Does a VPN Slow Down Your Internet?

Does a VPN slow down your internet? Why it happens at all

To understand why a VPN touches your speed, it helps to picture what a VPN actually does to your traffic. Normally, when you load a website or stream a video, your device sends data in a fairly direct path to your internet service provider (ISP), then out to the wider internet, then back again. A VPN inserts itself into that path in two specific ways, and both of them cost a small amount of time and processing power.

First, encryption. Before your data leaves your device, the VPN app wraps it in a layer of encryption so that nobody sitting between you and the destination — including your ISP — can read what you are sending or receiving. Encrypting and later decrypting that data is real computational work. Your device (or, on the other end, the VPN server) has to run mathematical operations on every packet. Modern phones, laptops, and routers have dedicated hardware that makes this fast, but it is never literally zero-cost.

Second, rerouting. Instead of your traffic going straight to its destination, it first travels to a VPN server — which could be in your own country or on the other side of the world, depending on which one you pick — gets decrypted there, and only then continues on to the website or app you are actually trying to reach. That extra hop adds physical distance, and distance adds time, because data (even moving near the speed of light through fiber-optic cable) still has to physically travel further and pass through more network equipment along the way.

Put those two things together — encryption overhead plus an extra stop along the route — and you get the reason a VPN connection is, in principle, always going to be at least a fraction slower than no VPN at all under otherwise identical conditions. The real question most people actually care about is not whether there is a slowdown, but whether it is large enough to notice in daily use. For most well-built modern VPN apps on a reasonable connection, the honest answer is: usually not.

How much slower is browsing and streaming with a VPN, realistically?

This is the part where it is tempting to throw out a single number — "a VPN slows you down by X%" — but that number would not be honest, because it depends entirely on your specific combination of connection, distance, server load, protocol, and device. What is fair to say is how the *type* of slowdown usually shows up in everyday use.

For most people on a broadband or modern mobile connection, connecting to a nearby, uncongested VPN server using a current protocol produces a change in speed that is genuinely hard to notice while browsing, checking email, using social media, or streaming standard or even high-definition video. The connection still comfortably clears the bandwidth those activities need.

Where a slowdown becomes more noticeable is at the edges: if you are already on a slow or unstable connection with little headroom to begin with, if you are doing something bandwidth-hungry like downloading very large files or streaming in 4K, if you connect to a VPN server that is very far away or overloaded with other users, or if you are using an older device that struggles with the encryption workload. In those situations, the added VPN overhead is competing for a resource — bandwidth, or processing power — that was already tight, so the effect is easier to feel.

The practical takeaway is this: the question is less "does a VPN slow you down" and more "how much headroom do you have, and how far is the server you picked." A fast connection with a nearby server and a modern protocol usually hides the slowdown almost entirely. A slow connection combined with a distant, congested server is where people notice buffering or lag.

Which factors cause the biggest VPN slowdown?

Not all sources of VPN overhead are equal. If you are trying to figure out where your own slowdown is coming from, these are the factors worth checking, roughly in order of how much they typically matter:

Server distance

This is usually the single biggest factor. The physical (and network) distance between you and the VPN server determines your baseline latency — how long it takes a signal to make the round trip. Connecting to a server in a nearby country will almost always feel faster than connecting to one on another continent, because you are adding less travel distance to every single request your device makes.

Server load

A VPN server, like any server, has finite capacity. If a lot of other users are connected to the exact same server at the same time, everyone sharing it gets a smaller slice of its available bandwidth and processing power. This is why the same VPN server can feel fast at 6 a.m. and noticeably slower during a provider's peak evening hours.

Protocol choice

The underlying VPN protocol — the technical rulebook the app uses to establish and maintain the encrypted tunnel — has a real effect on speed. Newer protocols are generally leaner and faster than older ones. This is covered in more depth below, but it is worth knowing that switching protocols inside your VPN app's settings, if it offers that option, is one of the few speed dials that is entirely in your control.

Encryption strength

Stronger encryption does mean more computational work, but this factor matters far less than people assume on modern hardware. Today's phones and computers have hardware acceleration built in specifically for the encryption standards VPNs use, so this is rarely the bottleneck it once was on older devices.

