Creative Problem Solving: Why the Best Business Ideas Come from the Cheapest Experiments

In 1999, Nick Swinmurn spent an afternoon driving between shoe stores in San Francisco looking for a pair of Airwalk Desert Chukka boots. One store had the right color but the wrong size. Another had the right size in the wrong color. He left empty-handed and irritated. Then he asked a question that most people would have filed under "shower thought" and forgotten: what if you could buy shoes online?

The obvious next step would have been to build an e-commerce platform. Negotiate wholesale deals with shoe brands. Rent warehouse space. Hire developers. Invest months and hundreds of thousands of dollars before a single customer placed an order.

Swinmurn did something else entirely. He drove to local shoe stores, asked permission to photograph their inventory, and posted the photos on a bare-bones website he called Shoesite.com. When someone placed an order, he drove back to the store, bought the shoes at full retail price, and shipped them to the customer himself. There was no warehouse. No wholesale agreements. No inventory. The entire "business" was a man with a camera and a car.

The customers had no idea. From their perspective, they were buying from a real online shoe store. And enough of them bought that Swinmurn had his answer: people would buy shoes on the internet. He renamed the site Zappos. Ten years later, Amazon acquired it for $1.2 billion.

Creative problem solving, the practice of finding solutions that conventional thinking would miss, rarely looks like genius. It looks like a photographer in a shoe store. The most reliable creative method in business isn't thinking harder about the problem. It's designing the cheapest possible experiment to test whether the problem is worth solving at all.

The Cheap Test Principle

In 2008, Drew Houston had a working prototype of a file-syncing tool, but every time he tried to explain what it did, people's eyes glazed. "It syncs your files across computers" sounded boring. It sounded like a feature, not a product. Houston needed to prove demand before he spent years engineering a full platform, but the concept was almost impossible to explain in words.

So he made a short demo video. Not a polished commercial. A screen recording of the product in action, narrated casually, with Easter eggs aimed at the Reddit and Digg communities he wanted to reach: references to "Chocolate Rain," the HD DVD decryption key controversy, even an XKCD comic. He posted it. Within twenty-four hours, the Dropbox waitlist exploded from 5,000 to 75,000 names. No advertising. No sales team. No finished product. Just a video that cost nothing and proved everything.

The pattern shows up wherever you look for it. A physicist named Andre Geim spent his career running the cheapest experiments he could design, seeking ideas that were low-risk but high in potential payoff, testing each one quickly, and cycling to the next. One of those experiments involved peeling layers of graphite with Scotch tape. The materials cost less than five dollars. The result was graphene, a material one atom thick and stronger than steel. It won the Nobel Prize in Physics.

Swinmurn validated a billion-dollar idea with photographs and a car. Houston validated a category-defining product with a short demo video. Geim won the highest prize in science with tape and a pencil. In each case, the breakthrough didn't come from the scale of the investment. It came from the cleverness of the test.

What Is Creative Problem Solving? (And Why Is It a Method, Not a Gift?)

The philosopher Karl Popper argued that the growth of knowledge comes not from proving things right but from proving things wrong. Bold conjectures, he argued, followed by severe critical tests designed to refute them. The faster you can disprove a bad idea, the faster you arrive at a good one.

This is what creative problem solving actually is. Not talent. Not inspiration. A structured cycle of generating hypotheses, fueled by daily ideation practice, and killing the bad ones as cheaply as possible. When you look at what Swinmurn, Houston, and Geim did, the structure is identical:

Observe a problem that bothers you or someone you're watching. Swinmurn couldn't find shoes. Houston kept forgetting his USB drive. Geim noticed unexplored physics in everyday materials.

Form a hypothesis about a possible solution. People will buy shoes online. People will pay for seamless file syncing. Graphite might yield interesting single-layer structures.

Design the cheapest experiment that could disprove the hypothesis. Photos on a website. A demo video. Scotch tape on graphite.

Analyze the results honestly. Did anyone buy the shoes? Did anyone sign up for the waitlist? Did the tape produce a single-atom layer?

Iterate. Adjust the hypothesis based on what you learned and run the next experiment. Swinmurn renamed the site, refined the selection, eventually built real infrastructure. Houston built the actual product after the video proved demand. Geim refined his technique across multiple experiments.

This five-stage cycle, observation to hypothesis to experiment to analysis to iteration, is the structure underneath every creative problem-solving success story. The idea that you need a completely original idea is a myth; what you need is a cheap way to test whether your version of an existing idea serves customers better. The companies that skip steps, that invest millions before running a single cheap test, are the ones that appear in case studies about spectacular failure.

Why Do Cheap Experiments Beat Expensive Plans?

Quibi raised $1.75 billion before its app was built. The concept was short-form premium video for mobile. Jeffrey Katzenberg and Meg Whitman assembled the most credentialed team in streaming history, hired the biggest names in Hollywood, and launched on April 6, 2020. The app reached number three on the App Store on its first day. By late May, an estimated 72,000 of its 910,000 free-trial users had converted to paying subscribers. Quibi announced its shutdown in October 2020, barely six months after launch. Katzenberg blamed COVID. Netflix added over 10 million subscribers that same quarter.

Quibi skipped the experiment. They went from hypothesis to full-scale build with $1.75 billion between the idea and the first piece of customer feedback. Every dollar spent before launch was a dollar spent on an assumption that had never been tested.

The neuroscience of creativity suggests why cheap experiments work better than expensive plans. Research on the default mode network, the brain's system for spontaneous association and idea generation, shows that creative insights emerge through rapid toggling between generating possibilities and evaluating them. A 2025 study on divergent thinking analyzed brain scans from 2,433 participants and found that the frequency of switching between the brain's generative and evaluative networks predicted creative ability across all samples.

Cheap experiments mirror this neural process externally. Each test is a generate-evaluate cycle. You hypothesize (generate), test (evaluate), learn, and cycle again. The faster and cheaper each cycle, the more cycles you can run, and the more creative your output. Quibi ran one cycle with $1.75 billion. Swinmurn ran dozens of cycles with a camera and gas money.

The Constraint Advantage

Swinmurn couldn't afford inventory, so he photographed someone else's. Houston couldn't explain his product in words, so he showed it in a video. Geim couldn't fund expensive physics experiments, so he used office supplies. In each case, the constraint didn't limit the creative solution. It produced it.

This pattern is well-documented. In 2019, Oguz Acar, Murat Tarakci, and Daan van Knippenberg published a cross-disciplinary review of constraint-driven creativity and found that constraints consistently fuel more creative output than unlimited resources. Bob Sutton at Stanford's d.school puts it more directly: "Want some creativity? Crank up the constraints."

Unlimited resources let you avoid the creative problem. You don't have to figure out how to validate demand cheaply when you have $1.75 billion, so you don't. The constraint forces the creative leap. No money for inventory? Find a way to sell without it. Can't explain the product? Find a way to show it. Can't afford lab equipment? Find a way to do the experiment for five dollars.

For entrepreneurs, this means that being under-resourced isn't a disadvantage in the creative problem-solving process. It's the process.