Advertisement · 728 × 90

What Is an Internet Speed Test?

An internet speed test measures how fast data travels between your device and the internet. SpeedIQ uses Cloudflare's global edge network to give you the most accurate real-world measurement — not a number inflated by ISP-favored servers. For the full context of why speed tests matter and what each metric tells you, see our complete internet speed guide.

A reliable speed test measures three things: download speed, upload speed, and latency (ping). Each matters for different activities. Speed tests typically take 10–30 seconds and run automatically once you click Start.

The Three Numbers That Matter

Download Speed

Download speed measures how fast data comes from the internet to your device, expressed in megabits per second (Mbps). Streaming, browsing, and downloading files all depend on this number. For 4K Netflix streaming you need at least 25 Mbps. For a household with multiple users, 100 Mbps or more is recommended. Gigabit connections (1000 Mbps) handle 4K streaming for the entire home plus heavy downloading without slowdown.

Upload Speed

Upload speed measures how fast you can send data from your device to the internet. Video calls, uploading files, and live streaming depend on upload speed. Most home internet connections have much lower upload than download — this is normal for cable and DSL connections. Fiber connections offer symmetric upload, meaning the same speed in both directions, which makes them ideal for content creators and remote workers — see our fiber vs cable comparison for the technology breakdown.

Ping and Latency

Ping measures how quickly your device receives a response from a server, in milliseconds. Lower ping means more responsive connections. For online gaming, anything under 30 ms is excellent and under 60 ms is good. Video calls feel natural at under 100 ms. Above 150 ms, real-time applications start to feel laggy. Ping depends heavily on physical distance to the server, the number of network hops, and the technology your ISP uses.

How to Run an Accurate Speed Test

Most speed tests give wildly different numbers because conditions change constantly. To get a reliable result, follow this checklist:

  • Connect via Ethernet, not WiFi. WiFi adds variable latency and reduces speed by 20–50% on most setups. Wired tests show your real line speed.

  • Close all other apps and tabs. Background updates, cloud sync, and streaming consume bandwidth and skew results.

  • Disconnect other devices. Phones, smart TVs, and consoles all share the same connection. Pause downloads and updates on every device during the test.

  • Test at different times of day. Speeds drop during peak hours (typically 6–11 PM). For a complete picture, test in the morning, afternoon, and evening.

  • Run the test 3 times. Take the average of three runs to smooth out brief fluctuations from server-side conditions.

Why Your Speed Test Result Differs From Your ISP Plan

If you pay for 1 Gbps but your speed test shows 400 Mbps, several real-world factors are likely involved.

The first is WiFi loss. Even modern WiFi 6 routers cannot deliver 1 Gbps consistently to a single device through walls. The second is peak-hour congestion — cable internet is shared infrastructure, so neighbors using the network reduce your available bandwidth. The third is router limitations. Old routers cannot route gigabit traffic regardless of your plan; an upgrade may unlock significant headroom. The fourth is ISP throttling: some ISPs throttle specific services like streaming or torrenting, which is detectable through targeted tests. The fifth is the device itself: an old laptop with a 100 Mbps Ethernet port physically cannot exceed that limit even on a 1 Gbps plan.

WiFi vs Wired: Always Test Both

The single most important test isolation is wired vs wireless. Connect your computer directly to your router with an Ethernet cable and run the test. Then disconnect, switch to WiFi, and run it again. The difference is your WiFi loss.

If wired matches your plan but WiFi is much slower, the problem is your wireless setup — router placement, channel selection, interference, or hardware. If both wired and WiFi are slow, the problem is upstream: your modem, cabling, or ISP itself. To dig deeper into wireless issues, see our guide to improve WiFi speed at home.

Mobile vs Home Speed Tests

Mobile data speed varies dramatically with location, network load, and signal strength. A 5G connection in an urban core can exceed 1 Gbps, while the same phone in a basement may struggle to hit 10 Mbps over LTE.

For mobile testing, run the test outdoors with a clear view of the sky to get the strongest signal. Make sure your phone is fully charged or plugged in — some devices throttle CPU on low battery, which reduces the speeds the modem can sustain. If you frequently get bad results in one location, check the carrier's coverage map; you may simply be in a dead zone.