Your own baseline connection and device

A VPN cannot make a slow connection fast, and it cannot outrun an underpowered device. If your starting point already has very little headroom, any added overhead is more noticeable, purely because there was less room to absorb it.

Wi-Fi vs. wired, and network congestion

This one is not really about the VPN at all, but it gets blamed on the VPN constantly. A shaky Wi-Fi signal, an overloaded home network with many devices, or general congestion on your ISP's side can all produce symptoms — buffering, lag, slow page loads — that look identical to VPN slowdown but have nothing to do with it. Before blaming your VPN, it is worth testing your connection without it to see if the same issue shows up.

Does server distance matter more than encryption?

In practice, yes — for most everyday use, the distance to your VPN server tends to have a bigger and more consistent effect on perceived speed than the strength of the encryption itself. This surprises a lot of people, because encryption is the part of a VPN that gets talked about the most, while server location is often an afterthought — just whichever option happened to be selected by default when the app opened.

The reason distance dominates comes down to basic physics and networking: every additional stretch of cable, every extra router or exchange point your data has to pass through, adds a small amount of delay (latency). Add those delays up across a long international hop, and the round-trip time for each request to your VPN server and back can become substantial — especially for anything that depends on many quick back-and-forth exchanges, like loading a complex webpage or playing an online game.

Encryption overhead, by contrast, is largely a fixed, small cost per packet that modern devices process very quickly thanks to built-in hardware acceleration. It does not scale up the way distance-related latency does.

The practical implication is simple and useful: if you want the fastest possible VPN experience for everyday tasks and you do not have a specific reason to connect to a server in a particular country (for example, to access region-specific content), pick the server closest to your actual physical location. That single choice usually does more for your perceived speed than any other setting in the app.

Can your choice of VPN protocol change your speed?

Yes, meaningfully so. The protocol is the set of technical rules a VPN uses to build and maintain its encrypted tunnel, and different protocols were designed with different priorities — some favor maximum compatibility and battle-tested security, others were built from the ground up with speed and modern efficiency in mind. If your VPN app lets you choose a protocol in its settings, this is genuinely worth understanding.

WireGuard (and provider variants of it)

WireGuard is a newer, leaner protocol built with a much smaller and simpler codebase than its predecessors. That simplicity translates into real-world speed advantages: less overhead per packet, faster connection setup, and generally the best throughput of the commonly available options. Several VPN providers, including ones on this site, ship their own tuned version of WireGuard under a proprietary name — NordVPN's is called NordLynx, for example — which is functionally WireGuard with provider-specific adjustments layered on top. If your VPN app defaults to or offers a WireGuard-based protocol, it is usually the fastest choice available.

OpenVPN

OpenVPN has been the industry workhorse for years — extremely well tested, widely trusted, and highly configurable. It is generally a bit heavier and slower than WireGuard, particularly over its TCP mode, though its UDP mode narrows that gap. Many people still choose OpenVPN specifically because of its long security track record, accepting a modest speed trade-off in exchange.

IKEv2/IPsec

IKEv2 is known for being fast to reconnect after a dropped connection, which makes it a common default on mobile devices that frequently switch between Wi-Fi and cellular data. Its everyday throughput sits in a similar range to OpenVPN for most users.

What this means for you

If speed is your priority and your VPN app offers a choice, a WireGuard-based protocol is the sensible default to try first. If it is not available or the app auto-selects a protocol for you, that is also fine for the vast majority of use cases — most reputable VPN apps today are tuned to pick a reasonable, fast default automatically rather than leaving users to hunt through a settings menu.

Can a VPN actually make your internet feel faster?

It sounds counterintuitive, but yes — there are specific, real situations where turning a VPN on can make a connection feel faster rather than slower, even though the VPN itself is technically adding overhead. The key phrase is "feel faster," because what changes is not the raw physics of your connection, but what your ISP is doing to your traffic before the VPN gets involved.

Some internet providers apply what is known as traffic shaping or throttling — deliberately slowing down specific types of traffic, such as video streaming or peer-to-peer file transfers, often during busy network hours, in order to manage overall network load. Because a VPN encrypts your traffic, your ISP can generally no longer see what kind of traffic it is looking at — it just sees an encrypted stream heading to a VPN server. If your slow streaming or downloads were caused by your ISP specifically targeting that kind of traffic, wrapping it in a VPN can restore full speed by making it invisible to whatever throttling rule was targeting it.

