New minimum speed for US broadband connections

What should be the new minimum speed for "broadband" in the U.S.?

This is the list of past minimum broadband speed definitions by year

year speed

1999 200 kbps in both directions (this was chosen as faster than dialup/ISDN speeds)

2000 200 kbps in at least one direction (changed because too many service providers had 128 kbps upload)

2010 4 mbps down / 1 mbps up

2015 25 Mbps down / 3 Mbps up (wired)
         5 Mbps down / 1 Mbps up (wireless)

2021 ??? / ??? (some Senators propose 100/100 mbps)

Not only in major cities, but also rural areas

Note, the official broadband definition only means service providers can't advertise it as "broadband" or qualify for subsidies; not that they must deliver better service.

At $50/month or less?

Maximize number of households of all demographic groups.

I’m not in the US but in Canada it’s been 50/10 since 2016 and we’re just “almost” there yet. IMO the target should have been more like 100/30 or even 50 of upload.

100/100 might be a bit short sighted considering it’ll take years to accomplish the necessary last-mile/distribution upgrades in rural areas.

I’d love to see 100/100, but I don’t see it happening anytime soon … especially for $50.

I pay $150/month for 300/8 at home and that’s the best upload I can get where I live … in a major city.

I get about 23/6 Mbps for $50/month here in Silicon Valley from my ATT DSL line.

I pay $150/month for 300/8 at home and that's the best upload I can get
where I live ... in a major city.

Outside of DC area, < $50/mo at 200/200. Because there is competition in the local market between Verizon and Cox.

Once OneWeb and Starlink become a thing, a lot of providers that have enjoyed a monopoly may have competition. This is on top of 5G in dense markets. Competition can drive the market faster.

     - Ethan

I don’t think it needs to change.

I second Mike.

I’d be interested to understand the rationale for not wanting to change the definition. Is it strictly the business/capital outlay expense?

At least 100/100.

We don’t like selling slower than 10g anymore, that’s what I’d start everyone at if I could.

FCC Definition of “broadband Internet” always lags behind the reality of actual user needs, by about a decade.

Various sources show that Internet bandwidth consumption increases at about 29% CAGR.

If you extrapolate from the previous increases and intervals of the FCC’s changes, the definition of broadband should be a minimum of 100Mbit/100Mbit in 2021.

When I hear incumbent providers insisting that 25/3 is still good enough, my answer is: “sure, I can agree with that, if you can do that PER DEVICE in the home.”

They don’t like that argument.

The only reason 25/3 is still the FCC definition is because of lobbying by those that are still limited by twisted pair copper infrastructure.

25/5 is very usable today. Defining acceptable latency and packet loss would be a plus.

Don’t think it needs to change? From 25/3? Telehealth and WFH would like to talk with you.

There’s very few things more draining than a conference call with someone who’s got a bad connection.


Abhi Devireddy

I fear there are too many areas that are still limited by *dsl
technology so trying to define a certain minimum for upstream
transmission rates is problematic. (Also a pet peave of mine since it
makes moving video and audio project files areound a PITA.)

Personally, I think we're probably best sticking with the current
figures until what is widely available as a top end service begins to
reflect different figures and I don't see that that has happened yet.


There are millions of people that have 0 mbps (or dialup, satellite, etc) and they can’t function day to day like everyone else in town.

Changing the definition of broadband to yet again, to a faster speed will do nothing for these people except slow the pace at which they get connectivity. Why do people “in town” need to go from 25/3 to 100/10 when we really should be focusing on the people with nothing?

Changing the definition to 100/100 kills every technology except for fiber. Every single cable internet connection suddenly becomes “not internet”. Do we really want another AT&T that ends up with all of the primary last mile technology to all the major cities again?

I’d like to see your data that backs up the statement that the broadband internet definition (of 25x3) lags behind actual user needs by a decade. Here are the TOP 4 residential users last month:

up down total
Fixed Wireless 25x4 93.628 3105.440 3199.068
Fixed wireless 25x4 290.000 2763.089 3053.089
Fiber 500 63.563 2063.782 2127.345
Fiber gig 24.752 1562.230 1586.982

Two wireless customers did MORE than two fiber customers. The wireless are on 25 meg and the fiber are on 500/1000 mbps plans.

The top wireless subscriber is DOUBLE the download usage of the gig fiber house. The highest upload user was wireless, which happens to be FIVE TIMES the highest usage of the fiber customer.

Here is an image comparing the top wireless and top fiber customer usage:

Please let me know what your data looks like, I would love to compare.

I’d be interested to understand the rationale for not wanting to change the definition. Is it strictly the business/capital outlay expense?

Clearly not a residential mass-market service.

“Bad connection” measures way more than throughput.

What about WFH or telehealth doesn’t work on 25/3?

Even among network operators, many people are disconnected from reality.