David Heinemeier Hansson wrote a post criticizing Google’s track record of killing products. The list on Killed by Google documents approximately 290 terminated projects. I’ve been thinking about this a lot lately.
The Products I Miss
Google Domains, Inbox by Gmail, Google Nexus, Google Reader - services with active, loyal user bases that were shut down anyway. Not because they were failing, but because they stopped fitting into a corporate priority list.
“Google kept Google+ for 8 years. They have money to burn.” So why kill things that work?
The Anxiety
The real problem isn’t any specific shutdown. It’s the uncertainty that comes with building on top of Google services. You can’t predict which features or products will disappear when priorities shift. Migration away from a discontinued service costs real time and effort - especially for hardware solutions or server-dependent applications.
What I’m Doing About It
I’m exploring alternatives and planning to self-host certain applications, even though it means taking on maintenance burden. The trade-off feels worth it.
Google’s current AI push worries me. New experimental projects tend to cannibalize attention from existing tools, and many of those experiments fail within a few years - but not before pulling resources away from things that were actually working.
The lesson: treat any Google service as temporary infrastructure. Plan your exit before you need one.