The Right Kind of Friction Protects Your Work

In the late 1970s, a British designer named James Dyson grew frustrated with his vacuum cleaner. It lost suction almost as soon as he started using it, and its bag, the very thing meant to collect the dust, kept clogging and choking the machine it was supposed to serve. Around the same time, he noticed a nearby sawmill using a giant cyclone to spin sawdust out of the air, and he wondered whether the same principle might work inside a vacuum. It seemed like a crazy idea, but it was worth a go. So he went out to his coach house and began building prototypes.

The first one didn’t work, and neither did the tenth, the hundredth, or the thousandth. Over roughly five years, Dyson built 5,127 prototypes, each a little different from the last and each teaching him something the previous one couldn’t. On paper, it was the least efficient way imaginable to build a vacuum cleaner: five years of wrong answers, stacked one on top of another. Yet Dyson has said many times that every failure taught him something, and that the answer was hiding inside the struggle.

There was no way for him to skip straight to number 5,127. The person capable of building that final prototype only existed because of the 5,126 that came before it. That’s the idea I want to explore here. Some friction looks like inefficiency in the moment and turns out to be the only path to something that matters, while other friction is administrative sludge that deserves to be removed as ruthlessly as possible. The real skill of effort is learning to tell the two apart, before a powerful tool makes the choice for you.

I’ve been working through a framework called VERVE. It comes out of my work on what I call the infinite lever, meaning AI, and it describes how we stay valuable, and frankly how we stay human, as AI reshapes the way we create and work. Voice is the first principle. Effort is the second, and the question at its center sounds simple, even though most of us have never answered it honestly: what do you actually do when you’re working?

The blind spotWhat do you actually do when you’re working?

I’m not asking about your title, your deliverables, your projects, or the meetings on your calendar. I’m asking what you really bring to the table while the work is happening. It’s the first question I ask any leader when I begin working with an organization, and many creative professionals and leaders can’t answer it. That blind spot is about to become very expensive. When you don’t understand the true value you add, you will inevitably automate the wrong things, handing off the very part of the work that made you valuable while clinging to the parts that were only ever in the way.

Here’s how that happens, slowly and without anyone noticing: we become so good at doing a thing that we begin to believe the thing we do is our job. The designer becomes so skilled at building layouts that she believes her job is building layouts. The writer becomes so fluent at producing clean paragraphs that he believes his job is producing paragraphs. The leader becomes so efficient at running meetings that she begins to believe running meetings is leadership. Our expertise is a gift, but it can also become a blindfold. The better we get at the visible part of the work, the easier it is to forget the invisible part that made us good in the first place.

The layout, the paragraph, and the meeting are containers. The real job is something much more fundamental: bringing your intuition, your purpose, and your experiences to bear on a problem in a way no one else can. The designer’s job is to see, the writer’s job is to notice, and the leader’s job is to discern. Those are uniquely human things, and they can’t be automated. So the first step in practicing effort is to define, honestly and precisely, what you do when you’re actually working, in terms of behavior you would recognize if you saw it. Until you can name it, you can’t protect it.

Support work is easy to spot once you start looking for it. By all means, automate it, delegate it, and get it out of the way, because every hour you reclaim is an hour you can give back to the thing only you can do. AI is genuinely great at this kind of sludge, and anyone who tells you to resist that help is asking you to protect the wrong thing.

Soul work is different. The temptation of this moment is to treat it as one more inefficiency to be engineered out of the process, but that friction isn’t a flaw in the process. It is the process. It’s where your thinking happens and where your capacity actually grows, and when you remove it, you don’t arrive at the same result faster. You arrive at a lesser result, and you slowly become a lesser creator. I call this the danger of results without reps, and it’s the Proximity trap: you end up standing next to work you didn’t really generate, able to produce the output but never having built the muscle, until eventually you lose the very ability that made you worth hiring.

This matters because your most valuable asset is your intuition. It isn’t magic, and it isn’t a hunch pulled from thin air. It’s the accumulated residue of every problem you’ve wrestled with, every failure you’ve learned from, and every pattern you’ve quietly noticed over years of doing the work. It’s what allows a seasoned creative director to look at a concept and know in three seconds that something feels off, long before she can explain why, and what allows a leader to walk into a room and sense the tension nobody is saying out loud. That kind of knowing can’t be downloaded or delegated, because it’s built one rep at a time. Outsource the friction, and you starve your intuition of the very experiences it needs to grow. That’s why Dyson’s prototypes were soul work, and why no one could have handed him number 5,127 on the first day and gotten the same inventor at the end.

The most effective artists and leaders seem to understand this instinctively. When Ernest Hemingway was asked how much rewriting he did, he said he rewrote the ending of A Farewell to Arms 39 times before he was satisfied. Asked what had stumped him, he answered simply: getting the words right. He could have settled for an ending that was good enough, and probably nobody would have noticed. But he understood that the struggle to find exactly the right words wasn’t standing between him and the work. It was the work.