VPN Speed and Speed Tests

Connecting to a VPN always reduces your speed because traffic is encrypted and routed through an intermediate server. A well-optimized VPN on a fast connection might cost you 5–15%. Distant servers, overloaded providers, or older protocols like OpenVPN TCP can drop speeds by 50% or more.

To test VPN impact correctly, first run a baseline speed test without the VPN. Then enable the VPN, connect to your usual server, and run the test again with the same speed test server. The percentage difference is your VPN overhead. For a complete methodology see our VPN speed test guide.

Reading Advanced Speed Test Metrics

Beyond the headline download/upload numbers, modern speed tests expose advanced metrics worth understanding.

Jitter is the variation in ping over time, measured in milliseconds. Low jitter (under 5 ms) means a stable connection ideal for VoIP and gaming. High jitter (over 30 ms) causes choppy video calls and rubber-banding in games even when average ping is acceptable.

Packet loss is the percentage of data packets that fail to reach their destination. Anything over 1% is a serious problem — it causes connection stalls, dropped calls, and game disconnects. Persistent packet loss usually points to faulty cabling, an overloaded router, or an ISP routing issue.

Bufferbloat measures how ping increases under load. A connection with low idle ping but very high under-load ping has bufferbloat — common on cable connections. The fix is QoS (Quality of Service) configuration on your router, which prioritizes interactive traffic over bulk transfers.

Speed Test Servers: Local vs Distant

Most speed tests automatically choose the nearest server to give you the highest possible numbers. This is fine for measuring your connection capacity, but it does not reflect real-world performance to services hosted further away. To test the route to a specific service — a game server, a hosting provider in another country — manually choose a server in that region.

SpeedIQ runs on Cloudflare's edge network, which has 300+ points of presence globally, so the test server is almost always within a few network hops of your device. This makes the result more representative of typical web browsing performance than tests using a single distant server.

Common Speed Test Mistakes

Even careful testers make these common mistakes that produce misleading results:

  • Testing once and trusting the number. A single test captures one moment. Always run 3 tests and average them.

  • Testing right after a router restart. Wait 2–3 minutes after restart for the connection to stabilize.

  • Testing while VPN is on but forgotten. Always check VPN status before running tests for raw line speed.

  • Comparing tests from different services. Different speed test providers use different server locations and methodologies. Stick to one tool for fair comparisons.

  • Ignoring upload when comparing connections. Many people focus only on download. For video calls, cloud backup, and streaming, upload speed often matters more.

When to Worry About Your Speed

If your wired speed is consistently more than 30% below your plan, contact your ISP. They can run a remote line test and may schedule a technician visit. Document multiple tests with timestamps to support your case.

If wired matches your plan but WiFi is consistently below 50% of wired, the issue is local — router placement, hardware, or interference. Buying a new router is often cheaper than upgrading your plan.

If ping spikes randomly or packet loss exceeds 1%, the issue is more serious. Check cable connections at the modem and router, then call your ISP if the problem persists.

Speed Test on Different Devices

The device you test on affects the result. A modern laptop with a gigabit Ethernet port can saturate gigabit connections. An older laptop with a 100 Mbps port physically caps at 95 Mbps regardless of your plan. Phones tested over WiFi rarely reach more than 600–700 Mbps even on the best networks.

For the most accurate test of your overall internet plan, use a modern computer connected via Ethernet. For testing the WiFi performance in a specific room, use the device you actually use there. The right test depends on the question you are trying to answer.

Conclusion

An accurate speed test takes more than clicking a button. Wire your device when possible, close other apps, run multiple tests, and compare wired vs WiFi to isolate where bottlenecks live. Pay attention to upload, ping, jitter, and packet loss — not just download speed. Once you know your real numbers, you can decide whether the problem is your plan, your router, your hardware, or simply expectations set by marketing materials.

Run a free speed test now using SpeedIQ at speediq.org to get accurate Cloudflare-backed measurements with full advanced metrics — or read our guide on what makes a good ping for gaming if latency rather than bandwidth is your concern.

Related Reading

Part of the Vatha network.