How to Measure Innovation

Quick Answer: To measure innovation, we need a system that goes beyond traditional financial metrics. Innovation accounting provides the framework: we track leading indicators at the project level, aggregate them into portfolio-level metrics, and use techniques like Monte Carlo simulation to forecast outcomes. The key is measuring progress toward validated learning, not just revenue: how fast are we validating assumptions, how quickly are we running experiments, and are our pivot-or-persevere decisions based on evidence?

Why Measuring Innovation Matters

The CEO asks the VP of Innovation, “What’s the ROI on that innovation lab we funded last year?” The VP pulls up a slide with the number of ideas generated, workshops held, and prototypes built. The CEO nods politely. Everyone in the room knows those numbers don’t answer the question. We’ve all been in that meeting.

The metrics we use to run established businesses (revenue, profit margin, market share) are lagging indicators. They tell us what already happened. For innovation projects, where we are deliberately exploring the unknown, those numbers are zero for months or even years. If we wait for revenue to prove an innovation project’s value, we’ve already missed the window to make good investment decisions along the way.

And yet, if we can’t measure innovation, we can’t manage it. We can’t make rational funding decisions. We can’t tell the board why the innovation portfolio deserves continued investment. We can’t compare one uncertain bet against another. Without measurement, innovation devolves into gut feel, politics, and whoever has the loudest voice in the room.

We can measure innovation. Not with a single magic number, but with a system. This guide walks through the five core disciplines that make up that system, with links to deeper articles on each. Let’s start with the foundation.

Innovation Accounting: The Foundation of Measurement

Innovation accounting is the discipline of quantifying progress on uncertain projects when traditional financial metrics are still zero. It was first popularized by Eric Ries in The Lean Startup, but the definition Eric gave was deliberately vague, and practitioners have been filling in the gaps ever since.

Innovation accounting answers one question: Is this project making real progress, or are we just burning cash?

To answer that, we need to shift what we consider “progress.” In a core business, progress means more revenue, more customers, more margin. In innovation, progress means reducing uncertainty. Every experiment we run, every customer interview we conduct, every assumption we validate (or invalidate) is a measurable step forward. The trick is quantifying those steps in a way that allows rational investment decisions.

A complete innovation accounting system operates at three levels:

1. Project level — Are individual projects validating their riskiest assumptions? Are teams running experiments efficiently? Are pivot-or-persevere decisions being made with evidence?
2. Portfolio level — Is the mix of projects across innovation horizons balanced? Are we investing appropriately across explore-and-exploit activities?
3. Capability level — Is the organization getting better at innovating over time? Are we building the skills, processes, and culture that make innovation repeatable?

Most organizations start with project-level accounting because it’s the most concrete. But stopping there is a mistake. Without portfolio and capability metrics, the innovation department can’t demonstrate its strategic value to the board.

For a deep introduction to the discipline, start with What Is Innovation Accounting?, which covers the history, the three levels, and the key principles. Then move to Innovation Accounting in Practice for real-world implementation patterns. When teams are ready to act on their data, How to Make Pivot or Persevere Decisions in Your Innovation Accounting walks through the decision framework step by step.

Innovation accounting gives us the language. Now: what exactly should we measure?

Innovation Metrics: What to Measure at Every Level

Picking the right metrics is where most innovation programs get stuck. The instinct is to grab whatever is easy to count (number of ideas generated, hackathons run, patents filed) and present those to leadership. The problem is that activity metrics don’t tell us if we’re creating value. Running 50 experiments is meaningless if none of them tested a real assumption.

We need to think about metrics in layers.

Project-level metrics focus on learning velocity and assumption risk reduction. How many critical assumptions has this project tested? What is the team’s experiment cycle time? How much uncertainty have we reduced since the last funding decision? These are the metrics that tell us whether a specific project deserves more investment or needs to pivot.

Program-level metrics look at the health of the innovation pipeline. How many projects are in each stage? What’s the stage-gate pass rate? What’s the average time from idea to validated business model? How much are we spending per stage? These metrics tell us if the innovation process is working.

