MediaTalk | Season 4 | Ep. 10 -What Drives Most U.S. Broadband Churn: Not Pricing, Bundling or Speeds

View Full Transcript

Mike Reynolds: Hi, I'm Mike Reynolds, a senior reporter covering the media industry with S&P Global Market Intelligence Tech, Media and Telecom news team. Welcome to "MediaTalk," a podcast hosted by S&P Global, wherein the news and research staff explore issues in the ever-evolving media landscape. Today, we're talking broadband with a pair of Kagan-ites. John Fletcher is an S&P Global Market Intelligence Kagan analyst who specializes in multichannel and broadband. Brian Bacon's a research analyst on a consumer insights team within S&P Global Market Intelligence, Kagan. The discussion is going to touch on a number of key dynamics: churn, speed, and the reasons why consumers change providers. We're also going to focus a bit on the burgeoning fixed wireless access realm, and possibly a big IPO for Mr. Elon Musk. Anyway, John, how are we doing today?

John Fletcher: Mike, I woke up this morning hoping that I could find someone to talk to about, uh, broadband churn, and my kids had no interest, so I'm very glad to be here.

Reynolds: Okay. Brian, how you doing today?

Brian Bacon: I'm doing great, Mike. Thanks.

Reynolds: Okay. Alright. Let's start with churn. You guys teamed in a couple of recent articles that can be found on our Cap IQ Pro website. This isn't a case of people shifting out for a streaming service because a favorite show has completed a season run. Broadband churn, if I'm right, is relatively stable, John, but it did increase in 2025.

Fletcher: Yeah. Broadband, it's a utility now. It's just like water and power. And so churn is the monthly rate at which people cancel their service. And none of the big internet service providers report on the figure — they reported years ago. And Brian is a wizard with the surveys. He figured out a way with arithmetic. We have a giant survey we do twice a year to estimate it for these ISPs, and what we learned doing this math—and we've done it for about 18 months now, and so we finally published it, it looked pretty good. We got some feedback from the industry. It said, yeah, this is looking pretty close—is that the churn rates for fixed broadband are right around 1% per month. And for people like us that always talk about this stuff, that's phenomenal. That's really good. That's a really low churn rate. The only service that's better than that are smartphone providers. Okay. Their churn rate's at 0.9%. We saw a slight uptick towards the end of last year because there is a little bit of a price war going on in broadband right now, which we think explains a little bit of that, but still really low. And, to your point, it's not streaming service churn. Yeah. Like, for example, I signed up for Fubo so I could watch the World Series and, no, I canceled it a month or two later. Or at least I hope I did. I better double-check because sometimes I forget.

But those churn rates are in the double-digit percentage rates. It's not uncommon for those services to have churn rates as high as 10%. But yeah, for broadband, it's very low, very consistent.

Reynolds: Alright, Brian, fixed wireless. It seems like a relatively significant uptick last year, right?

Bacon: Yeah. We did see a churn increase among all three of the top fixed wireless operators. The top three, which we believe is driven by customers trying out the service. And then we've seen similar rates among satellite broadband operators. But I'll let John speak to that.

Fletcher: Yeah, if you think about satellite broadband, you gotta think of them in two buckets. You've got the legacy operators, which they call the geo-satellite operators. They've been operating for a couple of decades. They have satellites 10,000 miles up in the sky. And then you got the new whiz-bang LEO, low Earth orbit, right? Services. And these are—

Reynolds: Amazon, right?

Fletcher: That's going to be Amazon, that's Starlink. And those are the big operators in the —  I don't think Amazon's launched quite yet, but Starlink has been around since 2020. And in terms of churn, what we saw from the data was Starlink had a churn rate on par with the rest of the big guys, around 1%, but the legacy operators were double, 3x that amount of churn. That plays out in the numbers too. If you look at the subscription growth rates, it's Starlink's going up and to the right. And the other two operators have been shrinking.

Reynolds: Gotcha. Brian, which companies are at the very lowest end when it comes to churn? What'd you guys find out?

