If AI software is going to take over a bevy of repetitive tasks and processes, then what should organizations be doing with the free time that creates?
David Schonthal, a professor at the Kellogg School of Management at Northwestern University, says the answer may lie in something the Japanese call “shokunin.”
We spoke with Schonthal recently to dive into what that means, and discuss his recent articles in MIT Sloan Management Review and Harvard Business Review.
1. Understanding Four Types of Friction that New Ideas Face
Key idea: Schonthal contents that the main barrier to innovation isn’t bad ideas, but predictable forms of human resistance.
“Friction theory is what we call it, and it was developed by myself and a Kellogg colleague of mine named Loran Nordgren. Loran is an experimental psychologist, and when we wrote this book…called The Human Element… [we were] trying to understand why it was that individuals and organizations say no to what are clearly good ideas…”
“It starts with this core foundational understanding that human beings are hard-wired to resist change for a number of different psychological and evolutionary reasons…”
“[We] break down resistance into four types of frictions. The first friction is the friction of inertia, which is a human being’s overwhelming tendency to stick with what they know, despite the fact that what they know is insufficient for the future… The second is the friction of effort, which is the real or often perceived amount of energy required for a change to take place, and it’s not just physical energy; it’s also cognitive energy… The third friction is emotional friction, which are the undesired negative feelings we inadvertently cause in the very people we’re trying to help… And then the fourth friction is something we would call reactance, which is a psychological phenomenon that effectively means people don’t want to be changed by other people. Doesn’t matter how good the idea is, I don’t want to feel like you are changing me.”
2. Shokunin: Use AI to Free Time for Craft and Core Competence
Core idea: Borrowing from the Japanese notion of the devoted craftsperson, Schonthal suggests that as AI automates routine work, organizations should reinvest that freed-up time in deepening the craft and quality at the heart of what makes them distinctive.
“I’m writing something right now with a collaborator of mine named Matt Holt. Matt is a an expert on Japan, [and] Japanese pop culture, and wrote a phenomenal book called Pure Invention about how Japanese pop culture came out of nowhere to take over the world. He and I are writing an article on this topic of shokunin. Shokunin is a Japanese term that effectively means craftsperson, and what we are talking about in this article is that if AI is able to take on a lot of the administrative functions that used to be done by humans, [and] if we have extra time available to us, what ought we spend that extra time on?”
How might AI allow you to spend more time doing that to create things that are truly special and resonant?
“One of the things that Matt and I put forward in this article, is that this is really an opportunity for people to go deeper into craft, to spend time on creating things of quality, creating things with a more human-centered angle to them…”
“If all of the other administrative stuff is taken off my plate, this actually gives me an opportunity to get deeper into craft and quality of things that are really important for my business. We used to say in business model design, never outsource the thing that makes you special. You can partner for just about anything. Never outsource your core competence.”
“There are elements of craft and ingenuity and creativity that are inherent to your organization that you can now spend more time on, whereas before it was always in battle with other things that were important to the core business. How might AI allow you to spend more time doing that to create things that are truly special and resonant?”
3. Question Zero: Using AI to Better Frame the Problem, Not Just Generate Ideas
Key idea: In a world where AI can generate endless ideas, Schonthal says real differentiation comes from asking a different, deeper first question — what he calls “question zero” — and using AI as a partner in framing the problem, not just ideation.
“…The most innovative companies in the world are solving different problems than everybody else, meaning they have framed the opportunity totally differently than everybody else. But…if everybody’s prompting AI with sort of obvious surface level prompts and certain surface level observations, you’re not going to get anything terribly rich.”
“Question zero is really figuring out the right question to begin with, [in order] to go on a more differentiated innovation journey. And that question is best served when it is a novel reframe of a problem that it’s likely your competitors, present or future, haven’t considered…”
“Let’s say AI is doing some listening in my data, and identifies some really interesting anomalies or behavioral patterns that are worth questioning. Maybe [I should be] doing a little bit of good old-fashioned jobs-to-be-done work, or design research, to find out if those things are in fact true. How are people framing those things in their minds? …I can then feed some of those reframed prompts into an AI and say, ‘All right, give me give me some alternative explanations for alternative ways of framing a problem such as A, B, or C.’”
Is there another problem inside of the data that maybe people haven’t considered?
“One example is Cursor, which is an AI-powered code editor… I think most people, when we think about code editors and AI, are thinking about Claude Code or GitHub Copilot… Those are about, how do we design code faster? …Cursor took a different tack, which was, is the problem that people can’t write code fast enough? …And do we want to get into an arms race with some of these other companies, Anthropic and Microsoft, in order to do that? Or is there another problem inside of the data that maybe people haven’t considered?”
“To make a long story short, what Cursor had observed in the data, and then what they validated through qualitative research, was there was a more foundational problem than writing faster code, which is a lot of the a lot of time was being spent by engineers just trying to read and understand the code that was written in the first place. …And so they reframed the problem from, ‘How do we help people write faster, better code,’ to, ‘How do we help designers understand the code base that they’ve been working with, make sense of it in a really intelligible and actionable way, and then take that forward?'”
Fetaured image by mostafa mahmoudi on Unsplash
You must be logged in to post a comment.