Are You Getting the Internet Speed You Pay For? How to Check (and What to Do If You're Not)
Most Americans pay for internet speeds they never actually receive. FCC data shows DSL providers consistently deliver less than advertised, while even cable and fiber can fall short during peak hours. Here is exactly how to test whether your ISP is delivering what you pay for, what counts as an acceptable gap, and what to do when your speeds come up short.
You are paying for a 300 Mbps internet plan. You run a speed test and see 180 Mbps. Is your ISP ripping you off, or is something else going on? The answer depends on how you test, when you test, and what "up to 300 Mbps" actually means in your contract.
The short version: most ISPs deliver 80 to 95 percent of advertised download speeds on average, according to the FCC's Measuring Broadband America reports. Fiber providers tend to hit or exceed their numbers. Cable providers are usually close but dip during peak hours. DSL providers consistently underdeliver. But those are averages — your experience depends on your specific setup, and the gap between what you pay for and what you get could be costing you real performance.
Measure your real-world speed, ping, jitter, and bufferbloat. Free, no signup required.
> Run Free Speed TestWhat "up to" speeds actually mean
Every ISP advertises speeds with the words "up to" in front of the number. This is not just marketing language — it is a legal distinction. "Up to 500 Mbps" means 500 Mbps is the maximum speed your connection can achieve under ideal conditions. It is not a guarantee that you will see that speed consistently.
Ideal conditions means: a wired Ethernet connection, one device on the network, no congestion on your ISP's network, and a server close to you. In practice, you will almost never hit those conditions simultaneously. The realistic expectation is 80 to 95 percent of your advertised speed on a wired connection during off-peak hours.
How different ISP types actually perform
The FCC's Measuring Broadband America program has been testing real-world ISP speeds since 2011. The most recent reports reveal a clear pattern by connection type:
| Connection Type | Typical % of Advertised Speed | Peak Hour Drop | Consistency |
|---|---|---|---|
| Fiber (AT&T Fiber, Verizon Fios, Google Fiber) | 95 to 105% | Minimal (1 to 3%) | Excellent |
| Cable (Xfinity, Spectrum, Cox) | 85 to 100% | Moderate (5 to 15%) | Good, varies by area |
| DSL (AT&T DSL, CenturyLink DSL) | 60 to 85% | Significant (10 to 25%) | Poor |
| Fixed Wireless 5G (T-Mobile, Verizon) | 70 to 95% | Significant (10 to 30%) | Variable |
| Satellite (Starlink, HughesNet) | 50 to 90% | Heavy (20 to 40%) | Highly variable |
Fiber providers often exceed their advertised speeds because the technology has massive overhead capacity. Cable providers are generally solid but share bandwidth with your neighbors, which causes slowdowns during evening hours. DSL degrades with distance from the telephone exchange — if you live more than a mile from the nearest node, you may never see your full plan speed.
How to test your speed accurately
A speed test is only useful if you do it right. Most people test on Wi-Fi with ten tabs open and three family members streaming, then blame their ISP for slow results. Here is how to get a number you can actually trust:
- Connect via Ethernet — Plug your computer directly into your router or modem with an Ethernet cable. Wi-Fi introduces 20 to 40 percent overhead that has nothing to do with your ISP. If your speed looks low on Wi-Fi, the problem might be your router or Wi-Fi setup, not your ISP.
- Close everything else — Shut down streaming, cloud backups, game launchers, and other devices. Any background traffic eats into your test result. You want to measure what your ISP delivers, not what is left over after your household uses some of it.
- Run the test at pong.com — Our speed test measures download, upload, ping, jitter, and bufferbloat in one pass. Unlike some tests, we use multiple connections to saturate your link and measure true throughput.
- Test multiple times — Run at least three tests spaced five minutes apart. A single test can be skewed by a momentary network hiccup. Average the results for a reliable number.
- Test at different times — Run tests in the morning, afternoon, and evening (7 to 10 PM). Peak-hour performance is what matters most since that is when you actually use the internet.
- Record your results — Write down or screenshot each test. You need data to make a case to your ISP if speeds are consistently low.
Why your speed might be lower than your plan
Before you call your ISP, rule out these common causes. Most of the time, the problem is inside your home rather than on the ISP's network:
Wi-Fi bottleneck (most common)
If you are testing on Wi-Fi, you are not testing your ISP — you are testing your router. A Wi-Fi 5 router maxes out around 400 to 600 Mbps in real-world conditions. If you are paying for a gigabit plan and testing over Wi-Fi 5, you will never see more than 500 to 600 Mbps even if your ISP delivers the full gigabit. Upgrade to Wi-Fi 6 or 6E, or test on Ethernet to isolate the ISP from the equation.
Old or cheap router
ISP-provided routers are often underpowered. If your router is more than four years old, its processor may not be able to handle the throughput of a modern plan. A router rated for 300 Mbps cannot deliver 500 Mbps regardless of what your ISP sends. Check your router's specifications against your plan speed.
Network congestion (peak hours)
Cable internet shares bandwidth with your neighbors. Between 7 and 11 PM, when everyone is streaming, speeds can drop 10 to 25 percent. This is normal for cable and not something your ISP will fix — it is inherent to the technology. If peak-hour slowdowns are severe, fiber is the only real solution.
Ethernet cable or port limitations
If you have a gigabit plan and test at exactly 100 Mbps on Ethernet, you probably have a Cat5 cable (not Cat5e or Cat6) or a 100 Mbps port on your device. Cat5 cables and older network adapters max out at 100 Mbps. Replace the cable with Cat5e or Cat6 and verify your device has a gigabit Ethernet port.
