How to Run an Internet Speed Test (And Actually Understand the Results)

Running an internet speed test takes about 30 seconds. Understanding what the numbers mean — and whether they reflect your real-world connection — takes a little more context. Here's what's actually happening when you run a test, what the results tell you, and why two people on the same plan can get very different readings.

What a Speed Test Actually Measures

A speed test works by sending and receiving data packets between your device and a nearby test server, then calculating how fast that transfer happened. Most tests measure three core values:

• Download speed — how quickly data moves from the internet to your device (measured in Mbps)
• Upload speed — how quickly data moves from your device to the internet
• Ping (latency) — the round-trip time for a signal to reach the server and return, measured in milliseconds (ms)

Some tests also report jitter, which measures the consistency of your ping over multiple packets. High jitter can cause choppy video calls or lag in online games even when average latency looks fine.

How to Run a Basic Speed Test 🖥️

1. Connect directly to your router via Ethernet if possible — this eliminates Wi-Fi as a variable
2. Close background apps and browser tabs that might be using bandwidth
3. Go to a test tool — Speedtest by Ookla, Fast.com (run by Netflix), and Google's built-in speed test (search "internet speed test") are widely used options
4. Click the test button and wait for results — most tests complete in under a minute
5. Run the test 2–3 times at different points in the day for a more representative picture

The test picks a nearby server automatically, though most tools let you choose a different server manually. Choosing a server farther away will generally show higher latency and sometimes lower throughput — useful if you're troubleshooting a connection to a specific region.

What the Numbers Mean in Practice

These are general reference points, not guarantees. Actual experience depends on how many devices share the connection, what apps are running, and how well your equipment handles traffic.

Why Your Speed Test Results Vary

This is where most confusion happens. Two people on the same 200 Mbps plan can test at 180 Mbps and 60 Mbps respectively — both correctly — because the path between device and server involves several variables:

Your hardware matters. Older routers, particularly those still using the 2.4 GHz band exclusively or running outdated firmware, can bottleneck a fast connection before the signal even leaves your home network. A Wi-Fi 5 or Wi-Fi 6 router handles throughput very differently than a 10-year-old unit.

Wi-Fi vs. wired connections produce different results. Wi-Fi introduces signal loss, interference from neighboring networks, and distance degradation. A speed test over Ethernet almost always shows faster, more consistent speeds than the same test over Wi-Fi from another room.

Device specs affect results. A smartphone's network adapter, a laptop's Wi-Fi card, and a desktop connected via Gigabit Ethernet all have different maximum throughput capabilities. Testing on an older device may reflect hardware limits rather than your ISP's actual delivery.

Time of day creates real differences. ISPs share bandwidth across neighborhoods. Peak hours — typically evenings — often produce lower test results than the same connection tested at 10am. This is called network congestion and affects cable and DSL connections more than fiber.

The test server itself is a variable. If the selected server is under load or geographically distant, results will look lower than your connection's actual capability.

Testing Wi-Fi Specifically vs. Your Whole Network 📶

If you're troubleshooting, it helps to isolate where a speed problem lives:

• Test wired first. Plug a laptop directly into the router and run a test. This is your baseline — the speed your ISP is delivering to your home.
• Test wirelessly from the same room as the router. Any significant drop here points to a router issue.
• Test from another room or floor. A larger drop here suggests a Wi-Fi coverage problem, not an ISP problem.

This sequence matters because the fix for "my ISP is under-delivering" is very different from "my router is struggling with signal distribution."

What ISPs Promise vs. What You Receive

Internet plans are typically sold with "up to" speeds — meaning the advertised number is a ceiling, not a guaranteed floor. Fiber connections tend to deliver speeds closest to advertised rates because they don't share bandwidth in the same way cable infrastructure does. Cable internet uses shared coaxial lines, so speeds fluctuate more with neighborhood usage. DSL speeds degrade with distance from the provider's central office.

If your speed test consistently shows significantly less than your plan's advertised speed — particularly on a wired connection during off-peak hours — that's useful data to bring to your ISP.

The Part That Depends on Your Situation

Running a speed test gives you a snapshot. Interpreting it accurately requires knowing your setup: what plan you're paying for, how your network is configured, what devices you're testing on, and what you're actually trying to do with your connection.

A household with three people video conferencing simultaneously has different requirements than someone browsing and streaming alone. A remote worker uploading large files cares more about upload speed than a household of streamers. What counts as "fast enough" — and where a bottleneck actually sits — shifts depending on those specifics.