Organization-level metrics are the hardest to define but the most important for long-term success. They measure innovation capability itself. Are teams getting faster at running experiments? Is the quality of hypotheses improving over time? Are we retaining innovation talent? Is cross-functional collaboration increasing?

One metric category that often gets overlooked is diversity. Research consistently shows that diverse teams produce more creative solutions and catch blind spots that homogeneous teams miss. Measuring the diversity of our innovation teams, their perspectives, and their approaches is itself an innovation metric.

For a comprehensive framework on which metrics to choose, read What Metrics Should Innovation Programs Measure?. To understand the hierarchy of metrics across levels, see Levels of Innovation Metrics. For guidance on tailoring metrics to specific functional areas, Applying Innovation Metrics and Objectives by Department provides practical templates. And for the diversity angle, Diversity Is an Innovation Metric makes the case for why it belongs on every innovation dashboard.

If a metric doesn’t change how we allocate resources, fund projects, or coach teams, it’s not a metric. It’s decoration.

Innovation ROI: How to Calculate and Communicate Returns

ROI is the question that never goes away. Every executive sponsor, every board member, every CFO eventually asks: “What’s the return on our innovation investment?”

It’s a fair question. But answering it requires us to be honest about what ROI can and cannot tell us in an innovation context.

Traditional ROI is simple: (Gains - Cost) / Cost. The problem is that for most innovation projects, the “gains” part of the equation is speculative. We’re investing in projects precisely because we don’t know what the return will be. If we already knew, it wouldn’t be innovation. It would be execution.

We can absolutely calculate ROI. We just have to match the method to the maturity of the project.

For early-stage innovation projects (Horizon 3, exploring entirely new business models), we should think about option value rather than traditional ROI. Each early investment is like buying a financial option: we pay a small premium now for the right (but not the obligation) to make a larger investment later, once we have more information. The value of the option is the value of the information we gain, not the revenue we expect.

For mid-stage projects (Horizon 2, scaling validated concepts), we can start to apply more traditional financial models, but with appropriate uncertainty ranges. Rather than claiming a project will generate $10 million in revenue, we might say there’s a 60 percent chance it generates between $5 million and $15 million, based on the assumptions we’ve validated so far.

For late-stage projects (Horizon 1 adjacencies), standard ROI calculations work reasonably well because we’ve reduced most of the uncertainty.

The key to communicating innovation ROI to leadership is matching the rigor of the calculation to the maturity of the project. Presenting a precise ROI number for an early-stage exploration project is dishonest. Presenting a range with explicit assumptions is credible.

We also need to talk about portfolio-level ROI. The innovation portfolio should be evaluated like a venture capital fund, not like a series of individual projects. Some projects will fail (that’s the point). Some will return modest results. A few will generate outsized returns. The question isn’t whether every project pays off. The question is whether the portfolio as a whole generates returns that justify the investment.

For the foundational case for innovation ROI, read The ROI of Innovation. To evaluate whether the team structure itself supports returns, see Is Your Innovation Team Contributing to ROI?. And for a specific look at whether traditional finance tools apply, Can You Use a Net Present Value Formula for Innovation Projects? explores the limitations and alternatives. To understand the true cost picture, How to Calculate the True Cost of Your Innovation Team breaks down the full investment, and How to Calculate Opportunity Cost puts it in context.

Innovation Portfolio Management: Managing a Portfolio of Bets

Innovation accounting and metrics tell us how individual projects are doing. Portfolio management tells us whether our collection of bets makes strategic sense.

Most large organizations don’t have an innovation measurement problem. They have an innovation portfolio problem. They have dozens (or hundreds) of projects at various stages, funded through different mechanisms, with no consistent way to compare them or decide which ones deserve more resources.

Good innovation portfolio management starts with a few principles:

Principle one: balance across horizons. A healthy portfolio has projects in all three innovation horizons. Horizon 1 (core improvements and extensions) generates near-term returns. Horizon 2 (emerging opportunities in adjacent markets) builds the next generation of revenue. Horizon 3 (transformative bets on new business models) creates future options. Most organizations are overweight in Horizon 1 because it feels safer. We’ve worked with companies running 40 innovation projects, 38 of which were incremental improvements to existing products. That’s not a portfolio. That’s a maintenance backlog with a fancy name.

