About VPN Shelf

In late 2023, my employer sent out a data breach notification. Third-party vendor, names and emails exposed, the usual boilerplate apology. Nothing catastrophic. But it made me actually sit down and look at what I was doing (or not doing) about my own network traffic.

I already had one VPN subscription that I mostly ignored. I downloaded the app, turned it on for a day, got annoyed by the speed hit, and turned it off. Not a serious test. So I started over and paid attention this time. I ran speed tests on the same servers every day for two weeks. I compared prices versus what was actually offered. I cancelled one subscription after realizing the "no-logs" policy was written in a way that probably didn't mean what I thought it meant. Signed up for three more. Kept a spreadsheet of results. My partner pointed out the spreadsheet and asked if this was normal. I said it was for research.

That was about two years ago. I've paid for and tested over a dozen services since then, and I still run comparisons on a recurring basis because providers change things without announcing them. What I do here is document what I actually find: real speed numbers from Seattle, billing gotchas, support experience, what each provider gets right and what it gets quietly wrong.

I'm not a cybersecurity professional. No formal networking or privacy training. I'm a software engineer who reads too much about threat models and prefers to settle product debates with measurements rather than marketing copy. If you want an academic analysis of VPN protocols, this isn't it. If you want to know which service held up over six months of daily use and which one I quietly unsubscribed from, that's what this site is for.

More on testing methodology on the author page.

This site uses affiliate links. If you buy a VPN through these links, I earn a commission at no extra cost to you. I only recommend services I have actually paid for and tested myself. For how affiliate relationships work here, see the editorial policy.