Does a VPN Make the Internet Slower?
Search this question and you’ll mostly get the same vague answer: “a little slower, but it depends.” True, but useless. We’ve actually run six real speed tests across two VPN providers and multiple countries. Instead of guessing, we pulled the real numbers.
The Direct Answer
Yes, a VPN typically makes your internet somewhat slower, because your data takes a longer route through an extra server. How much slower varies enormously. In our own tests, download speeds ranged from 1.9 Mbps on one server to 31.8 Mbps on another, same general setup, VPN Make the Internet Slower wildly different results.
What Our Own Tests Actually Show
Here’s every real download speed we’ve recorded so far, across NordVPN and ExpressVPN servers:
| Server | Provider | Download Speed |
|---|---|---|
| Glasgow, UK | NordVPN | 1.9 Mbps |
| ExpressVPN Manchester, UK | ExpressVPN | 8.4 Mbps |
| ExpressVPN Dallas (Automatic) | ExpressVPN | 8.6 Mbps |
| NordVPN Dallas | NordVPN | 11.2 Mbps |
| ExpressVPN Dallas (Lightway-TCP) | ExpressVPN | 24.9 Mbps |
| NordVPN Seattle | NordVPN | 31.8 Mbps |
That’s not a small spread. The fastest result is over 16 times faster than the slowest. If a VPN applied some fixed “tax” on your speed, every one of these would land in a similar range. They don’t. Which tells you the honest answer isn’t a percentage, it’s a list of factors that matter more than the VPN brand itself.
Our Fastest and Slowest Real Results, Side by Side
Seattle, on NordVPN, gave us the best number we’ve logged: 31.8 Mbps down.
Glasgow, also NordVPN, gave us the worst: 1.9 Mbps down, with ping climbing past 3,000 ms. Same provider. Completely different experience. That gap alone proves the server you land on matters more than which VPN app you’re running.
The Clearest Proof: Same Server, Same Minute, Different Setting
The strongest evidence we have isn’t even about comparing different servers. It’s from testing the exact same ExpressVPN Dallas server twice, minutes apart, changing only the protocol.
On Automatic, the default most people never touch, we got 8.6 Mbps down and a barely-usable 0.9 Mbps up.
Switch to Lightway-TCP, same server, same minute, same phone: 24.9 Mbps down, 13.8 Mbps up. Download nearly tripled. Upload jumped over 15 times. Nothing about the VPN provider changed. Only a setting did. You can read the full before-and-after breakdown for the exact numbers.
What Actually Causes the Slowdown
Based on what we’ve seen across every test:
Distance to the server. Seattle and Glasgow results weren’t close, and physical distance plus network routing explains a lot of that gap.
Protocol choice. The Dallas test above is the clearest proof we have. The right protocol mattered more than anything else we changed.
Server load at that moment. The same city can perform differently an hour later, since you’re sharing that server with other users.
Your own connection. Every test here ran over mobile data, which carries its own variability before the VPN even enters the picture.
Encryption overhead. Real, but smaller than people assume. It’s rarely the biggest factor in our results, distance and protocol consistently mattered more.
What We Haven’t Tested Yet
Needs my input: none of these six tests include a direct side-by-side with the VPN switched off on the same network at the same moment. That’s the one true apples-to-apples number we don’t have yet, a clean “with VPN vs without VPN” comparison on identical conditions. Until that test exists, we can show you the real range VPN servers produce, but not an exact “X% slower than no VPN at all” figure. Worth running next.
So, Does It Matter for You?
If you’re streaming standard video or browsing, even our slowest result would mostly get by, though 1.9 Mbps on Glasgow would struggle with much. If you’re gaming or on video calls, the protocol test above is the one to pay attention to: before assuming your VPN is just “slow,” try switching protocols first. It made a bigger difference than switching servers did.
Quick Answers
Does a VPN really slow down your internet?
Usually somewhat, yes, since traffic takes a longer path through a server. In our real tests, how much varied enormously, from barely noticeable to a significant drop, depending on server and settings.
How much slower does a VPN make your connection?
There’s no fixed number. Our own results ranged from 1.9 Mbps to 31.8 Mbps download across different servers and providers, a 16x difference under broadly similar conditions.
What causes a VPN to slow down more than expected?
In our testing, protocol choice made the single biggest difference, nearly tripling speed on one server just by switching settings. Distance to the server and server load also played large roles.
Does switching VPN protocol actually help speed?
Yes, significantly in our test. Switching from Automatic to Lightway-TCP on the same ExpressVPN server nearly tripled download speed and massively improved upload.
Is VPN encryption the main reason for slower speeds?
Not based on what we’ve seen. Distance, server load, and protocol consistently caused bigger swings in our results than encryption overhead alone.
Related:How Does a VPN Work on a Phone? Simple Real Proof (2026)
