Why Are My Speed Test Results Lower Than My Bandwidth Package

Why Are My Speed Test Results Lower Than My Bandwidth Package

Why Am I Not Getting The Advertised Speed That I'm Paying For When I Run A Speed Test?


There are quite a few 
  • Devices (phones, tablets, PCs, etc…) can have very different Wi-Fi and cellular radio capabilities. This means you might get one Speedtest result on one device and a different result on another, even using the same provider. Some devices may not be able to measure the full speed of your internet service. It’s also possible that your Wi-Fi router doesn’t support the full speed of your service.
  • Speedtest servers may perform differently. Generally, you will get faster speeds from servers closer to you. We recommend testing to a variety of test servers to get the most complete picture of your speed. Speedtest has the world’s largest testing server network, which means you will always have the benefit of testing to a server near your geographic location.
  • Other speed testing services use different servers in different locations than Speedtest, so differences in speeds between testing services are not uncommon.
  • Browsers (Chrome, Firefox, Edge, Safari, etc…) have different capabilities and may provide different results, particularly on high-speed connections.

Your connection is not properly isolated at the time you ran your speed test


To accurately benchmark your connection as best as possible when you run a speed test you need to be sure that there are no other devices on the network at the time of performing the speed test. The only device that should be connected to your network is the device you are performing the speed test on. If you have any other devices connected to the network there is a more-than-likely possibility that the aggregate utilization among all devices will make your test benchmark slower. Even when you aren't consciously using your devices they are more often than not consuming any available bandwidth in the background. A good example of this are gaming consoles. Certain gaming consoles have a "soft-shutdown" where they appear to be turned off but in actual fact they are hard at work in the background downloading the latest game updates or patches or similar. So to you running your speed test it looks like the console is off because there aren't any lights shining or flashing on the console, it's busy consuming as much available bandwidth as it can, you run your speed test and don't understand how you are receiving such a poor result.

Even if you only have one device, if that one device is doing something in the background that isn't part of your speed test, less total throughput will be available to the test and the results will be inaccurate as a result. 

To ensure that your speed test results are as accurate as possible you need to ensure the following:

- When you perform your speed test ensure that the only device on your network is the one you are running your speed test from
- If you are using a wired connection ensure that you are plugged directly into the Internet router or as close as possible to the router
- If you are using a wireless connection (WiFi) ensure that you are receiving full signal and that you are as close to your access point as possible




When you perform a speed test your device has to scan the Internet and find the best quality and most relevant server it can in order to provide the most accurate result. This scan normally takes place during the short pause between the time you press the start button and when you first start seeing speed results. During this period your device has determined what it thinks the best route would be together with what the best server location is. This connection has to travel over a large number of links and can often take place over great distances as well. As soon as the traffic that is used to perform your speed test has left your ISP's network it means that it has also left your ISP's area of responsibility as well. As the traffic leaves your ISP's network it is effectively in the wild and has to compete with all other traffic in all other instances all trying to reach their destination at the same point in time that your traffic is trying to reach the server you are testing to. Think about the links and routes as vehicles travelling down roads and joining freeways. If the freeway is flowing smoothly you will have little difficulty joining the freeway and moving onward to your destination, however if there is congestion and slow moving traffic your overall experience will be poor. Additionally if your destination is also congested or overwhelmed you will experience an even more slower experience. In this analogy you can also think of the server you are testing as a fuel station that has a lot of cars already there queueing in line to fill up. Your overall experience will be poor.

In this analogy the ISP is in charge of the roads in and around your house and there is very little that they can do to assist in improving way the traffic flows onto the freeway as well as how to improve congestion at the filling station. 

To ensure that your speed test results are as accurate as possible you need to ensure the following:



There are quite a few possibilities here. The below is list of the most common reasons there are discrepancies between the advertised rate and the realized rate:

  • The internet connection isn’t properly isolated. To accurately benchmark the connection you need to be sure that only 1 computer is on the network. Otherwise, the aggregate utilization among all devices will make your test benchmark slower. Even when you aren’t consciously downloading content, updates, synchronizations, etc. are occurring at all times on most modern devices. Even if you only have 1 device, if that 1 device is doing something in the background that isn’t part of your benchmark test, less total throughput will be available to the test and the results will be inaccurate as a result. You also need to be sure all devices/equipment within your network that are part of the testing path can handle the maximum throughput of the connection.
  • You’re using wireless or hardwired connection that is slower than your internet connection. Networks are only as fast as the slowest link in the communication path. Thus, if your wireless access point, Ethernet NIC, Switch, etc. are slower than the maximum throughput of the connection, you’ll be bottlenecked by how your testing platform connects.
  • There’s a signal quality issue between you and your ISP so the maximum throughput is less than the subscribed rate due to signal integrity issues. This is common in cable/coaxial connections when they’re experiencing ingress (interference). This is somewhat less common with technologies such as DSL as the copper pairs are generally dedicated to a specific account through to the node. Cable internet is similar to having one line that split as many times as needed to reach all the customers.
  • You may be benchmarking to a server that has a slow link somewhere between your ISP and the testing server. Ideally, you should use a testing server that is hosted by your ISP so the test stays entirely within their network. Realize, that advertised speeds are the maximum rate between your premise and the ISP ONLY. Once traffic leaves your ISP’s network, all bets are off. To put it another way, you may have a 100 mbps connection between you and your ISP, with perfect signal integrity and properly isolated, but there’s a good chance that most resources on the internet aren’t going to let you download from them at 100 mbps.
  • Your advertised rate is generally not guaranteed by most broadband providers. They sneak this into the contract and disclaimers using verbiage like “up to…” and “in times of network congestion.” Most internet providers oversubscribe their connections. What this means is that not every customer in the area is able to get their maximum speeds if they all attempt to do so at the same time. As a result, you slow down during peak usage periods.

Essentially, if the advertised rate and the realized rate don’t match, you’ve either got flawed testing methodology, a technical issue or your ISP doesn’t actually guarantee the advertised speed.


    • Related Articles

    • What Factors May Impact My SpeedTest Result

      What factors may impact my Speedtest result? This document is an extract directly from Ookla's speedtest.net knowledgebase the full article can be found here Speedtest measures the speed between your device and a test server, using your device's ...

    • How To Find Your Router's IP Address

      How to Find Your Router’s IP Address If you want to make changes to your router, you need to know your router’s IP address. Perhaps you want to change the network name, create a new WiFi password, or use a different channel to boost your internet ...

    • How to Add a Printer in Windows 10

      How to Add a Printer in Windows 10 Adding a printer to your Windows 10 laptop or desktop these days is user-friendly and requires little-to-no technical knowledge. You can add a printer via USB, your wireless network, or Bluetooth. Here’s how to add ...

    • Which Third-Party DNS Server Should You Use?

      Which Third-Party DNS Server Should You Use? If you want to use a third-party DNS server, some of the best options are the Google Public DNS, Cloudflare, and OpenDNS. Each of these providers has a primary and secondary DNS address that you can use. ...

    • How to Boot in Safe Mode in Windows 10

      How to Boot in Safe Mode in Windows 10 One of the ways to troubleshoot a Windows computer is to boot it in safe mode. According to Microsoft, you can pinpoint what is wrong with your computer by observing how it performs in safe mode. Specifically, ...