Does a VPN Slow Down Your Internet? (We Tested It)
A VPN typically slows your internet by 5–20% with a premium provider, or 50–80% with a free one. But the real impact depends on your protocol, server distance, and base speed. We break down exactly how much speed, latency, and upload you'll lose, when a VPN can actually make you faster, and how to test the difference yourself on pong.com.
Yes, a VPN slows down your internet. A premium VPN with a nearby server typically costs you 5–20% of your download speed. A free VPN can eat 50–80%. But here's what most articles won't tell you: in some cases, a VPN can actually make your connection faster — specifically when your ISP is throttling certain types of traffic.
The only way to know how much a VPN affects your connection is to test it. Run a speed test on pong.com with your VPN off, then turn it on and test again. The difference tells you exactly what your VPN is costing you — in download speed, upload speed, ping, and jitter. It takes 60 seconds.
Measure your real-world speed, ping, jitter, and bufferbloat. Free, no signup required.
> Run Free Speed TestWhy Does a VPN Slow Down Your Internet?
A VPN does three things that add overhead to every packet you send and receive. Understanding each one helps you minimize the impact.
1. Encryption overhead
Your VPN encrypts every packet before it leaves your device and decrypts every packet that arrives. This takes CPU cycles. On modern hardware with AES-NI acceleration, this adds less than 1ms of latency and negligible speed loss. On older phones or budget routers running VPN clients, encryption can eat 10–30% of your throughput because the processor can't keep up.
2. Server distance (the big one)
Without a VPN, your data travels directly from your device to the destination server. With a VPN, it first travels to the VPN server, then to the destination. If you're in New York connecting to a VPN server in London to reach a website hosted in Virginia, your data is crossing the Atlantic twice for no reason. Server distance is the single biggest cause of VPN speed loss.
This is also why server distance directly impacts your ping. Every 1,000 miles of distance adds roughly 10–15ms of round-trip latency. Connect to a VPN server across the country and you're adding 30–50ms. Connect to one across the ocean and you're adding 80–150ms.
3. Server congestion
VPN servers have limited bandwidth shared among all connected users. When a server is crowded, your speeds drop even if the server is nearby. This is why free VPNs are so slow — hundreds or thousands of users share the same handful of servers. Premium VPNs maintain more servers with fewer users per server, which keeps speeds higher.
How Much Speed Do You Actually Lose? Real Numbers
We compiled speed test data from multiple independent labs and our own testing to show what you can realistically expect. These numbers assume a 500 Mbps base connection — scale proportionally for your actual speed.
| VPN Scenario | Download Speed | Speed Loss | Ping Added |
|---|---|---|---|
| No VPN (baseline) | 500 Mbps | — | — |
| Premium VPN, nearby server | 425–475 Mbps | 5–15% | +2–8ms |
| Premium VPN, same country | 350–450 Mbps | 10–30% | +10–30ms |
| Premium VPN, different continent | 200–350 Mbps | 30–60% | +80–150ms |
| Free VPN | 50–200 Mbps | 50–80% | +30–100ms |
| VPN on router (no AES-NI) | 50–150 Mbps | 70–90% | +5–15ms |
Upload speed is usually hit harder than download because VPN encryption has to process your outgoing data in real time. If your upload speed matters for video calls, streaming, or cloud backups, pay extra attention to upload results when testing your VPN.
VPN Protocols Compared: WireGuard vs OpenVPN vs IKEv2
The VPN protocol you use matters more than which VPN company you choose. A mid-tier VPN on WireGuard often outperforms a premium VPN stuck on OpenVPN. Here's how the three main protocols compare in 2026 testing:
| Protocol | Speed Loss | Latency Added | Best For | Weakness |
|---|---|---|---|---|
| WireGuard | 3–10% | +1–3ms | Everything, especially gaming | Newer, fewer audit years |
| IKEv2/IPsec | 5–15% | +2–5ms | Mobile (fast reconnect) | Blocked by some firewalls |
| OpenVPN UDP | 10–30% | +5–15ms | Maximum compatibility | CPU-heavy, slower |
| OpenVPN TCP | 15–40% | +10–25ms | Bypassing firewalls | Slowest, TCP-over-TCP issues |
WireGuard runs inside the Linux kernel rather than in userspace like OpenVPN, which is why it processes packets with dramatically less overhead. In independent 2026 benchmarks, WireGuard pushes nearly four times the throughput of OpenVPN on the same hardware while adding roughly 1ms of latency compared to OpenVPN's 5–15ms.
When a VPN Actually Makes Your Internet Faster
This sounds counterintuitive, but there are real scenarios where turning on a VPN improves your speed test results. If your speed jumps when you connect to a VPN, it almost certainly means your ISP is throttling your traffic.
ISP throttling (the most common case)
Some ISPs deliberately slow down specific types of traffic — streaming video, gaming, torrents, or even speed tests from certain providers. A VPN encrypts your traffic so your ISP can't tell what type it is, which prevents selective throttling. Users in throttled regions have seen 30–50ms improvements in gaming ping even though the VPN itself adds 15–20ms of overhead, because the throttling reduction was larger than the VPN cost.
