Viaweb doing things that don't scale

In the summer of 1995, as the commercial internet was just beginning to take shape, two friends made a decision that would not only change their lives but also help define the future of online commerce. Paul Graham and Robert Morris, both programmers with backgrounds in artificial intelligence and computer science, decided to start a company called Viaweb. Their goal was ambitious yet straightforward: to create software that would allow merchants to build online stores without requiring technical expertise.

What made Viaweb remarkable wasn’t just its innovative technology but the unscalable, hands-on approach its founders took to get the business off the ground. In an era when “doing things that don’t scale” wasn’t yet a startup mantra, Graham and Morris intuitively understood that sometimes the path to building something big begins with doing things in small, manual, and decidedly inefficient ways.

The Birth of an Idea

The mid-1990s marked the early days of the World Wide Web. Netscape had just gone public, igniting what would become the dot-com boom. But despite the growing excitement about the internet’s commercial potential, e-commerce was still in its infancy. The few online stores that existed had been painstakingly built by hand, requiring significant technical knowledge and resources.

Paul Graham and Robert Morris saw an opportunity in this gap. What if they could create software that would allow anyone—regardless of their technical abilities—to set up an online store? This idea would eventually evolve into Viaweb, the first software-as-a-service company, though that term wouldn’t be coined until years later.

The concept was revolutionary for its time: instead of selling software that merchants would install on their own computers, Viaweb would host the software on its servers. Merchants would access and configure their stores through a web browser, and Viaweb would handle all the technical aspects—hosting, security, transactions, and more.

But having a great idea wasn’t enough. Graham and Morris faced a significant challenge: convincing merchants, many of whom were still skeptical about the internet’s commercial viability, to trust their businesses to this new platform.

The Unscalable Beginning

When Viaweb launched, online store builders didn’t exist. The few merchants who had ventured online had either built their sites themselves or hired expensive developers to do it for them. Graham and Morris were offering something entirely new, and they quickly discovered that simply making their software available wasn’t enough to attract users.

Their solution was decidedly unscalable: they would build the stores themselves.

“A consulting-like technique for recruiting initially lukewarm users is to use your software yourselves on their behalf,” Graham would later write, reflecting on this period. “We did that at Viaweb.”

This approach meant that instead of just providing tools and letting merchants figure things out, Graham and Morris would personally work with each early customer to set up their online store. They would input product information, configure payment systems, and design the store’s appearance—essentially doing all the work that their software was designed to let merchants do themselves.

This hands-on approach served several crucial purposes:

1. Overcoming skepticism: By offering to build the stores themselves, Graham and Morris removed a significant barrier to adoption. Merchants who were hesitant about the technology or unsure of their ability to use it could still get online.
2. Learning from users: Working directly with merchants gave the Viaweb founders invaluable insights into their needs, pain points, and preferences. This firsthand knowledge informed the ongoing development of their software.
3. Building relationships: The personal touch created stronger connections with early customers, who became not just users but advocates for the platform.
4. Demonstrating value: By handling the setup process, Graham and Morris could ensure that each store looked professional and functioned well, showcasing the potential of their platform to other prospective customers.

This approach was labor-intensive and time-consuming—the very definition of something that wouldn’t scale as the company grew. But it was exactly what Viaweb needed in its early days to gain traction in a market that was still finding its footing.

The Technical Innovation Behind the Scenes

While the hands-on approach to building stores for merchants was a key part of Viaweb’s early strategy, the technical innovation happening behind the scenes was equally important. Graham and Morris weren’t just manually creating static websites; they were developing a sophisticated platform that would eventually allow merchants to manage their own stores.

The technical foundation of Viaweb was built on Lisp, a programming language that was unusual for web applications at the time but gave the company a significant competitive advantage. As Graham would later explain in his essay “Beating the Averages,” using Lisp allowed them to develop features more quickly than competitors who were using more conventional languages.

“Robert and I both knew Lisp well, and we couldn’t see any reason not to trust our instincts and go with Lisp,” Graham wrote. “We knew that everyone else would be writing their software in C++ or Perl. But we also knew that that wouldn’t be a problem for us, because we were better programmers than our competitors, and we would be able to develop software faster.”

This technical edge was crucial because it allowed Viaweb to rapidly improve its platform based on the insights gained from working directly with merchants. Each manually built store provided lessons that were quickly incorporated into the software, making it progressively easier for merchants to eventually take control of their own online presence.

From Manual to Self-Service

As Viaweb gained traction, the founders began a gradual transition from manually building stores to enabling merchants to do it themselves. This evolution wasn’t just about scaling the business; it was about fulfilling the original vision of empowering merchants with tools that made e-commerce accessible.

The transition happened organically. After setting up a store for a merchant, Graham and Morris would show them how to make updates and manage inventory themselves. Over time, as the software became more intuitive and merchants became more comfortable with the technology, the level of hands-on support required decreased.

This gradual shift from a service-heavy model to a more scalable self-service approach was carefully managed. The founders recognized that different merchants had different levels of technical comfort and different needs. Some quickly embraced the self-service tools, while others continued to rely on support for more complex tasks.

The key insight was that the manual, unscalable work at the beginning wasn’t just a necessary evil to get the business started—it was a valuable investment in understanding the market and refining the product. By the time Viaweb was ready to scale, its platform had been shaped by real-world use cases and merchant feedback, making it far more effective than if it had been developed in isolation.

The Growth Strategy: Word of Mouth and Demonstrated Success

Viaweb’s approach to growth was as pragmatic as its approach to product development. Rather than spending heavily on marketing or sales, the company relied primarily on word of mouth and the demonstrated success of its existing merchants.

