Network speed vs bandwidth vs throughput: what's the difference?
When people talk about network performance, they often use network speed, bandwidth and throughput as if they mean the same thing. They don't. Each measures something different, and knowing the difference helps you choose the right internet plan, diagnose a slow connection and get the most from your network. This article explains what network speed, bandwidth and throughput are, how they relate to latency, and how they compare.
Table of Contents
- What is network speed?
- Common units of measurement
- What is bandwidth?
- What is throughput in networking?
- Bandwidth vs throughput: the key difference
- Where does latency fit in?
- Network speed vs bandwidth vs throughput: how they compare
- Optimising network metrics for video conferencing
- How Digital Samba handles these metrics for you
- Conclusion
- Frequently asked questions
What is network speed?
Network speed is the rate at which data moves between two devices on a network. It is usually what people mean when they talk about network or application performance: a faster network speed means quicker downloads, uploads and browsing.
Think of network speed as the flow rate of water through a pipe. Even a large pipe limits flow based on water pressure, the pipe's shape and any blockages. In the same way, network speed depends on several network and device factors, from the protocol used to transfer data to whether a device is wired or wireless and how well a server handles many clients at once.
Factors that affect network speed
Several things shape network speed:
- Hardware: the quality and age of routers, switches, cables and the device itself.
- Connection type: a wired Ethernet link is usually faster and steadier than Wi-Fi.
- Distance and signal: longer cable runs or a weak wireless signal slow things down.
- Server capacity: a server serving many users at once can only send data so fast to each one.
- Protocol overhead and congestion: the transfer protocol and competing traffic both eat into the usable rate.
Common units of measurement
Network speed is measured in bits per second (bps), the smallest unit for data transfer rates. In practice, speeds are usually expressed in larger units:
- Kilobits per second (kbps): 1,000 bits per second
- Megabits per second (Mbps): 1,000,000 bits per second
- Gigabits per second (Gbps): 1,000,000,000 bits per second
For example, a typical home internet connection may run at 100 Mbps, while a high-speed business connection may reach 1 Gbps or more.
What is bandwidth?
Bandwidth is the data transfer capacity of a network: the maximum amount of data that can be carried over a connection in a given time. Like speed, it is measured in bits per second (kbps, Mbps, Gbps).
Back to the water pipe: the wider the pipe, the more water can flow through it at once. The wider the bandwidth of a connection, the more data it can carry at once.
Bandwidth vs network speed
Bandwidth and network speed are related but not the same, and they are not directly proportional. Bandwidth is the capacity of the connection; speed is how quickly data actually moves across it. A connection with high bandwidth can still feel slow because of latency, congestion or hardware limits.
Factors that affect bandwidth
A connection's bandwidth comes down to a few things:
- The physical medium: fibre carries far more than copper or older wireless links.
- Your plan: internet providers cap bandwidth by tier, so your plan sets the ceiling.
- Network hardware: routers and switches have their own capacity limits.
- Shared or dedicated lines: a shared connection splits capacity between users, while a dedicated line reserves it.
- Interference: on wireless links, walls, distance and competing signals reduce usable bandwidth.
What is throughput in networking?
In networking, throughput is the amount of data actually transferred from one place to another in a given time. Where bandwidth is the capacity, throughput is what you really get once real-world conditions take their toll. It is measured in the same units as speed and bandwidth, usually bits per second (bps, kbps, Mbps, Gbps), and a higher throughput means more data is getting through successfully.
Throughput vs network speed
Network speed is the maximum rate the hardware can theoretically manage, a bit like the speed limit on a motorway. Throughput is the data that actually gets through under real conditions. A connection rated at 100 Mbps might deliver only 50 Mbps of throughput during heavy congestion, because real-world limits pull the figure down.
Factors that affect throughput
Throughput almost always comes in below bandwidth, for reasons like:
- Network congestion: competing traffic is the most common reason throughput drops below bandwidth.
- Packet loss and retransmission: lost packets have to be resent, which cuts the effective rate.
- Latency: high round-trip delay limits how much data protocols like TCP will send at once.
- Protocol overhead: headers and acknowledgements use part of the capacity.
- Hardware limits: an overloaded router or an old network card caps what gets through.
