📢 Transparency Note: We may earn commissions from links on this page. This doesn't affect our ratings or recommendations. Learn more
Home / Guides / Speed Test Methodology

VPN Speed Test Methodology

How we test VPN speeds, what metrics actually matter, and how to run meaningful speed tests yourself. Cutting through the marketing noise.

Updated: January 2025 8 min read

Quick Answer: What Speed Loss Is Normal?

A good VPN using WireGuard should reduce speeds by 10-20% on local servers. Expect 30-50% loss on distant servers. Anything over 50% loss suggests a problem with your VPN, server, or configuration.

In This Guide

Why VPNs Affect Speed

When you connect to a VPN, your data takes a longer journey and undergoes additional processing. Understanding why helps you set realistic expectations and optimize your setup.

1. Encryption Overhead

Every packet of data must be encrypted before leaving your device and decrypted at the VPN server. Modern protocols (WireGuard) minimize this overhead, but it still exists. On fast connections (500+ Mbps), encryption can become the bottleneck.

2. Server Distance

Data travels at the speed of light, but it still takes time. Connecting to a server 5,000 miles away adds latency—the time it takes for data to make the round trip. This is most noticeable in gaming and video calls.

3. Server Load

VPN servers have finite capacity. A server handling thousands of simultaneous connections will perform worse than a lightly-loaded one. Good VPNs show server load and let you choose less crowded options.

4. Protocol Choice

Different VPN protocols have vastly different performance. WireGuard is typically 2-3x faster than OpenVPN. Using the wrong protocol for your situation can dramatically reduce speeds.

How VPN Routing Works

💻
Your Device
Encrypts data
🔐
VPN Server
Decrypts & re-routes
🌐
Destination
Netflix, website, etc.

Without VPN: Your Device → Destination (direct)
With VPN: Your Device → VPN Server → Destination (extra hop + encryption)

Key Speed Metrics Explained

⬇️

Download Speed (Mbps)

How fast data arrives at your device. This is what most people care about—it determines how quickly websites load, videos buffer, and files download.

What you need:
  • 5-10 Mbps: Standard definition streaming
  • 25 Mbps: HD streaming (1080p)
  • 50+ Mbps: 4K streaming, fast downloads
  • 100+ Mbps: Multiple 4K streams, large file transfers
⬆️

Upload Speed (Mbps)

How fast data leaves your device. Important for video calls, streaming your gameplay, or uploading files to the cloud.

What you need:
  • 3 Mbps: Video calls (Zoom, Teams)
  • 5-10 Mbps: HD video calls, streaming to Twitch
  • 20+ Mbps: 4K streaming, fast cloud backups
⏱️

Latency / Ping (ms)

The time (in milliseconds) for data to make a round trip. Critical for gaming, video calls, and any real-time interaction. Lower is better.

What's acceptable:
  • <30ms: Excellent (competitive gaming)
  • 30-60ms: Good (casual gaming, video calls)
  • 60-100ms: Acceptable (most activities)
  • >100ms: Noticeable lag in real-time activities
📊

Jitter (ms)

Variation in latency over time. High jitter means inconsistent performance—your ping might jump from 30ms to 150ms randomly. This causes stuttering in video calls and rubber-banding in games.

What's acceptable:
  • <10ms: Excellent (stable connection)
  • 10-30ms: Good (minor fluctuations)
  • >30ms: May cause issues in real-time apps

Our Testing Methodology

We test VPN speeds under controlled, repeatable conditions. Here's exactly how we do it.

Test Environment

Base Connection
1 Gbps fiber (symmetric)
Test Device
Windows 11 PC (wired Ethernet)
Speed Test Tool
Ookla Speedtest CLI + iPerf3
Location
United States (East Coast)
1

Baseline Measurement

We run 10 speed tests without VPN to establish our baseline. Results are averaged to account for normal network fluctuations.

2

Protocol Testing

Each VPN is tested on WireGuard (or equivalent: NordLynx, Lightway) and OpenVPN. We use default settings to reflect typical user experience.

3

Server Location Testing

We test the same three server categories for each VPN: nearest server, US West Coast (cross-country), and UK server (international).

4

Multiple Test Runs

Each configuration is tested 5 times across different times of day (morning, afternoon, evening) over 3 days. Results are averaged.

5

Speed Loss Calculation

We calculate speed loss as percentage of baseline: (Baseline - VPN Speed) / Baseline × 100. Lower percentage = faster VPN.