Here's how to test for throttling: run a speed test on pong.com without your VPN. Note your speeds. Then connect to a VPN server in the same city and run the test again. If your VPN speed is higher than your no-VPN speed, your ISP is likely throttling. Read our full ISP throttling detection guide for a deeper walkthrough.
Bad ISP routing
Your ISP doesn't always route your traffic on the fastest path. Sometimes packets take detours through congested exchange points. A VPN can route around these bottlenecks if the VPN provider has better peering agreements. This is especially noticeable for gaming — connecting to a VPN server near the game server can sometimes reduce ping by bypassing your ISP's inefficient routing.
Bufferbloat mitigation (indirect)
A VPN doesn't fix bufferbloat directly. But because VPN traffic is encrypted, some routers handle it differently in their queuing, occasionally resulting in lower latency under load. This isn't reliable — the real fix for bufferbloat is SQM on your router — but it explains why some users report better loaded latency with a VPN on.
How a VPN Affects Gaming, Streaming, and Video Calls
Gaming
For gaming, the speed loss barely matters — games use very little bandwidth. What matters is ping and [jitter](/blog/what-is-jitter). A premium VPN on WireGuard with a nearby server adds 2–8ms of ping, which is acceptable for all but competitive esports. But connect to a distant server and you're adding 30–100ms, which makes fast-paced games unplayable. Our reduce ping guide covers this in detail.
Key insight from 2026 gaming tests: ping consistency matters more than the absolute number. A VPN with 45ms average ping and 2ms jitter feels smoother than a raw connection with 40ms ping but 15ms jitter. If your connection has high jitter without a VPN, a nearby VPN server can sometimes smooth things out.
Streaming (Netflix, YouTube, etc.)
4K streaming requires about 25 Mbps. Even with a 30% VPN speed loss, a 100 Mbps connection still delivers 70 Mbps — more than enough. You won't notice VPN-related buffering unless your base speed is already marginal for what you're streaming. If you are getting buffering with a VPN on, the issue is almost certainly server distance or your base speed being too low to absorb the overhead.
Video calls (Zoom, Teams, Meet)
Video calling is more sensitive to VPN overhead than streaming because it requires low latency in both directions. A VPN that adds 20ms of ping is fine. One that adds 80ms will cause noticeable delay in conversation flow. The bigger risk is jitter — VPN tunnels through congested servers cause packet delay variation that makes audio choppy and video freeze. Use a nearby server and WireGuard protocol for calls.
Speed tests
Speed tests are actually the best way to quantify your VPN's impact. Run a test on pong.com without the VPN and save the results, then connect to your VPN and run another. Pong.com measures download, upload, ping, jitter, and bufferbloat — so you'll see the full picture of what your VPN is doing to every dimension of your connection, not just raw speed. Learn how speed tests work to understand exactly what each metric means.
How to Test Your VPN Speed (Step by Step)
Here's the exact process to measure your VPN's impact. The whole thing takes about 2 minutes:
- Disconnect your VPN completely. Make sure it's fully off, not just paused.
- Go to pong.com and run a full speed test. Note your download speed, upload speed, ping, jitter, and bufferbloat grade.
- Connect your VPN to the server you normally use.
- Go back to pong.com and run the test again. Wait 10–15 seconds after connecting for the tunnel to stabilize.
- Compare the results side by side. Calculate the percentage difference for each metric.
- Optional: test a nearby server. If results are bad, switch to the VPN server closest to your physical location and test again. Distance is usually the fix.
Here's what to look for in your results:
| Metric | Acceptable VPN Impact | Problem Threshold | What to Do |
|---|---|---|---|
| Download speed | 5–15% loss | > 30% loss | Switch to closer server |
| Upload speed | 5–20% loss | > 40% loss | Switch protocol to WireGuard |
| Ping | +2–15ms | +50ms or more | Use server in same city/region |
| Jitter | +1–3ms | +10ms or more | Change server, check for congestion |
| Bufferbloat | Same grade | Dropped 2+ grades | Enable SQM on router, change server |
| Packet loss | 0% | > 0.5% | Server is overloaded, switch immediately |
10 Ways to Make Your VPN Faster
- Switch to WireGuard protocol. This alone can cut speed loss from 20–30% down to 5–10%. Check your VPN app's settings for protocol options.
- Connect to the nearest server. Every 1,000 miles adds 10–15ms of latency. If you don't need a specific country's IP, always pick the closest server.
- Use a wired connection. A VPN on WiFi stacks wireless overhead on top of VPN overhead. Ethernet removes one layer of speed loss.
- Avoid running VPN on your router (unless it has AES-NI). Consumer routers often have weak CPUs that bottleneck VPN encryption at 50–150 Mbps. Run the VPN on your device instead.
- Try split tunneling. Most VPN apps let you exclude certain apps or sites from the VPN tunnel. Route only sensitive traffic through the VPN and let everything else go direct.
- Switch servers if one is slow. VPN servers get congested. If your speed dropped, try 2–3 different servers in the same region before blaming the VPN itself.
