By Wiles Kase, Engagement Leader at Clareo, and Justin Kanthak, Lesaffre
Key takeaways:
- Reported hit rates (50%+) are vanity metrics. The real number is 5 to 10%, and chasing a higher hit rate just means padding launches with safe line extensions instead of real bets (see Yoplait vs. Chobani).
- Not all failures are equal. Judge them by the quality of the bet, not the outcome, or teams stop taking smart risks (Crystal Pepsi funded Pepsi Max and, decades later, Poppi).
- Zombie projects, not failures, damage portfolios. Set elimination criteria upfront and reward whoever retires risk cheapest and fastest, the way Poppi’s founders validated demand before chasing national retail.
Food R&D leaders live a strange contradiction. Their dashboards routinely report innovation success rates north of 50%. Yet their new-launch data show six of every seven new products disappear inside two years. Neither number is wrong. They measure different things.
A 10% innovation success rate in food is the price of real innovation, if you’re measuring the right things. Food R&D leaders should strive for bolder innovations that fail faster, yet in aggregate result in bigger wins. The more useful metric is a higher conversion rate, not a higher hit rate. This requires killing the wrong ideas faster, so the right ones can thrive.
Most food and consumer packaged goods (CPG) leaders already know their reported hit rates are misleading. Fewer have dealt with why. Here are three principles we have found from experience for amplifying innovation performance.
1. Measure what leads to growth
As a baseline, we estimate the real underlying hit rate for food innovation is between 5 to 10%, not 50%+. The estimate is grounded in industry experience and roughly tracks baseline category and market growth.
A manager staring at 10% will ask, “How do I improve it?” Many managers do so by padding the numbers with safer innovations. with more “hits” but not necessarily more “runs.” That is a distraction.
Mintel reports that only 35% of 2024 global CPG launches were genuinely new products; the rest were renovations, reformulations, and pack changes. In food and drink specifically, the share of true innovation as defined by Mintel (not simply new for the sake of it, but considered, relevant and built for the future) has fallen from roughly 50% in 2007 to 26% in 2024.
We see this battle between safe bets and real innovation in every sector. What is uniquely perilous about food is that “distraction by safe bets” lurks at every turn: line extensions, new claims, new colors (then removing colors), new sweeteners (removing sweeteners), and so on.
Take Yoplait. Between 2007 and 2017, Greek yogurt went from roughly 1% to 50% of the U.S. yogurt category. Yoplait, then General Mills’ flagship brand and the U.S. market leader with about 25% share in 2011, focused on line extensions: Yoplait Light, Yoplait Whips, new flavors, new packs. Chobani started from zero in 2007 and passed $1 billion in annual sales within five years of launch. Yoplait didn’t launch its own Greek line until early 2010, and never caught up. By 2016, Yoplait was down to 19% share. Meanwhile, the company’s innovation dashboards kept reporting success.
Redefine what counts as a “hit.” A real hit does two things. First, it creates material new growth, specifically excluding reliable but meagre sales gains. Second, it generates learnings that inform and improve the next round of innovation. Anything else is just product management with an edgier label, which can cost you the next Greek yogurt.
2. Reward the right kind of failure
Not all “low” hit rates are created equal. A 10% hit rate can mean a team is taking well-researched swings at meaningful consumer trends and the market is not ready. Or it can mean a team is launching products that any first-pass concept test should have stopped.
Crystal Pepsi (1992) was the right kind of flop. PepsiCo bet on the clean/clear-drink wave with 1,000 concepts and 3,000 formulations in development, generated ~$0.5B in year-one revenue, and applied the lesson (Pepsi Max in 1993, and now Poppi today) that taste and function, not visual novelty, were the durable levers.
The recommendation: judge every failure on the quality of the bet, not just the outcome. A defensible failure is fuel for the next opportunity. A sloppy failure is waste. If your review process treats them the same, teams will stop taking the defensible yet high risk small bets, because there is no upside for having taken them. Build a review that separates the two, and reward the teams that fail right.
3. Kill the zombies
Failures rarely drag down food R&D portfolios. It is the zombie projects that tank an innovation portfolio. Every dollar and hour spent on a doomed project are resources taken from other priorities. It is the classic sunk cost fallacy, though it really should be labeled the “R&D budget trap.”
The best innovation teams commit to “kill criteria” at the outset. They measure how cheaply and quickly the team retires uncertainty, not how many gates were passed. They reward the insight that cost $50,000 to discover, because it prevented a could-be five million dollar misfire. The right scoreboard is not “did this project succeed?” It is “how much risk did we retire per dollar spent?”
Learning velocity looks like Poppi. Before PepsiCo’s $2 billion acquisition, its founders spent years building consumer demand through direct channels — social, DTC, and independent grocers — answering the hard question (“will people pay for prebiotic soda?”) long before pursuing mass distribution. Each stage cost a fraction of a traditional CPG launch and retired a distinct risk. National retail was the last question they answered, not the first.
Enhancing learning velocity means finding the question most likely to kill an idea and answering it first, as cheaply as possible. For a food team, that usually means putting a product in front of real buyers with real money at stake before committing to a full launch. A small direct-to-consumer run, a trial in one city, or a limited placement with independent retailers will reveal more about genuine demand than another round of concept testing.
The recommendation: define kill criteria before you commit resources, and reward the team that applies the criteria with rigor and honesty. Free the budget to catch the next Chobani rather than a “New and Improved” disappointment.
The companies that figure this out are more likely to manage portfolios built on more and, ultimately, better bets. Fewer singles, more swings at the fence and, over time, more home runs.
Wiles Kase is an Engagement Leader at Clareo, working across the firm’s Mining, Energy, and Food practices to help clients build new businesses and manage innovation at scale. He has supported food clients looking to launch new products, enter new segments, and commercialize novel ingredients. Wiles leads venture investment due diligence for food clients and has led Seed and Series A investments in food, regenerative agriculture, and cleantech.
Justin Kanthak is a food industry veteran with expertise spanning culinary arts, food science, and commercial strategy. Throughout his career, he has led growth initiatives as a sales leader, guided global portfolios, and driven innovative solutions to market. Drawing on hands-on technical knowledge and global leadership experience, Justin writes about food innovation, market trends, product commercialization, and the evolving challenges shaping the global food and beverage industry.