Each store that Graham and Morris built became a showcase for what Viaweb could do. Satisfied merchants would refer others, creating a virtuous cycle of growth. This organic approach was well-suited to the early days of e-commerce when many businesses were still deciding whether to venture online.

The company also benefited from the growing interest in e-commerce more broadly. As more consumers became comfortable shopping online and more businesses recognized the potential of reaching customers through the internet, the market for Viaweb’s services expanded.

But even as the company grew, it maintained elements of its hands-on approach. New features and capabilities were often developed in response to specific merchant requests, and the company continued to provide a level of support that went beyond what was strictly necessary for the software to function.

This commitment to merchant success—even when it meant doing things that didn’t scale—became a defining characteristic of Viaweb’s culture and a key factor in its competitive advantage.

The Yahoo Acquisition and Legacy

Viaweb’s combination of innovative technology and merchant-focused service proved remarkably successful. By 1998, just three years after its founding, the company had become the leading online store builder, with thousands of merchants using its platform.

This success attracted the attention of Yahoo, which was looking to expand its e-commerce capabilities. In June 1998, Yahoo acquired Viaweb for approximately $49 million in stock, an impressive outcome for a company that had started with just $10,000 in seed funding.

After the acquisition, Viaweb was rebranded as Yahoo Store and continued to grow. At its peak, it would host more than 14,000 online stores, making it the most popular e-commerce platform of its era.

The legacy of Viaweb extends far beyond its direct business impact. The company pioneered the software-as-a-service model that would later become ubiquitous in the technology industry. It demonstrated the power of Lisp as a programming language for web applications. And perhaps most importantly, it showed that sometimes the best way to build a scalable business is to start by doing things that don’t scale.

Lessons for Entrepreneurs

The Viaweb story offers several valuable lessons for entrepreneurs, particularly those building new categories of products or services:

1. Start with manual processes to validate your concept: By building stores for merchants themselves, Graham and Morris were able to prove their concept and refine their product before investing heavily in automation and self-service tools.
2. Use unscalable approaches to overcome initial resistance: When introducing something new, potential customers often need more hand-holding than they will once the concept becomes familiar. Being willing to provide this extra support can be the difference between gaining traction and failing to launch.
3. Learn directly from users: The hands-on approach gave Viaweb’s founders insights that would have been difficult to obtain through more distant forms of market research. This direct knowledge informed product development in ways that created lasting competitive advantages.
4. Balance service and scalability: Rather than viewing manual processes and automation as mutually exclusive, Viaweb found ways to combine high-touch service with increasingly sophisticated self-service tools, creating a model that could grow while maintaining quality.
5. Focus on customer success: By ensuring that each merchant’s store was well-designed and functional, Viaweb created advocates who drove word-of-mouth growth. This focus on customer success proved more effective than traditional marketing would have been.

The Enduring Relevance of Viaweb’s Approach

Decades after Viaweb’s founding, Paul Graham would formalize the philosophy behind the company’s early approach in his influential essay “Do Things That Don’t Scale.” In it, he advises startup founders to embrace unscalable processes in the early stages of their companies—manually recruiting users, providing extraordinary levels of service, and doing things by hand that will eventually be automated.

This advice, drawn in part from his experience with Viaweb, has become a cornerstone of startup wisdom. It recognizes that the path to building something big often begins with small, labor-intensive steps that create the foundation for later growth.

“The most common unscalable thing founders have to do at the start is to recruit users manually,” Graham wrote. “Nearly all startups have to. You can’t wait for users to come to you. You have to go out and get them.”

In Viaweb’s case, this meant not just finding merchants but actively building their stores for them—a level of service that went far beyond what most software companies would consider reasonable. But this unscalable approach was precisely what allowed the company to scale eventually, creating a platform that would help thousands of merchants establish their online presence.

The Human Element in Technology

Perhaps the most profound lesson from Viaweb’s story is the importance of the human element in technology adoption. In 1995, setting up an online store wasn’t just a technical challenge; it was a conceptual leap for many merchants who had operated solely in the physical world for generations.

By personally building stores for these merchants, Graham and Morris weren’t just providing a service; they were guiding businesses through a transformative process. They were translating between the familiar world of retail and the emerging frontier of e-commerce, making the transition less daunting and more accessible.

This human bridge between established practices and new technologies remains relevant today. As artificial intelligence, blockchain, and other advanced technologies enter the mainstream, companies that can provide a human touch—doing things that don’t scale—may find themselves with a significant advantage in driving adoption and building lasting relationships.

Conclusion: The Paradox of Scaling Through Unscalable Means

The story of Viaweb presents us with a paradox: sometimes, the best way to build a scalable business is to start by doing things that don’t scale at all. By manually building online stores for merchants in their early days, Paul Graham and Robert Morris laid the groundwork for what would become a pioneering software-as-a-service company and, eventually, the most popular online store builder of its time.

This approach—hands-on, labor-intensive, and decidedly inefficient in the short term—provided the insights, relationships, and proof points needed to create a truly scalable platform. It’s a reminder that behind many technological innovations lies a period of unscalable, human-centered work that made those innovations possible.

In an age of automation and artificial intelligence, when the pressure to scale quickly is more intense than ever, Viaweb’s story offers a counterintuitive but powerful lesson: sometimes, the path to building something big begins with doing small things extraordinarily well, even if that means doing them by hand, one customer at a time.

As Paul Graham would later reflect, “Sometimes the slow way is the fast way.” For Viaweb, the seemingly slow approach of building stores for merchants by hand proved to be the fastest route to creating a revolutionary platform that would help define the future of e-commerce.