ISP throttling
Some ISPs intentionally slow down certain types of traffic — particularly streaming video, gaming, and torrents — during peak hours. If your speed test looks fine but Netflix buffers, throttling could be the cause. See our ISP throttling guide for how to detect and work around it.
Distance from ISP node
For DSL and some older cable networks, your speed depends on how far you are from the nearest network node. DSL loses speed significantly beyond a mile. Fixed wireless 5G depends on line of sight to the tower. If distance is the issue, no amount of troubleshooting inside your home will help — you need a different plan or provider.
What speed gap is acceptable vs. worth complaining about
Not every gap between your plan speed and your test result means something is wrong. Here is a practical framework for deciding when to act:
| Measured vs. Plan (wired test) | Verdict | Action |
|---|---|---|
| 90 to 100%+ | Normal — ISP is delivering well | No action needed |
| 80 to 90% | Acceptable — within industry norms | Monitor but not urgent |
| 70 to 80% | Below expectations — worth investigating | Check equipment, test at multiple times, call ISP if persistent |
| 50 to 70% | Underdelivering — legitimate complaint | Document results, contact ISP, file FCC complaint if unresolved |
| Below 50% | Serious problem — possible outage or fault | Contact ISP immediately, request technician visit |
What to do if your ISP is not delivering
If your wired speed tests consistently show less than 80 percent of your plan speed after ruling out equipment issues, here is how to escalate effectively:
Step 1: Document everything
Run speed tests at multiple times over three to five days. Record the date, time, and result of each test. Screenshot your results from pong.com. Note whether you tested on Ethernet or Wi-Fi, and what plan you are paying for. The more data you have, the stronger your case.
Step 2: Call your ISP
Contact your ISP's support line and reference your documented test results. Be specific: "I am paying for 300 Mbps. Over the past five days, I have averaged 190 Mbps on a wired connection across 15 tests." Ask if there are known issues in your area. Request a line test and a technician visit if remote diagnostics do not explain the gap.
Step 3: Request a plan adjustment or credit
If your ISP confirms that your line cannot deliver your plan speed (common with older cable and DSL infrastructure), ask to be moved to a lower-tier plan at a lower price. You should not pay for 500 Mbps if the infrastructure to your home can only deliver 300 Mbps. Many ISPs will offer a credit or plan change if you push for it.
Step 4: File an FCC complaint
If your ISP does not resolve the issue, file a complaint at consumercomplaints.fcc.gov. The FCC requires ISPs to respond to every consumer complaint within 30 days. In many cases, a complaint triggers a faster and more thorough response from the ISP than calling support does. This is free and takes about 10 minutes.
Step 5: Switch providers
If another provider serves your address, switching is often the most effective solution. Fiber is almost always better than cable for consistent speed delivery. Use pong.com's speed test history to compare what your current provider delivers versus what a new provider promises.
Do not forget about upload speed
Most people focus only on download speed, but upload speed matters for video calls, cloud backups, livestreaming, and working from home. Cable ISPs often advertise download speed prominently while burying upload speeds in the fine print. A plan advertised as "500 Mbps" might include only 10 or 20 Mbps upload.
Fiber connections are typically symmetrical — you get the same speed up and down. Cable connections are almost always asymmetric, with upload speeds one-tenth to one-twentieth of the download speed. If upload performance matters to you, this is one of the strongest arguments for switching to fiber.
| Activity | Upload Speed Needed |
|---|---|
| Zoom / video call (1080p) | 3 to 5 Mbps |
| Livestreaming (1080p) | 6 to 10 Mbps |
| Cloud backup (large files) | 10 Mbps+ preferred |
| Working from home (general) | 5 to 10 Mbps |
| Gaming (online multiplayer) | 1 to 3 Mbps |
| Email and browsing | Under 1 Mbps |
Why different speed tests give different results
You run a test on pong.com and get 250 Mbps. Then you try your ISP's built-in test and get 400 Mbps. Which one is right? The answer depends on what each test is actually measuring.
- ISP-hosted tests — These test the speed between you and your ISP's own server. The data never leaves your ISP's network, so it avoids any congestion at peering points. This gives you the highest possible number but does not reflect real-world performance, because all the websites and services you actually use are outside your ISP's network.
- Independent tests (pong.com, fast.com) — These test the speed between you and servers outside your ISP's network, which is closer to what real browsing, streaming, and gaming actually experience. Results may be slightly lower because the data crosses network boundaries.
- Server location — A speed test server 50 miles away will give different results than one 2,000 miles away. Pong.com automatically selects a nearby server for the most relevant measurement.
- Testing methodology — Some tests use a single connection (measuring one stream), while others use multiple parallel connections (measuring aggregate throughput). Multiple connections give a higher and more accurate result for modern broadband connections.
Frequently asked questions
?>Should I get exactly the speed I pay for?
?>Why is my Wi-Fi speed so much lower than my plan?
?>Does my ISP slow down my internet at night?
?>Can I get a refund if my ISP does not deliver full speed?
?>Is my ISP throttling my connection?
?>How often should I test my speed?
Bottom line
Paying for speed you do not actually receive is frustrating, but the fix starts with testing correctly. Most perceived ISP problems turn out to be Wi-Fi limitations, old equipment, or testing at the wrong time. Once you rule those out with proper Ethernet testing at pong.com, you will know whether the issue is yours to fix or your ISP's to answer for.
- Always test on Ethernet — Wi-Fi results do not reflect your ISP's performance.
- 80 to 95 percent of plan speed is normal — Do not expect exactly what the plan says.
- Test at peak hours — Evening performance is what actually matters.
- Fiber delivers most consistently — If available, it almost always hits or exceeds advertised speeds.
- Document before you complain — Multiple tests over several days gives you leverage with your ISP.
- File an FCC complaint if needed — It is free and forces your ISP to respond within 30 days.