How to Check If Your ISP Is Throttling Your Internet: VPN Test, Speed Comparison, and What to Do About It

You run a speed test and see 300 Mbps. Then Netflix buffers. YouTube drops to 480p at 9pm. Your video calls freeze on weekday evenings. The instinct is to blame your router or WiFi — but the culprit may be your internet service provider deliberately slowing specific types of traffic. This is called ISP throttling, and it’s more common than most users realize. Here’s how to detect it definitively and what your options are.

What Is ISP Throttling?

Throttling is the intentional reduction of internet speeds by your ISP for specific traffic types, services, or time periods. Unlike general network congestion — which affects all traffic equally when a cable node is overloaded — throttling uses deep packet inspection (DPI) to identify what you’re doing and selectively slow it down.

Common throttling targets include:

• Streaming services: Netflix, YouTube, and Disney+ are frequently throttled during peak evening hours (7–11 PM) to reduce bandwidth consumption on cable nodes.
• Video calls: Zoom, Teams, and FaceTime use sustained high-bitrate UDP streams that DPI can identify and deprioritize.
• Torrents and P2P: BitTorrent traffic is the most aggressively throttled category across virtually all major ISPs.
• After-data-cap slowdowns: If your plan has a soft data cap (common at 1.2 TB on Xfinity and Cox), your ISP may throttle all traffic once you exceed it — not just one service.

Throttling is distinct from the general congestion that slows cable internet during peak hours for all users. Throttling is targeted: your streaming slows while your speed test — which ISPs often exempt from throttling — still shows full speed.

Why Speed Tests Can Miss Throttling

This is the critical catch: Speedtest.net (Ookla) results are often unreliable for detecting throttling. Many ISPs host Speedtest servers inside their own network infrastructure. Traffic to those servers takes an optimized, high-priority path that bypasses the DPI filters applied to real internet traffic. You can see 500 Mbps on Ookla and simultaneously get 12 Mbps from Netflix — because the ISP treats speed test traffic differently from streaming traffic.

Run the speed test on this site to establish a baseline, but understand that no single speed test confirms or rules out throttling on its own.

Test 1: The Multi-Service Speed Comparison

The fastest way to detect service-specific throttling is to compare results from speed test tools that use different server infrastructure:

1. Run Speedtest.net and note your download speed. This uses ISP-adjacent servers and is likely to show your full plan speed.
2. Run Fast.com (Netflix’s speed test) immediately after. Fast.com routes through Netflix’s own CDN servers — the same infrastructure Netflix uses for streaming. If your ISP is throttling Netflix traffic, Fast.com will show significantly lower speeds than Ookla.
3. Run nperf.com or Cloudflare’s speed test (speed.cloudflare.com) as a third data point using neutral infrastructure.

If Speedtest.net shows 300 Mbps and Fast.com shows 30 Mbps, your ISP is almost certainly applying service-specific throttling to Netflix traffic. A difference of 50% or more between Ookla and Fast.com is the clearest non-VPN signal of throttling available.

Test 2: The VPN Speed Test (Most Reliable Method)

The definitive throttling test is the VPN comparison. When you connect through a VPN, all your traffic is encrypted inside a tunnel. Your ISP can see that you’re connected to a VPN server — but they cannot identify whether that encrypted traffic is Netflix streaming, a video call, or a file download. Because DPI can’t classify the traffic type, it can’t selectively throttle it.

Step-by-Step VPN Throttling Test

1. Connect via Ethernet to eliminate WiFi variables. Run the same test over WiFi if you want, but wire it first for a clean baseline. For more on why wired connections matter, see our guide on Gigabit Ethernet vs WiFi.
2. Without VPN: Run Speedtest.net, Fast.com, and your actual streaming service at peak hour (8–10 PM). Note all results.
3. Connect a VPN using the WireGuard protocol (NordVPN, Mullvad, or ExpressVPN all support it). Choose a server in your own country to minimize added latency.
4. With VPN active: Repeat the exact same tests immediately. Use the same Ookla server if possible.
5. Interpret the results:
   • Speeds are lower with VPN by 10–20%: Normal. VPN encryption adds overhead. WireGuard typically retains 88–94% of baseline speed on a good connection.
   • Speeds are roughly equal with VPN: No throttling detected.
   • Speeds are noticeably higher with VPN — especially on Fast.com or during peak hours: Strong confirmation of ISP throttling. The ISP can’t identify your encrypted traffic and lets it pass at full speed.

A free VPN trial (ProtonVPN offers unlimited free bandwidth) is sufficient for this test. You don’t need a paid subscription just to diagnose throttling.

Test 3: Time-of-Day Pattern Check

Throttling — especially on cable internet — often intensifies during peak evening hours when demand on shared cable nodes is highest. Run your speed test and Fast.com check at three different times:

• 6 AM (off-peak)
• 12 PM (mid-day)
• 9 PM (peak hour)

If your speeds drop by 30% or more at 9 PM compared to 6 AM — and the VPN test shows the drop disappears when your traffic is encrypted — you have strong evidence of time-based throttling rather than simple node congestion. Pure congestion would slow all traffic, including VPN traffic.

What to Do If You Confirm Throttling

Use a VPN as a Workaround

A VPN permanently prevents service-specific throttling because DPI can’t classify encrypted traffic. For streaming and video calls, a WireGuard-based VPN on a nearby server adds only 5–15ms of latency — imperceptible for most uses. It won’t help with data-cap throttling, since the total data consumed is still visible to your ISP even if the content isn’t.

Contact Your ISP

Call customer support and share your test results — specifically the Speedtest vs. Fast.com comparison and the VPN differential. Some throttling is a configuration error rather than policy. Getting it on the record also creates a paper trail if you need to escalate.

File a Complaint

In the US, you can file a complaint with the FCC (fcc.gov/consumers/guides/filing-informal-complaint) and the FTC. The FCC reinstated its net neutrality authority in 2024, making certain throttling practices legally actionable again. Your state attorney general may also have consumer protection jurisdiction over ISP practices.

Switch ISPs

In competitive markets, throttling is a reason to switch. Fiber ISPs (AT&T Fiber, Frontier Fiber, Google Fiber) use point-to-point connections rather than shared cable nodes and have substantially less incentive or infrastructure for service-specific throttling. Our fiber vs cable vs DSL comparison covers the technical reasons fiber connections tend to behave more consistently.

Understanding Your Results

Not every speed drop is throttling. Before concluding your ISP is at fault, check:

• Router placement and WiFi band: A 2.4 GHz connection in a congested apartment building can easily explain 80% slower speeds without any ISP involvement. Run all tests over Ethernet first.
• Your router’s WAN port speed: Some older routers have 100 Mbps WAN ports that cap all traffic regardless of your plan. See our guide on router WAN port bottlenecks.
• Modem signal levels: A degraded coax cable or weak DOCSIS signal upstream of your modem causes speed drops that look identical to throttling on a speed test.

Throttling is confirmed when speeds are consistently low for specific services, the VPN test shows a clear improvement during affected periods, and wired Ethernet tests rule out your home network as the variable. All three conditions together — not just one — make a compelling case.