The KPI Illusion: Why Innovation Needs Gut Feel Too
At a recent roundtable of corporate innovation leaders, the usual themes dominated: how to measure success, prove ROI, and bring accountability to initiatives that often feel like black boxes. There was no shortage of talk about dashboards, KPIs, and innovation funnels.
A lot of details were evaluated — is it better to look at revenue potential and volume, margin improvement potential, customer feedback, … and which proxies do we use?
But then someone said something that made the whole room pause:
“I understand the need for measurement — but what about expert judgment?”
That moment cut through the noise because it named the deeper tension many innovation teams feel but rarely articulate. In an era where innovation is under scrutiny and budgets are tight, organizations are doubling down on metrics. It feels safer. Rational. Controlled.
But here’s the problem: in our attempt to professionalize innovation, we may be sterilizing it.
The Corporate Pendulum: From Hype to Fear
A decade ago, innovation was in vogue. Everyone had labs, accelerators, and digital moonshots. It was a time of experimentation and ambition — but also of waste. Too many big bets went nowhere. Too many sleek prototypes led to nothing. In hindsight, it wasn’t always the ideas that failed — it was the system: disconnected from the business, unchecked in its spending, and unaccountable in results.
The backlash was inevitable. Boards asked for returns. CFOs demanded structure. The pendulum swung hard the other way: into the arms of KPIs, performance funnels, and quantified innovation scorecards. And while accountability is a good thing, the fear that fueled it has created new problems. Metrics have become a comfort blanket. We believe that if we can measure it, we can manage it. But in innovation, that’s often a dangerous illusion.
The problem isn’t that KPIs exist. It’s when they dominate too early — when judgment is deferred, and all conviction must be justified through numbers that don’t yet make sense. The result? Safe bets, incremental ideas, and a pipeline full of low-variance concepts that look good on paper — but won’t shift the company’s trajectory.
The Horizon Disconnect: Where Metrics Break Down
One reason KPIs are over-applied is a failure to differentiate types of innovation. The classic three-horizon model helps make this clear:
- Horizon 1 focuses on optimizing existing business models. Here, metrics shine. We have history, benchmarks, performance data. It’s about improving the known.
- Horizon 2 explores adjacencies — new customer segments, channels, or product extensions. Metrics still help, but assumptions must be more flexible.
- Horizon 3 is about future bets. Radical ideas. New ventures. In this space, metrics break down. You’re in unknown territory, where the right indicators are either unavailable or irrelevant.
The danger is that organizations apply Horizon 1 metrics — like revenue projections, CAC, or NPS — on Horizon 3 ideas. That’s like judging a seedling on whether it bears fruit. You end up killing the things that need time, space, and nurturing to grow.
It’s not that Horizon 3 should be a black box. But we need a different approach: one that balances qualitative insight, structured exploration, and evolving signals — not just rigid KPIs.
Future-Back Thinking and the Role of Expert Judgment
The best innovators don’t just extrapolate from today — they reason backward from tomorrow. They ask: what will matter in 2030 or 2035? What customer behaviors, technological shifts, or societal changes are already underway — and what does that mean for us now?
This “future-back” mindset isn’t fantasy. It’s strategic foresight. And it requires judgment — especially before data can validate anything.
Expert judgment gets a bad rap in data-driven cultures. But judgment doesn’t mean wild guessing. It means:
- Pattern recognition across markets and cycles
- Sensitivity to weak signals and early shifts
- The ability to form and hold a strategic point of view
Great founders, venture investors, and visionary leaders all operate from this space. They back ideas before the metrics exist — because they’ve seen this pattern before. They don’t abandon structure — but they know when to override it. And importantly, they hold their beliefs lightly: strong views, weakly held.
Corporate innovators must do the same. They must create space for conviction-based calls — but with built-in mechanisms to test and evolve quickly. That means:
- Making early-stage bets without performance guarantees
- Using lean experiments and qualitative signals to test hypotheses
- Pivoting or killing ideas fast if assumptions don’t hold
The goal is not to eliminate judgment. It’s to make it safe to use — then stress-tested in smart ways.
Building Innovation Systems That Respect the Unknown
So what does a smarter innovation governance model look like? It starts with sequencing — aligning the right tools with the right phase.
Phase 1: Explore with Judgment
Here, the goal is not performance — it’s insight. Use expert panels, domain leads, and future-back scenarios to identify meaningful problem spaces. Don’t ask for KPIs. Ask for clarity of thinking. Is there a real pain point? A trend with momentum? A founder or champion with deep conviction?
Phase 2: Validate with Light Signals
Now, begin testing assumptions. But stay lightweight. Use prototypes, fake doors, customer interviews, and limited pilots. Gather directional data — not definitive answers. The purpose is learning, not proving.
Phase 3: Scale with KPIs
Only once core assumptions are de-risked should you bring in hard metrics. Now the idea must earn its keep. At this stage, CAC, churn, revenue, and NPS are essential. You’re no longer validating — you’re optimizing.
Most corporate systems collapse because they apply Phase 3 governance in Phase 1. It’s no wonder innovation dies early.
We need to separate the governance of exploration from the governance of scale. And we need to train leaders not just in evaluating business cases — but in evaluating uncertainty, narrative coherence, and potential.
Final Word: Don’t Let the Spreadsheet Dictate the Strategy
Metrics matter. But they don’t lead. In innovation, they follow. They validate the bets that judgment helped you place.
If your process has no room for vision, conviction, or future-back thinking, you’re not leading innovation — you’re managing risk. And someone else, somewhere, is placing the bet you were too careful to make.
So yes, measure what matters. But don’t let the obsession with measurability blind you to what could be transformational. Use judgment. Learn fast. And remember: some of the best calls are made before the numbers say they’re safe.
Looking to work with a team of experts that know how navigate both the quantification and make expert judgment calls? Why not reach out: hello@minglabs.com
Writer:
Sebastian Mueller — Founding Partner, MING Labs
