You have a fast broadband plan, yet a video call still stutters or an online game feels sluggish. The culprit is often not a lack of speed but latency, the hidden delay that shapes how responsive your connection feels. It is one of the most important and least understood measures in networking.
Here is what latency is and why it matters.
What it is
Latency is the delay between sending a request and the first part of the response arriving, usually measured in milliseconds, written as ms.
In plain terms, it is the lag in a connection. When you click a link, send a message or move in an online game, your request travels across the network to a server and the reply travels back. Latency measures the round trip of that conversation. The lower the number, the snappier everything feels.
You may also see latency described as ping, the name of a tool that measures it. A "low ping" simply means low latency, which is why gamers prize it.
Latency is not bandwidth
The single most useful thing to understand is that latency and bandwidth measure different things, and confusing them leads to a lot of frustration.
- Bandwidth is how much data can flow at once, the width of the pipe.
- Latency is the delay before data starts arriving, the wait before anything comes through at all.
A classic analogy is post. Imagine sending a thousand letters by lorry versus a single urgent note by motorbike courier. The lorry has huge capacity, lots of bandwidth, but takes a long time to arrive, high latency. The motorbike carries almost nothing yet arrives in minutes, low latency. For a quick reply, you want the courier; for moving a mountain of post, you want the lorry.
This is why a connection with plenty of bandwidth can still feel laggy. If each request takes a long time to make the round trip, the experience is poor no matter how wide the pipe. For the other side of this story, see our guide to bandwidth.
What causes latency
Latency builds up from several sources along the path your data takes. The main contributors are:
- Distance. Data cannot travel faster than the speed of light, and in cables it is slower still. A server on the other side of the world is inherently further away in time than one nearby. This physical limit cannot be removed, only reduced by using closer servers.
- Network hops. Data rarely travels in a straight line. It passes through many pieces of equipment, called routers, each adding a tiny delay as it reads the destination and forwards the data on.
- Congestion. When a network is busy, data can queue, much like traffic at a junction. Peak evening hours often bring higher latency as everyone competes for the same routes.
- The connection type. Different technologies add different delays. Satellite links, for example, have high latency because signals travel vast distances up and back, even though their bandwidth can be good.
- The server and your device. The machine answering your request takes time to process it, and your own device takes time to send and receive. Both add to the total.
Every one of these adds milliseconds, and they stack up across the round trip.
Why latency matters
Whether latency matters depends entirely on what you are doing. For some tasks it is barely noticeable; for others it makes or breaks the experience.
Latency matters most for real-time activities, where a back-and-forth happens constantly:
- Video and voice calls. High latency causes awkward pauses, people talking over each other and audio that drifts out of sync.
- Online gaming. Fast games depend on instant responses. High latency, felt as lag, means your actions register late and the game feels unfair.
- Browsing. Loading a single web page can involve dozens of separate requests. High latency on each one adds up, making sites feel slow to respond even on a fast connection.
Latency matters far less for tasks that move a lot of data in one go, such as downloading a large file or streaming a film once it has buffered. There, bandwidth is the bigger factor, because a short initial delay barely registers across a long transfer.
A useful rule of thumb: bandwidth decides how fast you can move a lot of data, while latency decides how responsive everything feels moment to moment.
What counts as good latency?
There is no universal threshold, but lower is always better. As a rough guide, latency under about 50 milliseconds feels responsive for everyday use, and competitive gamers chase figures lower still. Once latency climbs into the hundreds of milliseconds, delays become noticeable in calls and games.
You can get a sense of your own latency with a simple online speed test, which usually reports a ping figure alongside your download and upload speeds. If that number is high even when nothing else is using the connection, the cause may be your connection type, your distance from servers, or congestion.
How latency is reduced
You cannot beat physics, but a lot of effort goes into trimming latency where possible. Common approaches include:
- Bringing content closer. Networks store copies of popular content in many locations worldwide, so your request travels a shorter distance. This is why a well-built site can feel fast no matter where you are.
- Better routing. Sending data along shorter, less congested paths cuts the number of hops and the time spent queuing.
- Wired connections. A cable to your router usually has lower and steadier latency than Wi-Fi, which is why serious gamers often plug in rather than connect wirelessly.
- Reducing the work. Websites and apps can be designed to make fewer round trips, so the unavoidable delay of each one matters less.
Small improvements at each stage add up to a connection that feels noticeably more responsive.
The bottom line
Latency is the delay between making a request and the first response arriving, measured in milliseconds, and it is what makes a connection feel snappy or sluggish. It is not the same as bandwidth: latency is about delay, bandwidth is about how much data flows at once, and a fast connection can still feel laggy if latency is high.
Distance, network hops, congestion and your connection type all add to it. Low latency matters most for real-time tasks like calls and gaming, and less for big downloads. Understand the difference, and you will know whether your next connection problem calls for more bandwidth or for lower latency, the quiet number that shapes so much of how the internet feels.