Principle two: stage-appropriate investment. Early-stage projects should get small, time-limited funding to run experiments and reduce uncertainty. As projects validate their assumptions, they earn the right to larger investments. This is the “metered funding” approach. It’s analogous to how venture capitalists invest: seed rounds are small, Series A is larger, and growth rounds are larger still. Each round is contingent on evidence of progress.

Principle three: kill projects early and often. The most expensive mistake in innovation isn’t funding a project that fails. It’s continuing to fund a project that should have been killed three quarters ago. Good portfolio management requires clear criteria for when to stop investing. And leadership needs to celebrate killing projects as a sign of discipline, not failure.

Principle four: governance that enables, not gates that block. Traditional stage-gate processes were designed for product development, not innovation. They tend to kill novel ideas early because novel ideas can’t pass the same gates as incremental improvements. Innovation portfolio governance needs to be flexible enough to let uncertain projects survive long enough to gather real data.

For a principles-first approach to portfolio management, read Managing Innovation Portfolios: Principles Before Tools. For a look at how governance frameworks have evolved, see Governance and Portfolio Management in 2023.

That covers most of what we need. But there’s one more piece: forecasting.

Innovation Forecasting: Predicting Outcomes Under Uncertainty

Forecasting innovation outcomes sounds like an oxymoron. But we’re not trying to predict specific outcomes. We’re trying to predict ranges of likely outcomes based on what we know today, and then update those predictions as we learn more. That’s fundamentally different from the deterministic forecasting used in established businesses, where we extrapolate from historical data. In innovation, we model uncertainty explicitly.

The most powerful tool for this is Monte Carlo simulation. Instead of calculating a single expected outcome, we run thousands of simulated scenarios, each with slightly different assumptions, and look at the distribution of results. This gives us statements like, “There’s a 70 percent chance this portfolio generates between $20 million and $80 million in new revenue over five years, and a 15 percent chance it generates more than $100 million.”

That kind of probabilistic forecast is far more honest (and far more useful) than a single-point estimate. It communicates uncertainty explicitly. It gives leadership the information they need to make resource allocation decisions. And it can be updated as new data comes in, getting more accurate over time.

The challenge, of course, is that building good forecasting models requires both the right data (from our innovation accounting system) and the right tools. This is where AI-powered forecasting tools come in.

Krobar.ai is our purpose-built innovation forecasting tool. It takes the data from an innovation portfolio (assumption validation rates, experiment results, stage progression) and runs Monte Carlo simulations to generate probabilistic forecasts. It answers questions like: “Given our current portfolio and learning velocity, what range of outcomes should we expect in 12 months?” and “If we increase our experiment budget by 20 percent, how does that change the distribution of outcomes?”

Innovation accounting data plus AI-powered simulation moves us from “we think this project is promising” to “based on 10,000 simulated scenarios, here’s the probability distribution of outcomes for this portfolio.” That’s a different conversation to have with the board.

For the mechanics of simulation-based forecasting, read Using a Monte Carlo Simulation to Forecast Innovation Outcomes. For the most common pitfall in business forecasting (overconfidence in point estimates), see The Most Common Error in Business Forecasting. And for a practical guide to making better estimates even without sophisticated tools, The Art of Guesstimation is a great starting point.

Ready to forecast your innovation portfolio? Try Krobar.ai and see what your data says about where your innovation efforts are heading.

Common Mistakes When Measuring Innovation

Even with the right frameworks, we see organizations make the same measurement mistakes over and over. Here are the ones that cause the most damage.

Using corporate growth metrics for innovation projects. We’ve seen companies set a $10 million revenue target for a Horizon 3 team that hasn’t even talked to customers yet. That’s like grading a kindergartener on their SAT scores. Revenue targets, market share goals, and profit margins are the wrong yardstick for early-stage innovation. Applying them prematurely kills projects before they have a chance to find product/market fit. For more on this trap, read The Danger of Corporate Growth Metrics.