Bacon: Verizon Fios, their fiber home internet service, had the lowest churn rate based on the results from our Q3 2025 media census survey. And they've just completed their acquisition of Frontier, which is another operator that had a really low churn rate.

Reynolds: Okay. Is there, as you look at this, John, providers — a number of tactics in play, I guess price among them, but some of the things they're doing, there's some really long-term contracts available now, right?

Fletcher: It's gotten kinda wild. Like I mentioned, there's a little bit of a—I'm calling it a price war, but it's also a bundling war. In late ‘25, Starlink dropped rates to $40 per month in some areas. Compared to satellite broadband ARPUs monthly—what people pay per month—that's less than half. It's usually about a hundred dollars plus for satellite broadband. That's a big deal. You mentioned the—they're calling them five-year rate locks. Comcast launched one last year for $50 a month for five years that your price won't change. But then T-Mobile really blew everyone outta the water.

They did a 10-year lock for people in their Fiber Founders Club. Ten years?! Where are we going to be in 10 years? Am I going to have grandkids in 10 years? That's a pretty big change. And this is all reaction to fixed wireless really pulling the rug out from everyone in 2020 and changing the dynamics, and the operators are all scrambling a little bit to keep these customers to stick with them. Another thing we haven't talked about is bundling other services with the same operator. That also is a big deal right now.

Reynolds: Okay? Yes. Comcast and Charter in the bundling game.

Fletcher: They offer a phone service. They all have smartphone services. They partner with the tier one operators like Verizon, AT&T, and T-Mobile. They rent their networks and offer a smartphone service to their existing Comcast and Charter customers. So I looked at the data three years ago, about 17% of Charter and Comcast's subscribers were doing this. They were using a phone service from their operator, and now that number is 40%. This is just three years later. So they're doing really well with bundling mobile phone services with their existing broadband customers. And as anyone who studies churn knows, the more services you have on the same bill, the lower your churn is going to be. The more things you can attach, the stickier the bundle is for people. Customers tend to stick around when you have more services attached to the same bill.

Reynolds: And probably harder to detach as well.

Fletcher: Yeah. A lot of phone calls and emails maybe to make that happen, but…

Reynolds: Oh, maybe that's the unspoken element here, that no one wants to make that phone conversation and try to get it canceled. That's a good point.

Reynolds: Alright, let's turn our attention here to speed a bit. They continue to tick up. The share of broadband subscribers seeking gigabit speed, I think, remained flat at about 30%. But according to Kagan's Media Census online consumer survey data conducted, I believe, in the third quarter, the share of broadband subscribers feeding speeds between 500 and 999 Mbps has increased from 22% in '24 to about 27% in '25. And over the same period, the share with plans between 100 and 499 Mbps decreased from 44 to 40%. So a lot of changes there. Brian, your thoughts? What did you guys find out?

Bacon: Yeah, so to lay this out, in our Q3 media census survey, we asked several questions to determine which package subscribers actually receive from their ISP. And based on those responses and the limited insights that are provided by the operators themselves, we estimate the package penetration rates in the same kind of way that we've been doing that for TV packages among the top multichannel operators.

Okay. So using that process, we estimate that less than a third of subscribers receive gigabit speeds. The growth that we're seeing is, like you said, that shift to the people in that 500 meg to just under a gig sort of speed tier range.

Reynolds: Okay. And is there a sense, Brian, that—who's making that change? Younger folks or older people, as we all need broadband more than ever. Is that happening maybe with older demographics as well?

Bacon: Yeah. To help with comparing across the ISPs, I grouped the packages into four tiers: less than a hundred megabits a second, 100 to 499, 500 to less than a gig, and then a gig plus. Younger subscribers had larger shares in those highest and lowest two tiers. And cost is probably the deciding factor. If we thought about it in the same way that, like, the cable TV bundle—it was expensive, right? But it had all the entertainment you needed for the whole family, and it made sense for households of multiple people, households with children, things like that. And so when you start to look at it from that way, it's like how people are portioning out their monthly spend on things like internet.

