Router Blinking logo Router Blinking
How-to guide

Verizon speed test

Verizon home internet customers — Fios and 5G Home — plus anyone using the public test at verizon.com/speedtest/. The My Verizon app speed-test steps cited are documented for 5G Home; mobile-device testing follows Verizon's wireless-device guidance.

By the Router Blinking editorial team · Last updated 2026-11-01 · How we check facts

Short answer

Run the browser test at verizon.com/speedtest/. Verizon describes it as a network test: for signed-in customers it measures from the router to the nearest network peering point, which is not the same measurement as Ookla-style device tests. For best results Verizon says to connect by Ethernet, disconnect VPNs, close streaming apps and idle the other devices. 5G Home customers can also run the test in the My Verizon app.

A network number, not a device number

Verizon states WiFi strength and device capabilities mean results are often lower than plan speed, and that its WiFi test does not check WiFi signal strength. A low WiFi number is not evidence of a line fault — and do not factory-reset a gateway to chase one.

#1 Verizon’s test measures the network, not just your device

Most speed tests answer “how fast is this device’s session right now.” Verizon’s own FAQ draws the distinction explicitly (S45): the Verizon speed test is a network test — for an authenticated customer it measures “directly from your router to the nearest network peering point”; for non-customers it still measures to the nearest Verizon peering point. Ookla, by contrast, “runs the test from your connected device, be it phone, laptop or tablet, to the nearest server run by Ookla.”

That single sentence changes how you should read every number on this page:

  • A Verizon test result largely bypasses your home WiFi. It is closer to the question “is the line delivering my plan?”
  • An Ookla-style test always includes the device and its WiFi path. It answers “what is this device actually getting?”
  • The two numbers are both real and will not match. Verizon’s own page notes the direction of the gap: “factors like wifi strength and device capabilities mean that results are often lower than plan speed” (S45).

So the working pattern is the same one documented for other providers: the provider’s test for the line, a wired or device test for the in-home half, and a fixed method for anything you want to compare over time.

#2 Run the browser test at verizon.com/speedtest/

The tool lives at verizon.com/speedtest/ and runs in the browser for any visitor — no sign-in required, though authentication is what switches the measurement into the router-to-peering-point mode described above (S45). Verizon’s page frames it simply: “Test your internet speed,” with download, upload and latency readouts.

Verizon’s own accuracy checklist, from the same page’s FAQ, is worth following to the letter:

  1. Connect your device directly to the router using an Ethernet cable, if possible.
  2. Disconnect from any VPNs.
  3. Avoid using video or music streaming apps while the test runs.
  4. Turn off other devices — the FAQ says yes, limiting internet use during the test, including idle devices, gives a more accurate reading.

That list matches the pre-test checklist on Verizon’s slow-speed support page almost word for word: Ethernet direct to the router, VPNs off, no other device connected, no streaming apps (S48). When two Verizon sources agree, it is a safe routine.

#3 Testing from the apps

Verizon splits app-based testing by product, and the distinction matters:

5G Home — My Verizon app. Verizon documents the exact path (S47):

  1. Open the My Verizon app and tap the Home tab.
  2. If needed, tap the 5G Home tab at the top.
  3. From the Home device section, tap the device to manage.
  4. Tap Wi-Fi networks, then Run speed test.

Two conditions are printed in the same article: you must be connected to your 5G Home Wi-Fi network to run the test, and “5G Ultra Wideband speeds range from 300 Mbps up to 1 Gbps.” The same KB warns that 5G Ultra Wideband “may fall back to 4G LTE speeds” based on location (S47) — a cellular characteristic no wired plan has, and the single most important thing to know before judging a 5G Home speed result.

Mobile devices — wireless considerations. Verizon’s wireless-device guidance adds the background-app warning: apps like OS and antivirus updates or file-sharing tools “can consume a significant amount of bandwidth” while you test (S46), plus the same 5G-to-LTE fallback note for cellular testing.

#4 Reading the results: download, upload, and the WiFi caveat

Verizon’s definitions on the test page are plain (S45): download speed “shows how long it takes to download or receive information to your computer or device from the internet”; upload speed “measure[s] how long it takes to send information from your device or computer through to the internet.”

The page’s most useful honesty is in its caveats:

  • Results are “often lower than plan speed” because of WiFi strength and device capabilities (S45).
  • Asked whether its test also checks WiFi signal strength, Verizon says no — signal strength is a different check from internet speed (S45).

