Build durable agents, data pipelines, and background jobs with an SDK
Learn about WorkflowsSince 2008, Hodinkee has become the preeminent destination for watch enthusiasts worldwide. Today, Hodinkee provides an interactive community, editorial content, insurance, and curated shopping for over 20 million annual visitors. Its small engineering team has shepherded the platform through every stage of growth, from passion project to the definitive voice in horology.
For engineering teams whose primary job is building product, every unnecessary infrastructure lever slows progress.
Hodinkee's team learned this firsthand when they moved from Heroku to AWS, seeking lower costs and better reliability. AWS delivered in terms of reliability and nominal cost reduction, but lacked a framework for Hodinkee’s lean development team to quickly deploy applications with standard levels of security and intuitive continuous deployment.
"It was way too much to be managing our infrastructure in AWS," recalls Brayden Sterrett, Head of Technology at Hodinkee. “We needed to pull a lot of levers just to get basic services set up, and even more to start optimizing things. It was just me and one other engineer at the time; it really started to slow us down.”
The question wasn't whether AWS was powerful enough. It was whether that power was worth the added complexity to their operations. On Render, the team found the right level of abstraction without sacrificing the long-term maturity they needed: 56% lower costs compared to Heroku and a platform where any engineer could deploy without deep AWS expertise.
“Heroku kept things relatively simple, but it became far too expensive for what it was,” Sterrett explains. “We reduced our infrastructure costs by about 30% with AWS; however, those savings came with a need to constantly tweak and maintain our infrastructure on a week-to-week basis. This degraded our development team’s velocity. There’s a real cost there. A time cost for my team, an opportunity cost for the company, and ultimately, a cost to our users who depend on us to build a great experience.”
Like many fast-moving teams, Sterrett and his team found that more control often works against you. It took three steps in AWS’s Parameter Store to manage environment variables. Securely accessing databases meant routing through a bastion host. Every configuration that should have been set up by default became an arduous manual process. The team found themselves maintaining infrastructure rather than building for their users.
Ultimately, Hodinkee’s migration to AWS meant the list price of their compute went down, but the savings were outweighed by the cost of complexity. They needed a cloud that combined the efficiency they required with a more intuitive, scalable developer experience.
"We needed a cloud platform that was simple enough to keep our team moving fast, affordable enough to actually reduce our costs, and easy enough to migrate to that we wouldn't have downtime or lose months in the transition."
Render met all three requirements. Native Ruby support meant existing applications could move over without rewrites. Render Blueprints provided a clean way to define infrastructure-as-code without managing Docker configurations for every project. Managed Postgres meant no separate database provider to coordinate.
What stood out most was how quickly things came together. "Projects were extremely easy to take out of Heroku, put into a Render Blueprint, set the environment variables, spin up the server, pg_dump, pg_restore, and we were off to the races," Sterrett recalls. “It usually took less than two hours to move over a service.” The deployment model clicked immediately: git push to build full-stack previews and automatic, zero-downtime deploys for changes merged to staging or production.
Today, Hodinkee runs 100% of its infrastructure on Render. Throughout the entire migration, their applications saw less than 15 minutes of downtime.
By migrating to Render, Hodinkee reduced infrastructure costs by 56% compared to Heroku, delivering on Sterrett’s original goal.
Sterrett also noted an equally meaningful operational shift: "Render is intuitive enough that no matter what level of infrastructure experience you have, you're able to operate. My team works out of Render every day. They're deploying, they're setting environment variables. Everyone can do it."
Today, Hodinkee's engineering team is growing, and simple infrastructure means new engineers can be productive from day one.
"I don’t see any barriers to bringing people onto the team because Render makes things straightforward," Sterrett says. "Many more engineers could easily operate out of the same environment."
Great infrastructure disappears. It doesn't demand attention or require specialized knowledge. It lets teams focus on what they're actually building.
For Hodinkee, that means continuing to grow the world's best destination for watch enthusiasts. For their engineering team, it means spending time on the platform their community loves, not the servers behind it.
As Sterrett puts it: “It’s been refreshing. We're saving money, shipping faster, and actually enjoying the way we work.”