This is not a guarantee — not every ISP throttles, and not every slowdown you experience is caused by throttling in the first place. General network congestion, an oversubscribed local node, or simple distance to the content server can all cause slow streaming with no throttling involved at all, and a VPN will not fix those causes. But if you have noticed that a specific type of activity (streaming in particular) is consistently slower than the rest of your internet use, testing it with a VPN on is a reasonable, low-effort way to see whether traffic shaping was the culprit.

How can you test whether your VPN is actually slowing you down?

Rather than guessing, you can measure this directly with a simple before-and-after comparison, and it only takes a couple of minutes.

Why a single test is not enough

Internet speed naturally fluctuates throughout the day because of local network congestion, how many devices in your home are active, and how busy your ISP's network is in your area at that hour. A single speed test taken once with the VPN off and once with it on can be misleading if the two tests happened to land in different congestion conditions. For a more reliable read, run each test two or three times a few minutes apart and compare the averages, ideally at a similar time of day.

Step 1: Test your baseline speed with the VPN off

Close the VPN app completely, wait a few seconds for your connection to settle, then run a speed test using any reputable browser-based speed test tool. Note the download speed, upload speed, and ping (latency) it reports.

Step 2: Connect to your VPN and test again

Turn the VPN on, connect to the server you would normally use, and run the exact same speed test again under the same conditions — same device, same Wi-Fi or wired connection, roughly the same time of day.

Step 3: Compare, then adjust one variable at a time

Compare the two sets of results. If the difference is small, that is a normal, expected amount of VPN overhead and nothing to worry about. If the difference is large, try changing one variable at a time to isolate the cause: switch to a nearer server, try a different protocol if your app offers a choice, or try a different server in the same country in case the one you picked happened to be congested. Re-test after each change so you know which adjustment actually helped.

A note on what "normal" looks like

It is normal for latency (ping) to increase somewhat with a VPN on, since your traffic is taking a longer physical path. It is normal for download and upload speeds to dip modestly. What is not normal is a connection becoming unusable — constant buffering, pages that will not load, or speeds dropping to a small fraction of your baseline. If you see that, something specific is wrong (an overloaded server, a poor protocol choice for your situation, or a local network issue), and it is worth troubleshooting rather than just living with it.

Do extra VPN features like a kill switch or double VPN slow things down further?

Most of the everyday safety features built into a modern VPN app have little to no measurable effect on speed, but a few specific ones are worth understanding because they genuinely do add overhead by design.

Kill switch

A kill switch blocks your device from sending any traffic outside the VPN tunnel if the VPN connection unexpectedly drops, so you never accidentally leak unprotected traffic. It sits there passively monitoring the connection and only acts if something goes wrong, so it has no meaningful ongoing effect on your speed while the VPN is connected normally.

DNS leak protection

This routes your DNS lookups (the process of translating a website name into its actual server address) through the VPN's own DNS servers instead of your ISP's. The effect on speed is generally negligible, though in rare cases a VPN provider's DNS servers may respond slightly slower than your ISP's if they are geographically farther from you — usually not something you would notice in practice.

Double VPN / multi-hop

This is the one feature on this list that has a real, by-design speed cost. Double VPN (sometimes called multi-hop) routes your traffic through two separate VPN servers instead of one, encrypting it twice, for an extra layer of privacy that makes it harder to trace traffic back to you even if one server were somehow compromised. Because your data is now traveling a longer physical path and being encrypted and decrypted twice, this feature reliably produces a bigger speed reduction than a standard single-server connection. It is a deliberate trade-off for a specific, higher privacy use case, not something to leave switched on by default if speed matters more to you day to day.

Ad and tracker blocking

Some VPN apps include a built-in feature that blocks ads, trackers, or known malicious domains at the network level before pages even finish loading. Because this can mean fewer requests and less data actually being downloaded per page, it sometimes makes browsing feel faster overall, rather than slower, even though the VPN itself is still adding its usual small amount of overhead underneath.

