Thoughts / Growth

How Zappos started by taking photos at local shoe stores

As founders, we often fall into the trap of thinking we need everything to be perfect before launching. The sophisticated tech stack, the polished design, the seamless automation – we convince ourselves these are all essential for day one. But some of the most successful companies started with surprisingly simple solutions, and Zappos is a perfect example of this mindset.

Back in 1999, Nick Swinmurn had what seemed like a crazy idea: what if people could buy shoes online?

Today, this sounds completely normal – we buy everything online. But in the late 90s, this was borderline absurd. The common wisdom was that no one would ever buy shoes without trying them on first.

This is the story of how Zappos proved everyone wrong, and what we can learn from their unconventional approach.

The birth of an idea

It started when Swinmurn went to a mall looking for a pair of shoes. After visiting multiple stores without finding what he wanted in his size, he wondered if there might be a better way.

This frustration led to a business hypothesis: people might be willing to buy shoes online if the process was convenient enough.

But how do you test such an assumption without investing millions in inventory and warehousing?

The $0 warehouse solution

Instead of raising venture capital and building a complex infrastructure, Swinmurn came up with an ingeniously simple solution.

He went to local shoe stores, asked permission to take pictures of their shoes, and posted these photos on his website, which he called Shoesite.com (later renamed to Zappos).

Here’s where it gets interesting: when someone ordered shoes from his website, he would physically go to the store, buy the shoes at full retail price, and ship them to the customer.

Yes, he was losing money on every sale. But more importantly, he was learning.

The hidden genius of this approach

As a designer myself, what fascinates me about this strategy is its elegant simplicity.

Instead of trying to solve every problem at once (inventory management, warehouse operations, supplier relationships, etc.), Swinmurn focused on validating one critical assumption: Will people buy shoes online?

This approach had several brilliant aspects:

First, it required almost no upfront investment. The only real costs were a basic website and the time spent photographing shoes.

Second, it provided immediate market feedback. Every order (or lack thereof) was a direct signal about customer behavior.

Third, it allowed for rapid iteration. Swinmurn could easily adjust his offering, pricing, and presentation based on real customer interactions.

Fourth, and perhaps most importantly, it let him start immediately. No waiting for inventory, no complex supplier agreements, no warehouse setup.

From manual to magic

Once Swinmurn had proof that people would indeed buy shoes online, he brought on Tony Hsieh as CEO. Hsieh, who had recently sold his company LinkExchange to Microsoft for $265 million, saw the potential in what was then still a small, scrappy operation.

Together, they started building what would become the real Zappos – with proper inventory, warehouses, and what would eventually become their legendary customer service culture.

But none of this happened overnight. They gradually transitioned from the manual, store-to-store process to a more scalable operation.

The power of manual first

In the startup world, we often talk about building “scalable solutions” from day one. But Zappos shows us that sometimes, starting with a completely manual, unscalable process can be the smartest approach.

This “manual first” approach has several advantages:

• It lets you validate your core assumptions before making major investments
• It provides rich, qualitative feedback about your customers’ needs
• It forces you to deeply understand every aspect of your business
• It helps you identify what actually needs to be automated (versus what you think needs to be automated)

Modern applications

Today, we might call Swinmurn’s approach a “Wizard of Oz” MVP – where you manually simulate what will eventually be automated. We see this pattern repeated in many successful startups:

• DoorDash‘s founders started by buying food themselves and delivering it
• Instacart began with the founders doing the grocery shopping
• Airbnb started with just photos of the founders’ apartment

Also read Doing things that don’t scale: 10 tips for startups

A different way of thinking about MVPs

As a designer and founder, I’ve learned that sometimes the best designs aren’t the most sophisticated – they’re the ones that most efficiently validate your core assumptions.

The Zappos story challenges us to think differently about what constitutes a viable first version.

Instead of asking “How can we build this perfectly?” try asking:

• What’s the core assumption we need to validate?
• What’s the simplest way to test this assumption?
• What manual processes could we use before building automation?
• What parts of our solution actually need to be perfect on day one?

Also read What is the definition of a MVP?

The $1.2 billion validation

In 2009, Amazon acquired Zappos for $1.2 billion. From a simple website with photos of other stores’ shoes, it had grown into one of the most successful e-commerce companies in history.

But none of that would have happened if Swinmurn had waited until he had the perfect solution.

For me, this story perfectly encapsulates what startup entrepreneurship is all about – finding clever, resourceful ways to validate assumptions and solve problems.

It’s about being willing to do things that don’t scale, to start small, and to learn quickly.

The next time you’re working on a startup idea, ask yourself: What’s your equivalent of taking pictures at local shoe stores?

What manual, unscalable process could you use to test your assumptions? Sometimes, the best way to build something big is to start surprisingly small.

Also read: How Dropbox started: The MVP strategy that launched a giant