BH3D Logo
The Entrepreneurial Journey of Clara.io

The Entrepreneurial Journey of Clara.io: Finding Product-Market Fit in Unexpected Places

By Ben Houston, 2024-12-04

The Entrepreneurial Journey of Clara.io0:00 / 0:00

Initializing audio...

This is the story of Clara.io, a online 3d editor we made in 2012 to 2015.

Teaser Image {500x300}

In late 2011 and early 2012, a history of creating successful VFX tools (Deadline and Krakatoa), and more recently my own VFX software tool company Exocortex, I saw an opportunity. The web was transforming how we used software, with applications like Google Docs and Gmail leading a migration from desktop to cloud. As someone deeply familiar with 3D editors and digital content creation tools (DCCs), I could see the tremendous potential in moving these traditionally desktop-bound applications to the web.

The advantages seemed clear. Cloud-based rendering would make powerful computing resources accessible to everyone, even those using low-end devices like Chromebooks. The collaborative nature of VFX work would benefit immensely from a Google Docs-style approach – no more passing zip files back and forth or struggling with merge conflicts. Plus, the endless cycle of downloading, installing, and updating massive software packages could be eliminated with a web-based interface.

The Beginning

In 2012, we made a bold move. Instead of seeking external funding, we leveraged the profits from our existing company, Exocortex, and made creative use of Canadian government programs for entrepreneurs. We cobbled together funding through SR&ED tax credits, a loan from the Business Development Bank of Canada, and grants from IRAP and the Ontario Centers of Excellence. This approach meant we maintained ownership of our vision.

The development journey was challenging but enlightening. A crucial early partnership with Seneca College in Toronto, particularly with visionaries like Catherine Leung & Dawn Mercer, helped us bridge the gap between our desktop development background and the new world of web-based tools. A key hire, Wayne Larson, whom I met at a local Ottawa JavaScript meetup, brought crucial expertise in JavaScript application design.

Exocortex dev team in 2013 {500x300}

After a year of intense development, we launched at SIGGRAPH 2013. Our established reputation in the VFX industry as plugin developers proved invaluable, helping us secure significant media coverage both in traditional outlets and social media. The response was electric – presenting to an auditorium of around 200 people, we got a ton of attention and accolades for delivering a fully functional 3D editor in a web browser. Our partnership with Chaos Group to integrate V-Ray, the leading renderer at the time, added to the excitement.

Following the launch, we secured two angel investors, including Marc Petit, former SVP of Autodesk, which further validated our vision and provided valuable industry expertise.

Looking at the UX design on launch day, still hurts a little bit, its so rough and dirty. But we quickly polished it up to something that I was proud of, especially for 2013/2014:

Clara.io in 2014 {500x300}

Early Success and Challenges

The initial response was overwhelming, with thousands of invite requests in the first week alone and lots of media cover (like this CG Channel article.) We introduced innovative features like a searchable model library and real-time pixel streaming with V-Ray – cutting-edge technology for 2014. Within a year and a half, we had attracted more than 100,000 users. (We ended up around 1.5M registered users by the time we sunset Clara.io in 2022.)

Clara.io Homepage {500x300}

But we faced a significant problem: while students and beginners loved Clara.io, professionals couldn't use it for serious work. Browser memory limitations (around 1GB) and inefficient data structures meant the platform struggled with complex projects. Professional 3D artists, deeply invested in their existing tools and workflows, saw no compelling reason to switch to a platform that couldn't match their current capabilities.

The Unexpected Pivot

Just as we were grappling with these challenges, our backup plan was disappearing. Autodesk announced they were sunsetting Softimage, the platform for which we built most of our desktop plugins. Our bridge back to the VFX world was burning behind us, and we had borrowed significant money to fund Clara.io's development.

Then came an unexpected opportunity. Steelcase, a major US furniture manufacturer, approached us about using Clara.io to visualize their configurable office chairs. This request revealed a crucial insight: for products with numerous configuration options, visualization removes customer doubt and confusion. Our technology could be adapted to store individual components as separate scenes and materials, assembling them in real-time based on user choices.

Visiting Steelcase in Grand Rapids, Michigan {500x300}

The resulting partnership with Steelcase, involving an annual licensing fee, showed us a new direction. While we had set out to create a collaborative 3D editing platform, we discovered a more viable business in enterprise product configuration. This pivot led to the development of Threekit, transforming our journey from a struggle to find paying users into a successful enterprise software story.

Sun-setting Clara.io

7 years after we had pivoted away from the 3D artist focused Clara.io towards the enterprise market with Threekit, we made the hard decision to sunset the Clara.io website at the end of 2022. A shutdown notice was sent out giving everyone time to move off of it. It was discussed here on Hacker News, here on Reddit, also here on Reddit and covered by CG Channel.

The Lesson

Clara.io's story illustrates a crucial lesson in entrepreneurship: sometimes your initial vision isn't where you find success. While we set out to revolutionize 3D editing, we instead found our niche in helping enterprises solve a specific, valuable problem. The technical capabilities we developed for one purpose proved perfectly suited for another, more commercially viable application.