The practical guidance here is straightforward: leave a kill switch and DNS leak protection on by default, since their speed cost is effectively zero and the safety benefit is real. Reserve double VPN for situations where you specifically want that extra layer of protection and are willing to accept a bigger, noticeable speed trade-off in exchange.

What can you actually do to reduce VPN slowdown?

If your testing shows a bigger gap than you would like, there are several concrete, practical things you can try, roughly in order of how much impact they tend to have.

Pick a nearby server

As covered above, this is usually the highest-impact change available to you. Unless you specifically need a server in another country (for accessing region-locked content, for example), choose the closest available location to your physical location.

Avoid obviously congested servers

Some VPN apps show a live load indicator or a recommended/"lightest load" server option. If yours does, use it. If it does not, simply avoid servers you already suspect are popular default choices, and try an alternative location in the same country if one is available.

Switch protocols if you have the option

If your VPN app lets you manually choose a protocol, try a WireGuard-based option first, since it is generally the fastest widely available choice today.

Use a wired connection instead of Wi-Fi when it matters

Because Wi-Fi has its own separate sources of speed loss — signal strength, interference, distance from the router — plugging in via ethernet removes an entire category of variables and makes it much easier to see what the VPN itself is actually doing to your speed.

Check for background apps competing for bandwidth

Cloud backups, automatic software updates, and other devices on your network downloading large files in the background can all eat into the bandwidth available to whatever you are actively trying to do, and the effect can easily be mistaken for VPN overhead.

Restart the VPN app or your router

This sounds almost too simple, but a stale connection to a specific server, or a router that has been running for a long time without a restart, can quietly degrade performance in ways that a fresh connection resolves immediately.

Consider split tunneling if your app supports it

Split tunneling lets you choose which specific apps route through the VPN and which connect directly to the internet as normal. If you only need VPN protection for certain activities, routing everything else outside the tunnel means those apps are not affected by VPN overhead at all.

Does a VPN affect gaming and video calls more than browsing?

Yes, and the reason is not really about raw speed at all — it is about latency and consistency, which matter far more for real-time activities than they do for loading a webpage or streaming a pre-buffered video.

When you load a webpage, your device requests the page once, downloads it, and displays it — a slight delay of a fraction of a second is rarely something you would consciously register. Streaming video works similarly: the player downloads a buffer of content slightly ahead of what you are watching, so small, brief fluctuations in speed rarely cause visible interruption.

Online gaming and live video calls are different. Both depend on a continuous, rapid back-and-forth exchange of small packets of data, and both are far more sensitive to two things a VPN can add: latency (the delay before a packet arrives) and jitter (how much that delay varies from one packet to the next). A VPN that adds even a modest amount of extra latency, because your traffic is now routing through a server that may be out of the way of the most direct path, can be felt directly as input lag in a game or as a slight delay or stutter in a call. Jitter is often the more noticeable culprit in video calls specifically — it is what causes audio to briefly cut out or video to freeze for a moment even when your download speed test looks perfectly fine.

If gaming or video calls are a priority, the same advice from earlier applies with extra force: pick the closest possible server, use a wired connection over Wi-Fi if you can, and consider a WireGuard-based protocol, since its efficiency tends to produce more consistent latency, not just a higher peak speed. Split tunneling is also particularly useful here — routing your game or call app outside the VPN tunnel entirely, while keeping other traffic protected, removes the added latency from the activity that is most sensitive to it.

Do VPNs slow down phones more than computers?

Not inherently, but there are a few real differences worth knowing if most of your VPN use happens on a phone rather than a desktop or laptop.

Modern smartphones have hardware-accelerated encryption built into their chipsets, just like modern computers do, so the raw computational cost of running a VPN is not usually the limiting factor on either platform. What differs more is the network environment a phone typically operates in. Phones move between Wi-Fi networks and cellular data constantly, and cellular connections — particularly on weaker signal or older network generations — already have more variable baseline latency and speed than a stable home broadband or ethernet connection. Add VPN overhead on top of an already-variable cellular connection, and the combined effect is more noticeable than the same VPN overhead added to a strong, stable Wi-Fi or wired connection.