Starting from scratch every time. Too many innovation programs ignore the measurement infrastructure that already exists in the organization. Financial reporting systems, product analytics, customer research databases — these are all sources of data that can feed into innovation accounting. Starting from scratch wastes time and creates unnecessary organizational friction. See Why Are You Starting Your Innovation Program from Scratch? for a better approach.

Measuring activity instead of progress. Counting the number of ideas, workshops, or prototypes tells us nothing about whether we’re creating value. Activity metrics feel productive but they’re vanity metrics in disguise. Measure learning (assumptions validated or invalidated) and decisions made (pivots, perseveres, kills). If the numbers on the dashboard don’t make anyone uncomfortable, they’re the wrong numbers.

Not connecting innovation metrics to business strategy. If the metrics dashboard doesn’t link back to the corporate strategy — which growth areas are we pursuing, what strategic risks are we hedging — it’s just a reporting exercise. The innovation portfolio should be a direct expression of the corporate strategy, and the metrics should reflect that connection.

Lessons Learned

No organization needs to implement all five disciplines at once. Start with project-level accounting — can we measure whether individual projects are making progress? Then build up to portfolio-level metrics. Then add ROI communication for the board. Then layer in forecasting as the data matures.

The organizations that measure innovation well share one trait: they treat measurement as a decision-making tool, not a reporting burden. Every metric should change a decision. If it doesn’t, cut it.

Need help building an innovation measurement system for your organization? Our coaching and consulting team works with corporate innovation leaders to design and implement innovation accounting, metrics, and portfolio management systems tailored to their context. Get in touch to start a conversation.

FAQ

What is the best single metric for measuring innovation?

There is no single best metric. Innovation measurement requires a system of metrics at different levels: project-level metrics (assumption validation rate, experiment cycle time), portfolio-level metrics (stage distribution, kill rate, time to validation), and organization-level metrics (innovation capability, talent retention). The best single starting point is the number of critical assumptions validated or invalidated per unit of time, because it directly measures learning velocity.

How do we measure innovation ROI when projects haven’t generated revenue yet?

For early-stage projects, we use option value rather than traditional ROI. Each early investment buys information that helps us make better future investment decisions. We can quantify this by tracking how much uncertainty we’ve reduced per dollar spent. As projects mature and move closer to market, we shift to more traditional financial models with explicit uncertainty ranges (for example, “60 percent probability of generating $5 million to $15 million in three years”).

How often should we review innovation metrics?

Project-level metrics should be reviewed at every team sprint or experiment cycle (typically every one to two weeks). Portfolio-level metrics should be reviewed monthly or quarterly by the innovation board or steering committee. Capability-level metrics (organizational innovation health) should be reviewed annually. The cadence matters less than the action: every review should result in at least one explicit decision (continue, pivot, kill, scale, or reallocate).

What is innovation accounting and how is it different from regular accounting?

Innovation accounting is a method for measuring progress when traditional financial metrics like revenue and profit are still zero. Regular accounting looks backward at what happened. Innovation accounting looks forward at what we’re learning. It tracks leading indicators (validated assumptions, reduced uncertainty, experiment velocity) rather than lagging indicators (revenue, market share). It operates at three levels: individual project progress, portfolio health, and organizational innovation capability.

Do we really need AI tools for this, or can we just use spreadsheets?

Spreadsheets work fine for project-level accounting. But once a portfolio has more than a handful of projects, the combinatorics get ugly fast. AI-powered tools like Krobar.ai use Monte Carlo simulation and machine learning to generate probabilistic forecasts from innovation portfolio data. Instead of single-point estimates (“this project will generate $10 million”), AI models produce probability distributions (“there’s a 70 percent chance this portfolio generates between $20 million and $80 million”). As the AI ingests more data from your innovation accounting system, the forecasts become increasingly accurate. This is a significant improvement over spreadsheet-based planning and gut-feel estimation.