The same instinct shows up in business. In 2004, Jeff Bezos told his senior team at Amazon that they would no longer use slide decks in meetings. Anyone proposing an idea would write a narrative memo of up to six pages instead, and every meeting would begin with the whole room reading it in silence before anyone said a word. To an outsider, that must have looked like a staggering waste of executive time. Yet writing a complete narrative forces you to connect your ideas and confront the gaps in your logic in a way that slides let you hide.

Bezos added friction on purpose because he knew friction was where the thinking happened.

So where does your thinking happen? I ask because an idea is gaining quiet momentum, usually in the way we talk about AI, that work itself is the enemy, and that the ultimate goal is to remove as much effort from our lives as possible so we can finally be free. Work, at its best, is one of the great gifts of being human. It gives us identity, purpose, and the chance to put our perspective into the world where other people can interact with it. When we strip all of the effort out of our lives, we don’t become more free. We become less ourselves. None of this means we should glorify busyness, but it does mean being wise about which kinds of effort we keep.

The deadline testSitting with the problem is the efficient move.

The hardest place to protect soul work is under pressure, when every instinct tells us to prove our value by responding right now. Carla Johnson has spent more than two decades helping organizations innovate by working directly against that instinct. She coined the term innovation architect, and she’s the CEO of Rethink Labs. Her curiosity compass describes four zones of thinking, and the one most of us default to under pressure is the efficiency zone, where the questions are about how much time we have, what the budget is, and when the deadline hits. As Carla put it, you can’t be curious when you’re that focused on details, because it doesn’t give your brain room to explore and expand.

Her alternative is to stop, sit back, and spend ten or fifteen minutes doing nothing except asking questions about what’s right in front of you, beginning with whether you’re even solving the right problem. Even when it turns out to be the right problem, a team that spends that time asking questions walks away with a completely different perception of it. Because everyone arrives with their own experience and background, they also leave with a more cohesive picture of the next best step.

I find that striking, because it’s almost the inverse of how the AI conversation tends to go. AI can make us efficient in the short term by handing us a quick result. In the long term, it can make us less efficient if we’ve abdicated the strategic thinking. Building a little apparent inefficiency into the front of the process lets us marinate in the problem and approach it from the core of who we are, rather than reactively filling in the blank because there’s pressure to respond.

Carla also names the thing that most often keeps us from that kind of exploration: our own expertise. We’re very good at knowing what we know, and that makes us blind to what’s possible. So she asks people to swap hats and imagine they’re a Lego designer or a magician, and she’s watched very left-brain, framework-minded people embrace the absurdity almost immediately, because they feel emotionally disconnected from the outcome.

I’ve seen the same thing in an exercise I run with teams. I ask how we could make commercial air travel fun, and they come up with a see-through floor with cameras on the bottom of the plane, or exercise bikes in the back, and everyone has a great time. Then I hand them the problem their own manager gave them, and they freeze. They have too much experience and too much knowledge, and they begin discounting ideas for fuel efficiency and turbulence before they’ve even been spoken. Carla calls this brand detachment disorder, the reflex that says an idea will always work over there but it’s not how we do things here. She says that telling ourselves “we’re fine” may be one of the biggest lies we ever tell ourselves.

She also told me about a software company in Atlanta that set aside a week twice a year for the entire company to work on innovation, and the chief technology officer insisted it include everyone, not only the tech people. A woman on the finance team came back from the first week and asked what opportunity she might have been missing in her own work. She realized she spent 40 hours every month, a full week, building one report by hand. So she taught herself a programming language, and now the report takes 12 minutes. Carla describes it as doing 15 months of work in 12, because she gave herself back a week every month.

That’s support work, and she was exactly right to eliminate it. The two ideas belong together: we remove the friction that’s in the way of the work so we have the time, the energy, and the curiosity to embrace the friction that is the work. The ideas that matter most often take years to ripen. No prompt will ever produce them on demand, because they’re the result of intuitive layers that accumulate over time.

If you’d like a ready-made worksheet for this, every episode of Daily Creative comes with a free Action Guide at toddhenry.com/start, because it’s not what you know but what you do that matters. And if you want to go further on the practice of curiosity, Carla’s book Rethink Innovation is a good place to start.

Go back to that coach house, and to the long stretch of years when James Dyson was building prototype after prototype after prototype. If someone had handed him the finished design on the first day, he might have had a better vacuum sooner. But he never would have become the inventor capable of building the next thing, and the thing after that. To anyone watching from the outside, the failures looked like waste. They were the slow, unglamorous reps that built the person who could finally see the answer.

Those 5,126 failures were the work, and they were what made him.