Detailed content
🎯 Why Measure Innovation — and the Dangers of Measuring It Wrong
"What gets measured gets managed" is one of the most misapplied phrases in innovation. When applied without discernment, it leads organizations to measure what's easy to measure—and optimize the metric instead of the actual outcome. Innovation becomes theater: lots of Post-its, few products. Lots of approved projects, little new revenue.
💎 The Measurement Paradox
Innovation needs to be measured to be managed — but the wrong measurement destroys it before it produces value. The Goodhart’s law says: “When a measure becomes a target, it ceases to be a good measure.” In the context of innovation, this means teams will start maximizing the metric—not the outcome.
- • Goal: 10 ideas per month → team generates 10 bad ideas just to hit the target
- • Goal: 5 experiments run → teams run trivial experiments without a real hypothesis
- • Goal: 300% ROI on innovation → only low-risk (and low-innovation) projects are approved
📊 The Problem in Practice
- 94% of executives are dissatisfied with their company’s innovation performance (McKinsey, 2023)—but most continue to measure the number of projects and the budget invested
- 72% of companies cannot connect innovation initiatives to measurable business results (Deloitte Insights, 2022)
- Organizations that measure innovation results (not just activities) are 2.4x more likely to deliver above-industry-average growth (BCG, 2024)
✓ Measure to Learn
- ✓Metrics tied to value hypotheses, not activities
- ✓Metrics that change behavior when they change
- ✓Metrics portfolio (input + output + outcome)
- ✓Tolerance for negative metrics (learning)
✗ Measuring to Show Off
- ✗Counting the number of ideas generated as a proxy for innovation
- ✗Measure budget invested instead of value generated
- ✗A pretty dashboard for the board with no operational use
- ✗Metrics that never show poor results
💡 Metrics test
For each metric you use, ask: "If this number were cut in half, what would we do differently?" If the answer is "nothing," the metric is decorative. If someone at the company has already optimized the number without improving the outcome, the metric has become theater.
⚖️ Input vs. Output vs. Outcome Metrics
The most common mistake in measuring innovation is confusing the three levels: what we put into the process (input), what the process produces (output), and the real value the output generates (outcome). Organizations that measure only inputs believe they're innovating because they invested. Those that measure only outputs believe they're innovating because they produced something. Only those that measure outcomes know whether they truly innovated.
🔧 Input Metrics — The Fuel
Measure allocated resources: R&D budget as a % of revenue, headcount dedicated to innovation, number of partnerships with startups and universities, training hours in innovation methodologies. They’re necessary to ensure investment, but guarantee nothing about results. A company can spend millions on R&D without innovating; another can innovate with minimal resources.
📦 Output Metrics — What the Process Delivered
Measure what came out of the innovation process: number of experiments run, MVPs launched, patents filed, features delivered, idea-to-project conversion rate. They’re closer to the outcome, but can still be misleading. A product launched that nobody uses is output without outcome. A hundred experiments without a clear hypothesis are activity without learning.
🎯 Outcome Metrics — Value Generated
Measure real impact on the business and the customer: revenue generated by products launched in the last 3 years (revenue from new products), cost reduction through process innovation, NPS change attributable to improvements, market share gained in new segments. They’re the hardest to measure (they require causal attribution), but they’re the only ones that validate whether innovation really happened.
🔽 The Innovation Funnel and Conversion Rates
The innovation funnel maps how ideas become products—and how many die at each stage. This isn’t pessimism; it’s realism. A 1% conversion rate at the final stage can be healthy if the earlier stages filtered well. The problem is when you don’t know the rates—or when no idea makes it to the end.
📊 Funnel Benchmarks
- Clayton Christensen's rule of thumb: for every 3,000 raw ideas, reach ~125 formal evaluation, ~9 development, ~4 customer testing and only 1.7 launches with commercial success
- Companies that systematically measure the funnel launch products with 30% more frequently and abandon weak initiatives 40% earlier (Stage-Gate International, 2023)
- The most common bottleneck is the transition from idea → prototype: approval bureaucracy kills more innovations than a lack of ideas
Funnel Stages and What to Measure at Each One
Idea Generation
Rate: 100% → ~4%Measure: volume of ideas submitted, diversity of sources (departments, hierarchy), ideas per employee/month
Screening and Selection
Rate: ~4% → ~0.3%Measure: average evaluation time, documented selection criteria, % rejected with feedback, % of ideas from outside the core business
Prototyping and Experimentation
Rate: ~0.3% → ~0.13%Measure: idea-to-prototype cycle time, average experiment cost, % of hypotheses validated vs. refuted, iteration speed
Development and Launch
Rate: ~0.13% → ~0.06%Measure: time-to-market, post-launch success rate, new product NPS, revenue from new products in 12/24 months
🛠️ Bottleneck Diagnosis
Map where ideas die. If most fall at screening, the problem is selection criteria or lack of sponsorship. If they fall at prototyping, the problem is resources or an experimentation culture. If they reach launch but don’t generate revenue, the problem is insufficient market validation. The funnel is a diagnostic tool, not just a funnel.
🧭 Culture Metrics: Psychological Safety, % of Ideas Implemented, and ESAT
The most critical innovation metrics aren’t on the financial dashboard—they’re in the culture. Without a supportive culture, no methodology works. But culture isn’t abstract: it can and should be measured. Three dimensions are especially revealing: the team’s psychological safety, the actual rate of idea implementation, and employee satisfaction with the innovation environment (ESAT).