Bandwidth vs throughput: the key difference
The most common mix-up is bandwidth versus throughput, so it's worth stating plainly: bandwidth is the maximum data your connection can carry, and throughput is how much actually gets through. Your provider might sell you 100 Mbps of bandwidth, but a speed test during busy hours could show 60 Mbps of throughput.
Bandwidth is the potential; throughput is the reality. The gap between them comes from congestion, packet loss, latency and protocol overhead, which is why your measured speed rarely matches the figure on your plan.
Where does latency fit in?
Network speed, bandwidth and throughput all describe how much data moves. None of them captures delay, and that's where latency comes in. Latency is the time a packet takes to travel from one point to another, usually measured in milliseconds.
You can have plenty of bandwidth and still feel a laggy connection if latency is high, which is why real-time applications like video calls and gaming care about it so much. For a closer look, see our guide to latency and bandwidth, and the related measure of jitter, which is the variation in that delay.
Network speed vs bandwidth vs throughput: how they compare
Here is how the three core network performance metrics line up side by side, with latency added for context.
| Metric | What it measures | Motorway analogy | Typical unit |
|---|---|---|---|
| Network speed | How fast data can move (theoretical rate) | The speed limit | bps / Mbps |
| Bandwidth | Maximum data the link can carry at once | The number of lanes | bps / Mbps |
| Throughput | Data actually delivered in real conditions | The speed you really drive in traffic | bps / Mbps |
| Latency | Delay for a packet to travel | Time waiting at junctions | milliseconds (ms) |
In short: bandwidth is the capacity, throughput is what you actually get, speed is how fast it moves, and latency is the delay along the way. The three data-rate metrics are all measured in bits per second, which makes them easy to compare, while latency is measured in time. Throughput is usually lower than bandwidth, so it is the number that best reflects real-world performance.
Optimising network metrics for video conferencing
For video conferencing, all of these metrics matter at once. Enough bandwidth lets you carry high-quality video and clear audio, steady throughput keeps that data flowing in real time, and low latency keeps conversation natural rather than stilted. If any one of them falls short, you get blurry video, choppy audio or awkward delays.
For businesses that need guaranteed performance, a dedicated connection such as a leased line from Elevate provides stable bandwidth and fewer congestion problems, which keeps online meetings consistently smooth.
How Digital Samba handles these metrics for you
If you are building video into your own product, Digital Samba's embedded video SDK manages these network conditions so you don't have to. It adapts to changing network speed, makes the most of the bandwidth available, and keeps data moving smoothly even when real-world throughput dips, so calls stay stable on imperfect connections.
You can start building with a free developer account and add real-time video with a few lines of code.
Conclusion
Network speed, bandwidth and throughput each measure network performance in a different way: speed is how fast data can move, bandwidth is how much the link can carry, and throughput is how much actually gets through, with latency adding the delay along the way. Knowing the difference helps you pick the right internet plan and track down the real cause of a slow connection. For real-time video, Digital Samba's SDK takes care of these trade-offs for you.
Frequently asked questions
What is the difference between bandwidth and throughput?
Bandwidth is the maximum amount of data a connection can carry; throughput is how much actually gets through in real conditions. Throughput is usually lower than bandwidth because of congestion, packet loss and latency.
Is network speed the same as bandwidth?
No. Bandwidth is the capacity of the connection, while speed is how fast data moves across it. A connection can have high bandwidth but still feel slow if latency or congestion is high.
What is throughput in networking?
Throughput is the amount of data successfully transferred from one point to another in a given time, measured in bits per second. It reflects real-world performance, which is usually lower than the connection's rated bandwidth.
How is throughput different from latency?
Throughput measures how much data gets through, while latency measures how long a single packet takes to travel. You can have high throughput and high latency at the same time, which is why real-time apps watch both.
How do you measure throughput?
You can run a speed test, which reports real throughput alongside your rated speed, or use a tool like iPerf to send controlled traffic between two devices and measure the rate that actually gets through.
Why is my throughput lower than my bandwidth?
Real-world factors eat into capacity: network congestion, packet loss, latency, protocol overhead and hardware limits all mean you rarely get the full bandwidth your plan advertises.
Share this
You May Also Like
These Related Stories

TCP vs UDP: Understanding the Two Core Transport Protocols

Latency vs Bandwidth: What’s the Difference