Reynolds: And I guess in terms of that way too, if you don't have the financial resources at the moment, you're probably going lower, right?

Bacon: Yeah. Lower-income households represented a larger share of subscribers to the lower-tier plans, as you would expect. I think the interesting follow-up would be to find the order of priority. So do consumers put their wireless data plan first? Maybe. 'Cause we have seen that of people that are mobile only. So do they maybe do that first and then figure out how much broadband they can afford? Do they need to have gigabit speeds or something like that?

Reynolds: Yeah, I think maybe another thing too—I shouldn't say this as an old video guy—broadband is more important to some than video, God forbid. Anyway, Brian, as you look at the different players here, which company is perhaps exceeding the most in getting the higher gig customers?

Bacon: Yeah. Looking across the different operators, it really shows the differences. For example, Frontier had the largest share of subscribers with a gigabit plan. But their listed plans are almost all gigabit speeds. Okay. I think out of six of them, there's four of them that are gig plus, so most of your subscribers are going to end up in there.

Reynolds: That makes sense. Makes sense. How about at the other end of the spectrum? Who's selling slower speed packages?

Bacon: Yeah, you got operators like Charter. They've got a large base of subscribers, but they have the twofold challenge of trying to retain those subs, but also moving them up to higher speed and therefore higher price plans. Some operators are a tale of two systems, like AT&T. They've got a mix of fiber and also some legacy DSL in there.

Reynolds: All right. I guess it's interesting to see in terms of—Cox and Charter are supposed to come together at some point this year, so it'll be interesting to see how they work together in terms of their broadband and video and, I guess, mobile offerings. They also talked about, in some cases, one side of the company is better at some of the business offerings while the other has—they've got some features that they believe are compelling. So that'll be something to keep our eye on as well.

Alright. Let's see. As you said, Verizon and Frontier together—how's Verizon doing here, Brian?

Bacon: So Verizon, the share receiving gigabit speeds remained about the same at 37%. But we did add a two-gig plan option in the 2025 survey, which pulled about 5% out from that gig. So if you look at it year over year, it seems like that gig plan took a hit, but it's really just that they shifted up to the two-gig plan.

Reynolds: Alright, and AT&T?

Bacon: Pretty stable. Overall, there's some small movements from their Internet 100 plan to the 500 meg and one gig options, though.

Reynolds: Alright. We talked a little bit about Charter. What about the other big cable player, Comcast?

Bacon: After years of tracking their TV packages, I've learned that Comcast likes to change the names of their plans and have different names for basically the same plan in different markets. Okay, so it's difficult. So I had to change some of the answer options from the '24 to '25 surveys and try to cover all the bases when putting it together after fielding. But that's enough of my problems. In the end, there was some movement from the 100 to 499 meg tier.

Reynolds: Okay.

Bacon: Getting to that 500 meg to just under a gig tier, which went from 19% to 29% of Comcast subs.

Reynolds: Huh? Okay. Alright. Good for them. Alright, we've talked about churn and speed. We'll be talking about them again as we go into the next section here. John, why do people move from one provider to the next? What are a couple of the top reasons here?

Fletcher: It's funny you say the word move because that's it. When we went into this, we had all this churn information. We were able to ask these thousands of surveyed people who left their operator. We asked them why. We gave them a laundry list of reasons, and 19% of them—at the highest, this is the most common reason—is because they physically moved to another address where their current ISP doesn't operate. And the second most common reason is they left for a cheaper price. And speed doesn't come in until about fifth or sixth down on the list of reasons. And it's funny 'cause I'm here, I'm sitting here listening to you guys talk about broadband speeds. And I remember when the 56K modem came out and that was a big deal. I'm just chuckling to myself. And now the slowest tiers are a hundred megabits, just okay. Pretty wild how things have changed over the years.

