The Debunked Stat Behind a Popular Design Thinking Pitch

On a recent episode of Students Incorporated, the student-produced podcast at the International Community School of Bangkok (ICS Bangkok), co-host Premi cited one of the most repeated statistics in business media: nearly 30,000 new products launch every year, and 95% of them fail. The number came, she said, from Clayton Christensen, the late Harvard Business School professor best known for "disruptive innovation." It's a good line for making the case that companies need a better process for coming up with ideas. It's also, according to Christensen's own account, not something he ever said.

Researchers George Castellion and Stephen K. Markham went looking for the source of the "95% fail" claim years ago and asked Christensen directly. He denied ever citing that figure and said the real failure rate was lower, adding that the business world "has not done a good job empirically validating this fundamental fact." The number keeps circulating anyway, attached to his name in blog posts, consulting decks and, now, a high school podcast, because it's a useful number, not because anyone checked it.

That's a small irony sitting inside a much bigger one. The episode's first segment was built around design thinking, the human-centered problem-solving framework popularized by the design firm IDEO and Stanford's d.school. And design thinking, as a brand, has spent the past several years absorbing exactly the kind of scrutiny that stat should have gotten decades ago.

A framework under audit

The most visible sign of that scrutiny is IDEO itself. According to Fast Company, the firm's revenue fell from roughly $300 million in 2019 to under $100 million in 2023, a decline steep enough that IDEO cut a third of its staff and closed its Munich and Tokyo offices before installing a new CEO, Michael Peng, in 2025. Fortune reported in May 2026 that Peng now questions whether the thing IDEO built its identity on is even a differentiator anymore. "Customer centricity, the thing that IDEO has always stood for, just seems like it's table stakes now," he told the magazine, noting that more than half of companies already believe they're customer-centered, whether or not that's true. His answer isn't to retire the philosophy but to rebuild IDEO's pitch around it for an era when AI, not empathy interviews, is what clients are asking about.

The public argument got louder in 2018, when Pentagram partner Natasha Jen gave a talk at the 99U conference with a title blunt enough that it's still shorthand in design circles for the whole backlash, a flat, one-word rejection of the methodology itself, not just how companies execute it. Jen, whose client list includes the Guggenheim Museum and Harvard's Graduate School of Design, argued the methodology had calcified into a single ritual, Post-it notes on a whiteboard, standing in for the harder, less photogenic work of expert critique and testing. The talk went viral in design circles and gave language to a complaint that IDEO's own Michael Hendrix later conceded had merit: companies were staging what he called "the theater of innovation" without doing the underlying work.

None of that means the concept the ICS Bangkok hosts described is worthless. Earlier in the same segment, host Mr. Jason had opened by reading the field's own definition into the record, from Tim Brown, IDEO's executive chair: design thinking as "a human centered approach to innovation that draws from the designer's toolkit to integrate the needs of people, the possibilities of technology, and the requirements for business success." From there, the hosts, Mia, Frank, Premi and a co-host identified on air only as Highlight, moved through entertainment, sports, social issues and business before closing the segment with the framework distilled into five stages: empathize, define, ideate, prototype and test. The sports discussion produced the episode's sharpest data point. Mr. Jason, using stadium construction as a case study, cited research on the top five most expensive stadiums ever built, noting that SoFi Stadium in Inglewood, California cost $5.5 billion, still the most expensive sports venue ever constructed, a figure that works out to roughly $78,000 per seat. That stat holds up, and it's more relevant now than when it aired: SoFi is hosting eight matches in the 2026 FIFA World Cup this summer, part of a wave of North American stadium spending that other reporting has since scrutinized over public funding and whether fans actually benefit from it.

Where the exercise actually earned its keep

The stronger half of the episode, and the part with none of design thinking's branding baggage, was the second segment: a live demonstration of a technique called Worst Possible Idea, led by guest Fabrice, who joined specifically to teach it. The premise inverts how most brainstorms start. Instead of chasing the best idea immediately, which tends to produce overthinking or silence, the group first competes to generate the worst, most unworkable idea it can, then mines that list for anything usable. "The worst idea technique flips that completely," Fabrice told the hosts, describing how conventional brainstorming pushes people toward the best answer right away and how that pressure is exactly what stalls a room.

It helps remove the fear of failure and gets people laughing, which makes creativity flow even more freely.

Unlike "design thinking," which has become an umbrella term applied to nearly any structured creativity process, Worst Possible Idea traces to a specific person: Bryan Mattimore, a co-founder of the innovation consultancy the Growth Engine Company, who is widely credited with coining the exercise as a way to short-circuit groups stuck trying to sound smart in front of each other. Innovation trainers who still teach the exercise, largely unchanged, point to negativity bias as the reason it works: people are naturally faster and more fluent at imagining what goes wrong than what goes right, so the technique leans into that instinct instead of fighting it. It also strips out status anxiety. Nobody has to defend a bad idea they offered on purpose, which is precisely what makes a room full of teenagers, or executives, willing to say the quiet, ridiculous thing out loud.

The hosts ran two live examples on air. Asked for the worst way to handle customer service at a coffee shop, the group proposed an hour-long line, rude robot baristas and deliberately handing customers the wrong drink. Flipped, those became a faster ordering system, an entertaining AI chatbot for the wait, and a "surprise drink of the day" promotion. Asked for the worst way to fix communication breakdowns between departments, they landed on canceling every meeting, writing messages in code and delivering memos by carrier pigeon. Reversed, that produced shorter, more essential meetings and a more engaging internal messaging system. None of it was profound in isolation, but the mechanism, start bad, get specific, then reverse, is a repeatable move that a team can actually run without a facilitator's certification or a room full of Post-its.

Host Mr. Jason used his own history to make a related point about the difference between having a framework and actually applying it. Early in his career as a designer, he pitched his employer on sending him to a web design conference in Seattle, a trip that led directly to new client work and, eventually, a new department at the agency. Years later, launching his own co-working space, he skipped the empathy step entirely. "I neglected to walk in the shoes of my customers," he said, meaning he never fully understood the pain points of the people who'd actually use the space. The business closed after about a year and a half, in the wrong location, built on assumptions instead of the kind of listening design thinking is supposed to force.

The takeaway that survives the debunking

Put together, the episode inadvertently modeled the exact correction that design thinking's critics have been asking for. The Christensen stat is decoration, the sort of unverified number that makes a pitch sound more rigorous than it is. The five-stage framework is broad enough to bend around almost any subject, useful as a vocabulary but not as a method a team could run cold. What's left standing is the specific, teachable, attributable technique: a named inventor, a clear mechanism, and two on-air examples that turned genuinely bad ideas into usable ones in a few minutes each. That's a smaller claim than "a human-centered approach to innovation," but it's one that doesn't need a professor's name attached to it to hold up.

← Back to all posts