🛡️ How to Measure Psychological Safety
Amy Edmondson (Harvard) developed the standard psychological safety scale—7 questions answered on a Likert scale by each team member. The team’s average score is the indicator. Sample questions:
- •"If I make a mistake on this team, it won't be used against me"
- •"Members of this team are able to bring up problems and tough issues"
- •"On this team, it's safe to take risks"
- •"My skills are valued and put to use on this team"
Score < 3.5 on a 5-point scale: danger zone — the team won’t experiment. Score > 4.2: healthy zone for innovation.
Psychological Safety
Measured using Edmondson’s survey. Apply it by team (not aggregated — the company average masks toxic teams). Track quarterly. Correlate it with the team’s experiment rate: the correlation is usually strong and reveals where culture blocks innovation.
% of Ideas Implemented
Of the ideas submitted to the innovation program, what percentage make it to implementation (not just approval)? A rate of 0.5%–2% is healthy in large companies. A zero rate means the idea program is theater. A rate of 30%+ may indicate that screening is too loose — bad projects are being implemented.
ESAT (Employee Satisfaction)
An NPS version for employees, focused on innovation: "On a scale of 0 to 10, how much do you feel your company supports and enables innovation?" Break it down by hierarchy level — frontline employees often see reality differently from the C-level. A large gap = a cultural implementation problem, not a messaging problem.
⚠️ Trap: Aggregated Culture Metrics
Measuring psychological safety or ESAT as a company-wide average hides the problem. A company where 20% of teams are highly innovative and 80% are dysfunctional can have an "acceptable" average score. Always break it down by team, department, and hierarchical level. Variance matters as much as the average.
📈 Innovation Accounting (Lean) and Actionable Metrics
Eric Ries, in “The Lean Startup,” introduced the concept of innovation accounting: a measurement system tailored to the high-uncertainty context of startups and innovative projects, where traditional financial metrics do not apply because there is no track record, no established revenue, and the business model is still being discovered.
⚙️ The Three Steps of Innovation Accounting
Innovation accounting works in three progressive phases, each with its own metrics that reflect the stage of learning:
✓ Actionable Metrics
- ✓Linked to a specific action that can be taken
- ✓Change means something different will be done
- ✓Ex.: D1 activation rate (suggests changing onboarding)
- ✓Ex.: % of users who complete a critical step (reveals friction)
- ✓Ex.: NPS segmented by user cohort
✗ Metrics for Reporting Only
- ✗Good for reporting, not for decision-making
- ✗Change doesn’t indicate what should change in the product
- ✗Ex.: total number of registered users
- ✗Ex.: monthly aggregate pageviews
- ✗Ex.: "overall engagement" without a precise definition
🛠️ Actionability test
For each metric, complete this sentence: "If [metric] drops from X to Y, we will [specific action]." If you can't complete the sentence with a real, concrete action, the metric is for reporting, not for taking action. A good actionable metric is one that, when it changes, tells the whole team what to do next.
🎭 Vanity Metrics and Number Theater — What to Avoid
Vanity metrics — a term popularized by Eric Ries — are metrics that make you look good to the board without revealing anything about the business's real health or the innovation's real progress. They are numbers that always go up, never indicate a problem, and never lead to decisions. Number theater is when the entire organization knows the metrics are empty but keeps reporting them because that's what the C-level wants to see.
🔍 Anatomy of a Vanity Metric
A vanity metric has three identifiable characteristics: (1) it always grows or rarely falls, regardless of what the company does well or poorly; (2) it has no denominator — and is an absolute number without context (1,000 downloads vs. 1,000 downloads / 10,000 visitors = 10% conversion — this second figure is actionable); (3) it doesn’t distinguish cohorts — treats all users as if they were equal.
- •“We have 100,000 registered users” — without knowing how many actively use the platform
- •"We generated 500 ideas at the hackathon" — without knowing how many were tested or implemented
- •"We invested R$ 5M in innovation" — without correlating it with revenue or learning generated
- •“We trained 2,000 employees in design thinking” — without measuring behavior change
How the Numbers Theater Takes Hold
Pressure for visible results
Leadership wants to see “innovation happening.” The innovation team selects metrics that grow easily to show progress.
Metrics become targets
What was an indicator becomes a goal. Teams start optimizing the number instead of the outcome. Goodhart’s Law in action.
Silent skepticism
Everyone knows the numbers are empty, but no one speaks up. The culture of psychological safety deteriorates. Real innovators leave—or stop putting themselves out there.
Disruption (or stagnation)
Market crises expose the emptiness of the metrics. Or the company stagnates while competitors who measured the right things move forward.
✗ Common Vanity Metrics
- ✗Number of registered users (without DAU/MAU)
- ✗Total ideas submitted (without an implementation rate)
- ✗Innovation budget in absolute $
- ✗Number of training sessions held
- ✗Downloads / pageviews without conversion
✓ Actionable Alternatives
- ✓DAU/MAU ratio (real engagement)
- ✓% of ideas that reach prototype and launch
- ✓Innovation ROI (new revenue / invested budget)
- ✓Post-training behavior change (90-day survey)
- ✓Retention rate and LTV of new products
💡 Golden Rule
Before adopting any innovation metric, ask: “Could this metric show that we’re doing poorly?” If the answer is no, it’s a vanity metric. Honest metrics show both the positive and the negative. And only honest metrics build a real culture of innovation.
📋 Module Summary
Next Module:
2.1 - Design Thinking — the next learning path begins with the methodology that puts the user at the center of the innovation process