When Google shut down goo.gl in 2018, I didn’t lose a single link. I just never used it.
Back in 2009, colleagues kept saying: “Why build your own? Google made a free URL shortener—reliable, fast, forever.”
The logic seemed solid. Google wasn’t some garage startup. World-class infrastructure, engineering teams, reputation. Why waste time rolling your own?
In April 2018, they shut it down. Not because Google was evil - just a business decision. Maintenance costs money, no monetization path, priorities shifted.
What I saw happen:
Companies scrambling through years of documentation, marketing materials, email campaigns. QR codes on printed collateral leading nowhere. Some just accepted the broken links.
The ironic part? Most migrated to Bitly. Another free public service.
The pattern repeats constantly:
Facebook’s Parse, Google Reader, Docker Hub rate limits, Heroku free tier, certain Bitnami images. The list goes on.
A free service lives exactly as long as it’s profitable for the company to run it. That’s not cynicism—that’s business.
Where’s the line?
For MVPs, proof of concepts, quick validation - free services are perfect. Testing an idea for 3 months? Absolutely use whatever works.
But if those links are in customer documentation, if support depends on them, if recovery takes weeks—you need control.
You don’t have to self-host everything. Could be a paid tier with SLA. Could be your domain on their infrastructure. The key is understanding what happens when the service disappears.
After 15 years in infrastructure, my rule:
Free - for experiments.
Critical - under your control.
Everything between - conscious risk.
You probably know this already. But maybe somewhere in your infrastructure, there’s still a dependency that could vanish tomorrow.
Think for yourself. Even if I’m the one saying it.