Practically: a WiFi result measures the weakest link in that moment — the radio path to the room you are sitting in, the device’s own radio and processor, and the line. A wired run removes the first two. Verizon’s slow-speed page names the in-home variables explicitly, including one most guides forget: “the quality and condition of the wiring which connects your computer(s) to the Verizon network” (S48).

#5 The benchmark Verizon itself cites

Asked “what are good upload and download speeds,” Verizon’s answer does not point to a Verizon tier — it cites the FCC: “The FCC’s benchmark for high-speed fixed broadband specifies download speeds of at least 100 megabits per second and upload speeds of at least 20 megabits per second” (S45). The same answer adds the household framing: up to 21 connected devices in the average home, so bandwidth needs scale with the house, not just the plan.

That gives you a two-step reading for any Verizon result:

  1. Against the FCC benchmark — is the connection adequate at all (100/20)?
  2. Against your plan — is the wired number close to the tier you pay for?

For what specific activities need, the FCC activity table — browsing at 1 Mbps, HD streaming at 5–8, 4K at 25, online multiplayer at 4 — is the reference Verizon’s own page points toward.

#6 Fios vs 5G Home: what to expect from each

Verizon sells two very different home products, and the sources describe their speed behavior differently:

  • Fios is Verizon’s fiber product. The test page’s server-to-device wording and the network-test description describe a stable, wired path; Verizon’s own materials treat the connection to the network as the consistent part, with WiFi as the variable.
  • 5G Home is cellular. The documented characteristics are a 300 Mbps–1 Gbps Ultra Wideband range and possible fallback to 4G LTE by location (S47). A 5G Home result that dips across the day is consistent with that documented behavior — the connection shares the cell network, and Verizon’s factor list includes network congestion and outages (S48).

If you are on 5G Home, run the app test (S47) where the router actually sits, note the time of day, and compare several runs before drawing conclusions. If you are on Fios, the wired Ethernet run is the number that means “the line.”

#7 If the result is low: Verizon’s documented path

Verizon’s slow-speed page provides the sequence (S48):

  1. Rule out the obvious share of the problem first — WiFi-versus-wired, running applications, device count, and the wiring inside the home are all named factors.
  2. Use the Guided Solutions Tool. Verizon points its slow-speed readers there to identify and resolve Wi-Fi issues, with technician scheduling built in if needed.
  3. Re-run the speed test under the documented conditions — Ethernet direct, VPN off, other devices idle, no streaming (S45, S48).
  4. Check the equipment’s own signals before restarting anything. A gateway mid-boot or reconnecting will fail tests for reasons no speed routine fixes — the CR1000A light guide, the G3100 chart and the Fios ONT blinking guide cover those states, and the outage guide lists Verizon’s official check (the Guided Solutions Tool).
  5. Then, and only then, a single restart — and never a factory reset for a speed number; nothing in Verizon’s speed documentation calls for one. The restart versus factory reset guide explains the difference.

#8 What If Scenarios

Q: Why is my Verizon result different from Ookla’s? Because they measure different paths, by design. Verizon measures router-to-peering-point for authenticated customers (a network test); Ookla measures device-to-nearest-Ookla-server (S45). Expect the Verizon number to be higher and more stable, and the Ookla number to track your WiFi conditions.

Q: My 5G Home speed drops in the evening. Is it broken? Cellular capacity is shared, and Verizon documents both the congestion factor and the Ultra-Wideband-to-LTE fallback behavior (S47, S48). Test several times of day with the app, note the pattern, and take that log to support rather than a single slow result.

Q: The WiFi test is slow but Ethernet is fine. Do I need a new router? Not necessarily — first check placement and distance, which Verizon names as the usual WiFi culprits (S45), and remember its own note that the speed test does not measure signal strength (S45). The Verizon router light guides help rule out fault states before you shop.

Q: Should I run the test while other people are home? For a number you plan to act on, no: Verizon’s guidance is to idle other devices and stop streaming during the run (S45, S48). For a realistic picture of the household at 8 pm, the opposite — but label the two kinds of runs differently; they answer different questions.

Review note (2026-11-01): the peering-point distinction, the FCC 100/20 citation and the 5G Home steps are from Verizon’s live pages and KBs on this date; the slow-speed page footer dates to 2020. Re-verify verizon.com/speedtest/ behavior and KB steps at each review — provider test implementations change without notice.

Sources

Last checked 2026-11-01. Report a correction.