation as an exercise in collecting evidence. Based on systems thinking, he challenges teams to predict second-order and higher order effects across customers, suppliers, regulators, and culture—and then test those assumptions quickly and cheaply.
Biology provides a good metaphor: Some species spread risk widely, others concentrate resources from reproduction. Companies need to do both — run as many low-cost probes as possible to map the landscape, while reserving focused investment for validated directions. Distinctions such as pre-mortems, red teaming, and real options analysis hardwire learning into decision making.
The aim is not to remove uncertainty but to digest it. Each iteration purchases information that allows the next bet to be more intelligent.
Oversee the portfolio—and institutionalize the kill
“Risk is part and parcel of being in business,” says Bev White, the CEO of Nash Squared. Her firm’s Digital Leadership Report discovered that almost a fifth of AI investments are now mainstream or large. That scale increases the stakes—especially given that relatively few generative AI pilots make it to production—so governance needs to make sure ambition doesn’t also make for recklessness.
Impactful innovators clearly set stage gates, monitor unit economics early, and normalize kill rates of lagging bets. Blameless postmortems turn failure into organizational memory. Ring-fenced budgets and cross-functional squads will, at the same time, protect promising projects from quarterly whiplash.
External benchmarks help. IBM Says Most Firms Are Now “Voicing an Inclination to Adopt” AI, but Challenges Loom Large IBM’s Global AI Adoption Index supports executive surveys that an impressive load of enterprises are using or investigating AI, but the problems of skills, data quality, and security still top the list of barriers. It turns out that leaders who invest in capability-building, in product management, data governance and AI ethics, are better at getting the return on risk.
Turning ideas into action
Establish your risk appetite and guardrails upfront. Tranche funding based on learning milestones, not just timelines. Reliable at the core, that is, to free space for daring gambles.” And measure what matters — the impact on customers, the speed to insight and the cost to learn, not just outputs shipped.
It pays to operationalize innovation. As these leaders demonstrate, the winning formula is courage with guardrails: Dream big, test relentlessly and learn fast or risk being disrupted by competitors who do, say the authors.