⚠️ Important Caveats

  • Your results will differ. Your base internet speed, location, and ISP all affect VPN performance.
  • Server performance varies. The exact server you connect to matters—VPNs have thousands of servers with different loads.
  • Time of day matters. Servers are slower during peak hours (evenings, weekends).
  • We can't test every server. Results represent typical performance, not every possible scenario.

Latest Speed Test Results

Results from our January 2025 testing cycle. Baseline connection: 940 Mbps down / 940 Mbps up.

VPN Protocol Local Server Cross-Country UK Server Avg. Loss
NordVPN NordLynx 812 Mbps 654 Mbps 498 Mbps -30%
NordVPN OpenVPN 312 Mbps 278 Mbps 215 Mbps -71%
Surfshark WireGuard 798 Mbps 612 Mbps 467 Mbps -33%
Surfshark OpenVPN 298 Mbps 256 Mbps 198 Mbps -73%
ExpressVPN Lightway 756 Mbps 598 Mbps 445 Mbps -36%
Mullvad WireGuard 834 Mbps 678 Mbps 512 Mbps -28%
PIA WireGuard 712 Mbps 534 Mbps 389 Mbps -42%

Key Findings

  • WireGuard is dramatically faster than OpenVPN. Average speed loss dropped from 72% (OpenVPN) to 34% (WireGuard). Always use WireGuard when available.
  • Mullvad edges out competitors on raw speed. Despite having fewer servers, their WireGuard implementation is exceptionally fast.
  • NordVPN and Surfshark are nearly identical. Both owned by Nord Security, they use similar infrastructure and deliver comparable speeds.
  • ! International connections take the biggest hit. Connecting to UK from US East Coast costs ~45% of speed even with WireGuard.

How to Test Your VPN Speed

Our results are a starting point—your mileage will vary. Here's how to run meaningful speed tests yourself.

Step 1: Establish Your Baseline

Disconnect from VPN and run three speed tests at speedtest.net. Average the results. This is your maximum possible speed—no VPN can exceed it.

💡 Use a wired Ethernet connection if possible. WiFi adds variability that makes comparisons harder.

Step 2: Test With VPN Connected

Connect to your VPN (use WireGuard protocol if available). Run three more speed tests and average the results. Calculate speed loss:

Speed Loss = (Baseline - VPN Speed) / Baseline × 100

Step 3: Test Different Servers

Try at least three server locations: your nearest server, a server in another region, and a server in another country. Performance varies significantly by server.

Step 4: Test Different Times

Run tests at different times of day. Evening hours (6-10 PM local time) typically show slower speeds due to higher server load. Morning tests often show peak performance.

How to Improve VPN Speed

1. Switch to WireGuard

If your VPN app defaults to OpenVPN, switch to WireGuard (or NordLynx for NordVPN, Lightway for ExpressVPN). This alone can double your speeds.

2. Connect to Nearest Server

Don't use "auto-connect" if it chooses distant servers. Manually select the server closest to your physical location for best speeds.

3. Avoid Overloaded Servers

NordVPN and Surfshark show server load percentages. Choose servers with lower load (<50%) for better performance.

4. Use Wired Connection

Ethernet is faster and more stable than WiFi. If your VPN feels slow on WiFi, try a wired connection before blaming the VPN.

5. Restart Your Connection

Sometimes you land on a slow server. Disconnect and reconnect to be assigned a different server. Try a few times if speeds are poor.

6. Disable Extra Features

Features like Double VPN, Onion over VPN, or MultiHop add security but reduce speed. Use standard servers for everyday browsing.

Frequently Asked Questions

Can a VPN ever increase my speed?

Rarely, but yes. If your ISP throttles certain traffic (like streaming or torrents), a VPN can bypass this throttling by hiding what you're doing. Some users see faster Netflix speeds with a VPN for this reason.

Why are my VPN speeds much slower than your tests?

Your base internet speed is the limiting factor. If you have 100 Mbps internet, you can't get faster than that with a VPN. Also check: Are you using WireGuard? Is your server nearby? Are you on WiFi?

Is ping more important than download speed for gaming?

Yes. Games need very little bandwidth but require fast response times. A connection with 50 Mbps download and 30ms ping will game better than 500 Mbps with 100ms ping. Prioritize nearby servers for gaming.

Do "optimized servers" actually help?

Sometimes. VPNs label servers as "optimized for streaming" or "P2P optimized." These often have more bandwidth allocated or better routing. Worth trying if you're having issues with specific activities.

The Fastest VPNs of 2025

WireGuard changed everything. Modern VPNs now deliver speeds that were impossible just a few years ago.

Related Guides