Battery life is the other practical difference. Running a VPN continuously on a phone does use some additional battery, since the device is doing extra encryption work and, in some cases, keeping a more persistent background connection alive. This is rarely dramatic with a well-built VPN app, but it is worth knowing if you notice a mobile VPN app draining battery faster than expected — switching to a more efficient protocol like WireGuard, where the app offers the choice, generally helps on this front too, since it was designed in part to be lighter on both CPU and battery than older protocols.

The practical upshot: if a VPN feels slower on your phone than it does on your laptop at home, the cause is more likely to be the cellular network itself, or a weak Wi-Fi signal, than anything different about how the VPN works on mobile.

Is a small speed loss a fair trade-off for using a VPN?

This really comes down to what you are using a VPN for, and it is worth being honest about the trade-off rather than pretending it does not exist. A VPN encrypts your traffic and hides your browsing from your ISP and from anyone else monitoring the network you are connected to, which is genuinely valuable on public Wi-Fi, when traveling, or simply as a baseline privacy habit. That protection has a real, if usually small, cost in the form of some added latency and a modest reduction in top-line speed.

For most everyday activities — browsing, messaging, social media, video calls, and standard streaming — a well-configured VPN on a decent connection produces a difference that is genuinely hard to notice in practice. For a narrower set of activities where every bit of bandwidth or the lowest possible latency matters most — competitive online gaming, very large file transfers, or working with a connection that was already close to its limits — the trade-off is more likely to be noticeable, and it becomes a more personal decision about whether the privacy and security benefit is worth it for that specific activity.

A reasonable middle ground many people land on is using split tunneling (where available) or simply toggling the VPN off for the specific tasks where speed matters most, while keeping it on as the default for everything else. That way you are not forced into an all-or-nothing choice.

Do free VPNs slow you down more than paid ones?

Generally, yes, and the reason is straightforward economics rather than anything mysterious. Running a global network of fast, well-maintained VPN servers costs real money — bandwidth, hardware, and staff to keep it all running. Paid VPN services fund that infrastructure directly through subscriptions. Free VPN services have to fund it some other way, and that "other way" very often means a smaller server network shared by a much larger number of free users, deliberately capped speeds to encourage upgrading to a paid tier, or business models that raise their own separate privacy questions entirely, which is a topic worth researching on its own before choosing a free option.

Among the four providers covered on this site, the practical differences in day-to-day speed come down to the same factors discussed throughout this article: server network size and spread, how modern the protocols on offer are, and how congested a given server tends to be. NordVPN and Proton VPN are both generally positioned around strong performance in our own editorial assessment, each with a large, actively maintained server network and modern protocol support (NordVPN's NordLynx and Proton VPN's own WireGuard implementation among them) — Proton VPN also happens to offer a genuine free tier, though free-tier server selection is typically more limited than the paid plans across the industry. PureVPN and FastestVPN both lean toward value-oriented positioning with broad device support; in our own editorial assessment their everyday speed is solid for typical browsing and streaming, if generally a step behind the fastest options in the category. As with every claim about pricing or ratings on this site, we deliberately do not publish specific speed numbers or star ratings here, since real-world results vary by location, server, and time of day — the accurate way to compare is to check each provider's own current server list and, where offered, try their money-back guarantee window on your own connection before committing.

The bottom line on VPNs and speed

A VPN slowing down your internet is not a myth, and it is not a sign that something is broken — it is simply the expected cost of adding encryption and an extra routing hop to your traffic. For the overwhelming majority of everyday use, on a reasonable connection, with a nearby server and a modern protocol, that cost is small enough to live comfortably in the background of normal browsing, messaging, and streaming.

Where it becomes worth actively managing is at the extremes: already-slow connections with little headroom, latency-sensitive activities like gaming and video calls, very large file transfers, or specific privacy features like double VPN that trade speed for an extra layer of protection on purpose. In every one of those cases, the fixes are the same handful of levers covered throughout this article — server distance, server load, protocol choice, wired versus Wi-Fi, and which extra features you actually need turned on.

The most useful habit is simply testing your own setup rather than assuming. Run a quick speed check with the VPN off, then on, using your actual server and protocol. If the gap is small, you now have real confidence instead of a guess. If it is larger than expected, you have a specific, fixable starting point rather than a vague sense that "VPNs are slow."