- Close unnecessary background apps. Cloud syncing, system updates, and background streaming all compete for your VPN tunnel's bandwidth. The same advice from our slow internet guide applies doubly when on a VPN.
- Check your base speed first. If your connection is already slow without a VPN, the VPN will make it worse. Run a speed test without the VPN to establish your baseline. If it's poor, fix that first — our causes of slow internet guide can help.
- Update your VPN app. VPN providers regularly improve their apps with better protocol implementations and server routing. An outdated app may be 10–20% slower than the current version.
- Consider your [DNS settings](/blog/what-is-dns-how-it-affects-speed). Some VPNs use slow DNS servers. If page loads feel sluggish even though speed tests are fine, your VPN's DNS resolution may be the bottleneck.
Free vs Paid VPN: The Speed Difference Is Massive
Free VPNs are the single biggest source of "VPN made my internet slow" complaints. The speed gap between free and paid VPNs is not subtle — it's often 3–5x slower.
| Factor | Premium VPN ($3–12/mo) | Free VPN |
|---|---|---|
| Download speed loss | 5–15% | 50–80% |
| Ping added | 2–15ms | 30–100ms |
| Server count | 3,000–6,000+ | 5–50 |
| Users per server | Low (managed capacity) | High (overloaded) |
| WireGuard support | Yes (most providers) | Rarely |
| Bandwidth cap | Unlimited | 500MB–10GB/month |
| Speed cap | None | Often capped at 1–10 Mbps |
Free VPNs are slow for structural reasons: they have fewer servers, more users per server, older protocols, and many impose hard bandwidth or speed caps. If you've tested your VPN and the speed loss is above 30%, you're likely on a free or bottom-tier plan. Upgrading to a paid VPN with WireGuard support is the single most impactful change you can make.
VPN Speed Loss by Connection Type
Your internet connection type affects how much a VPN slows you down. Faster connections lose a smaller percentage but more raw Mbps. Slower connections lose a larger percentage and feel the impact more.
| Connection Type | Typical Speed | With Premium VPN | Noticeable? |
|---|---|---|---|
| Fiber (1 Gbps) | 900–950 Mbps | 750–900 Mbps | Only in large file transfers |
| Cable (300 Mbps) | 250–300 Mbps | 210–270 Mbps | Barely noticeable |
| DSL (50 Mbps) | 40–50 Mbps | 30–45 Mbps | Noticeable for 4K streaming |
| 5G Home (200 Mbps) | 150–200 Mbps | 120–175 Mbps | Minimal impact |
| Satellite (50 Mbps) | 30–50 Mbps | 15–35 Mbps | Very noticeable, high latency stacks |
Satellite internet deserves a special callout. Satellite connections already have 500–700ms of latency due to the distance to orbit. Adding a VPN on top stacks another 10–50ms plus encryption overhead on hardware that may struggle with it. VPNs on satellite connections are often painfully slow. If you're on satellite, check your internet speed stats to see how your connection compares.
4 VPN Speed Myths, Debunked
Myth 1: "A VPN will halve your speed"
This was true in 2018 with OpenVPN on underpowered hardware. In 2026, premium VPNs on WireGuard lose 5–15% on nearby servers. The idea that VPNs cut your speed in half is outdated by almost a decade.
Myth 2: "More encryption = more speed loss"
AES-256 and ChaCha20 (used by WireGuard) both run at wire speed on modern CPUs with hardware acceleration. The encryption strength has virtually zero impact on speed. The protocol's architecture (kernel vs userspace, connection handling, packet overhead) matters far more.
Myth 3: "VPNs fix slow internet"
A VPN only makes your internet faster if the slowness is caused by ISP throttling. If your internet is slow because of a bad router, WiFi issues, DNS problems, or your ISP delivering less speed than promised, a VPN will make it worse, not better. Always diagnose the root cause of slow internet first.
Myth 4: "The VPN speed test on the VPN provider's site is accurate"
VPN providers often run their own speed tests that route through optimized servers, giving unrealistically good results. Always test with an independent tool. Pong.com tests against real-world infrastructure, not an ISP-optimized or VPN-optimized path — so you get numbers that match your actual experience. Learn how different speed tests compare and why results vary between them.
Frequently Asked Questions
?>Does a VPN slow down WiFi specifically?
?>Does a VPN increase ping for gaming?
?>Which VPN protocol is fastest?
?>Can my ISP see I'm using a VPN?
?>Why is my VPN so slow at night?
?>Does a VPN affect bufferbloat?
?>Should I run the VPN on my router or my device?
?>How do I know if my VPN speed loss is normal?
Bottom Line
A VPN will slow your internet by 5–20% if you use a premium provider with WireGuard and a nearby server. Free VPNs cost you 50–80%. The actual impact depends on server distance, protocol, your base speed, and whether your ISP is throttling you.
The only way to know the real impact on your connection is to test it. Run a speed test on pong.com without the VPN, then with it on. Compare download, upload, ping, jitter, packet loss, and bufferbloat. If the numbers are worse than acceptable, switch to WireGuard, pick a closer server, or upgrade from free to paid. If the numbers are better with the VPN on, your ISP is throttling you — and now you have proof.