In this article
- What Mbps Actually Measures
- Download vs. Upload Speed
- Why a Speed Test Number Rarely Matches Your Plan Exactly
- Wired vs. Wi-Fi Speed Test Differences
- How Many Devices Your Plan Can Realistically Support
- Latency: The Number Speed Tests Underemphasize
- A Simple Process for Narrowing Down a Slowdown
- Frequently Asked Questions
"Up to 500 Mbps" on your internet bill and "87 Mbps" on a speed test result you just ran can both be technically accurate at the same time, which is confusing if you don't know what's actually happening in between. Here's a plainer walk-through of what these numbers represent and why they diverge.
What Mbps Actually Measures
Mbps stands for megabits per second — a measure of how much data can move between your device and the internet each second. It's worth noting this is measured in bits, not bytes: a 100 Mbps connection transfers roughly 12.5 megabytes per second at best, not 100 megabytes, since there are 8 bits in a byte. This trips people up when comparing a connection speed to, say, a file size shown in megabytes.
Download vs. Upload Speed
Download speed measures how fast data comes to you — loading web pages, streaming video, downloading files. Upload speed measures the reverse — sending files, video calls, and increasingly, cloud backups and video conferencing, which have made upload speed more relevant to everyday use than it used to be. Many consumer internet plans, particularly on cable connections, offer significantly higher download speeds than upload speeds, which is worth checking if your household does a lot of video calling or uploads large files regularly.
Why a Speed Test Number Rarely Matches Your Plan Exactly
Several things sit between "the speed your plan promises" and "the number your speed test shows":
- Network congestion, both on your provider's broader network during peak hours and within your own home if multiple devices are active simultaneously.
- The specific server your speed test connects to, and the general internet routing between you and it, which can vary between test providers and even between runs.
- Your router's own processing capacity, which on very high-speed plans (500+ Mbps) can occasionally become the bottleneck if the router itself is older or lower-end.
- Wi-Fi overhead, covered in more detail below, which is often the single biggest gap between plan speed and real-world results.
Wired vs. Wi-Fi Speed Test Differences
A speed test run on a device connected directly to your router via ethernet cable measures something meaningfully different from the same test run over Wi-Fi. Wi-Fi introduces overhead from radio interference, distance from the router, the number of other devices sharing the same wireless band, and the specific Wi-Fi standard your device supports — a phone with an older Wi-Fi chipset will never hit the same speeds as a newer laptop, even sitting in the exact same spot. If your wired speed test roughly matches your plan but your Wi-Fi results are notably lower, that points toward a Wi-Fi coverage or interference issue rather than a problem with your actual internet plan.
How Many Devices Your Plan Can Realistically Support
There's no fixed number, since it depends heavily on what those devices are actually doing — ten smart plugs use negligible bandwidth compared to two people simultaneously streaming 4K video and one on a video call. As a rough guide, standard-definition video calls and browsing use relatively little bandwidth individually, while 4K streaming, large file transfers, and competitive online gaming (more sensitive to latency than raw speed) tend to be the activities that actually strain a household's connection when several happen simultaneously.
Latency: The Number Speed Tests Underemphasize
Most speed test results emphasize download and upload numbers prominently, with latency (often called "ping," measured in milliseconds) shown smaller or further down the results. Latency measures the delay before data starts moving, rather than how much moves once it does, and it matters disproportionately for real-time activities — video calls, online gaming, remote desktop use — where even a fast download speed doesn't compensate for a laggy, high-latency connection. If video calls feel delayed or gaming feels unresponsive despite a speed test showing decent Mbps numbers, latency is worth checking specifically rather than assuming raw speed is the whole picture.
A Simple Process for Narrowing Down a Slowdown
- Run a speed test wired directly to your router, if possible, to establish your actual baseline separate from Wi-Fi.
- Run the same test over Wi-Fi in the same room as the router to isolate whether distance is a factor.
- Run it again farther from the router, in the room where the problem is actually occurring, to see how much Wi-Fi coverage is contributing.
- Check how many devices are active on your network at the time, and whether anything is doing something bandwidth-heavy in the background (a large backup or update, for instance).
- If wired speed itself is well below your plan, the issue more likely sits with your provider or modem rather than your router or Wi-Fi setup.
Frequently Asked Questions
Should my speed test result exactly match my plan's advertised speed?
Not exactly — some gap between advertised and measured speed is normal and expected due to network overhead and testing variability. A consistently large gap, especially on a wired connection, is worth raising with your provider.
Does upgrading my internet plan fix Wi-Fi dead zones?
No — plan speed and Wi-Fi coverage are separate issues. A faster plan won't reach a room your router's signal doesn't cover; that's a coverage problem better addressed with router placement, an extender, or a mesh system, covered in our mesh Wi-Fi guide.
Why does my speed test vary so much between runs?
Network conditions genuinely fluctuate — congestion on your provider's network, the specific test server's load, and activity from other devices on your own network all shift somewhat from moment to moment. Running a test a few times at different points in the day gives a more reliable picture than a single result.