For most the dozen years I’ve been writing this blog, the biggest cable companies accounted for almost all of the growth in broadband customers. Quarter after quarter, and year after year, the big cable companies were the source of almost all new net broadband customers.
This started to shift a few years ago when FWA cellular home broadband from T-Mobile, Verizon, and more recently, AT&T entered the scene. For the last couple of years, almost all of the net broadband growth in the country came from the FWA technology and these three carriers.
It’s clear that we’re now entering a new stage the industry where cable broadband losses are accelerating, where FWA growth hasn’t slowed, and where the big telcos are growing again because of their expansion of fiber.
Consider the following statistics that show the net change in broadband customers over the last year, and for the latest quarter, for the largest cable companies, largest telcos, and FWA carriers. There are a few big companies missing from this comparison like Cox, Mediacom, and Windstream, since those companies are privately held and don’t publicly report customer counts.
These numbers show that telcos other than Lumen are growing again. The numbers for telcos don’t tell the whole story because net customer changes in the table include both DSL losses and fiber gains. For example, during the last year, AT&T added over 1 million customers to fiber.
These trajectories don’t bode well for the big cable companies. There were a lot of predictions made last year that FWA growth would slow down, and that doesn’t seem to be the case yet in 2025. The telcos are all picking up steam in terms of adding fiber customers. It’s going to be interesting over the coming years to see how the biggest cable companies fare in battling everybody else. Charter has decided to fight the trend through the merger with Cox. We’ll have to wait and see what the rest have in mind.
I’d like to see someone chart the price vs growth numbers without including speed as a factor.
My perspective is that this is essentially all price. Cablecos don’t come down in price without some inventive (ACP basically), FWA is trying to steal customers so their price is low, and fiber services are doing more or less $50 services to gain customers as well. The theme I see is that it’s all price. Most people aren’t picking fiber because it’s better, they’re picking it because they get 1-3 years at $50 and a free install. FWA they’re getting for 35-45 bunded with the cell plan with no install fee. Cable is $85+ for many people and the cablecos like to do incremental price hikes that infuriate people.
We see people coming from cell based FWA because of unreliability, but that means they switched to it in the first place and almost all of them did it based on price.
Cablecos have to decide when it’s time to drop prices and lose that revenue, or just lose customers until they’ve lost the revenue anyway.
Ha! Everybody would love that. Cable companies are already handing out a lot of low prices in the form of promotions and special pricing to keep customers. But this is proprietary and nobody knows the extent to which they do it. Big fiber guys and the FWA guys now doing the same.
I think in major markets that we’ve already reached the point where price is king, but the general public has no idea what the magic target price ought to be.
Speed is a factor for some folks, many of who will pay extra for gigabit speeds regardless of the cost, but many other people are happy with ‘good enough’ speed as long as it works when they want to use it.
That’s the cable company dilemma – a big chunk of customers buy based on speeds, and cable companies have slow upload, while another big chunk of the market buys by price. It’s hard as hell to market to both of these groups along with the folks in the middle who aren’t paying that much attention.
Another comparative number I would like to see is the error rate or speed of through-put. My neighbors and I have experienced a noticeable difference between FTTH (FiOS) and the cable provider, and T-Mobile FWA… Yes, T-Mobile FWA is cheap, but the transmissions were slowed by what seemed like a high rate of errors in transmission.
Is this data available anywhere?
@Ronald Isaacson, they definitely don’t want to talk about per-packet reliability or jitter in cellular FWA. The entire deployment model is at issue, ie in-home devices with dramatically reduced signal levels as a result. LTE/5G is reasonably good at handling those things but often by a robust error correction and retransmission scheme.
I operate traditional FWA as well as LTE FWA, there’s a SUBSTANTIAL difference between the two on OTA error rates and retransmissions and jitter and every other metric you can measure.
As noted yesterday, the AT&T wireline data included in the above table is incorrect. That figure includes FWA results, and therefore FWA is doublecounted in the table.
AT&T can be estimated to have lost (71,000) wireline connections in 1Q25 and lost (399,000) wireline connections over the last 12 months.
You may be right, but I can’t see how to get to your number from the AT&T 1Q 2025 financial report. AT&T has the most obtuse reporting of customers of any of the big ISPs. Here is what they released along with the announcement of 1Q 2025 earnings:
Fiber net additions 261,000
Non-fiber net additions (I assume DSL) (124,000)
Net of the two 137,000
FWA net additions 181,000
They also made the statement in the earnings: “Consumer Wireline also delivered positive broadband net adds for the seventh consecutive quarter”, which I read as referring to the 137,000 net wireline additions.
There is no source I can find that verifies their total number of wireline broadband customers as a way to doublecheck the math.