Reynolds: I wonder again, John, you'd mentioned earlier the five- and ten-year plans that are out there. Will they travel to a new home with you?

Fletcher: That's a great question. If that operator's there, I'm sure there's fine print that explains details of that.

Reynolds: Oh yeah. You gotta check that boilerplate for sure. No doubt about that. Brian, what about service? That's gotta be up there, right?

Bacon: Quality of service issues was the third most selected. And it was especially important among former HughesNet subscribers — 31% indicating that's why they switched their ISP. Then you got a quarter of former WOW (Wide Open West) subs switched for that reason. So reliability is really important in the streaming world.

Reynolds: John, we had mentioned that cost is not the biggest factor in speed, but at the same time, those are considerations.

Fletcher: Yeah, I mean, quality service is a big one. But speed's kind of in the middle. And the interesting thing is, across the pundits and the executives, it's all about convergence. That's the buzzword right now. And that's what we talked about earlier—bundling lots of different services through the same provider. That's really important to the people who offer these services, but according to the people who left their operator, the bundle deal and promotional offer, those are the bottom reasons for why people would leave their current provider. So yeah, moving, speed, and price—and obviously quality of service—are way more top of mind with people as far as reasons why people leave their current provider.

Reynolds: Alright. We mentioned convergence. Are there some overall takeaways, John, you have in terms of this demographic group is leaving for these reasons as opposed to other demographic groups leaving for other reasons?

Fletcher: Yeah. Yeah. A couple of things that pop out: we look at a bunch of different demographic variables, which are all available in the reports we did, but older survey takers were more likely to switch for a cheaper rate, which, when you're on a fixed income, that makes a lot of sense, right? They're also more sensitive to price increases, so they left an operator because their current operator bumped up the price. Another thing that I thought was interesting, and I can relate with this one, is households with children are more likely to cite quality of service issues for reasons to churn. And if you think about anyone who has anyone under the age of 18 living in their house, they're on their smartphone, they have their iPad, and they're watching a streaming service all at the same time. That really can tax a network. So I'm assuming that's what's going on with quality of service for families. And then also, we also notice that higher-income households are also more likely to cite quality of service issues. Maybe they're more particular than other households. But yeah, there's a laundry list of different things to look at in terms of ethnicity, income, education, et cetera, for kind of what's happening with churn rates based on those variables as well.

Reynolds: Alrighty. Let's take a closer gander at fixed wireless access. At the close of the third quarter, Kagan's media census data has put FWA residential accounts across the top 10 US states at 8.1 million. Can you talk about that?

Fletcher: The media census data is excellent. We're able to get regional information for broadband customers. And as we've talked about before, the DNA of fixed wireless and satellite broadband is rural. That's the sweet spot. Connecting Aunt Myrtle on her farm to watch her cat video—she couldn't do with that 56K modem. How are we going to get Myrtle high-speed internet? Fixed wireless came up and then satellite. And what we've seen is it's evolved over time. Rural is definitely a sweet spot for fixed wireless and satellite broadband. But what we're seeing is there's a lot more customers in suburban and urban areas. But in terms of fixed wireless, specifically looking at this data, you see the top three states are Texas, Florida, and California. Those are some of the most highly populated states, sure, as well. But if you think about—just think about those states in your head—they're pretty, they're geographically large. You definitely have huge cities, but you also have a lot of people living in rural areas there. So that also lends well towards fixed wireless. And it's popular in the Western states, in the Midwest, and in the South. Especially where we have rural households, lower-income households. Fixed wireless still does really well there, but what we're seeing from T-Mobile is a lot of their fixed wireless customers are coming from suburban areas and urban areas. And the spectrum that T-Mobile has is that 2.5 gigahertz. They have a hundred megahertz worth of it — it's the old Clearwire spectrum. And it's — if you can imagine like a giant freeway that's empty, that was 2.5 gigahertz in 2020, and now they're starting to populate it with cars, and it's a giant empty freeway, whether it's in a rural area or it's in an urban area, that 2.5 gigahertz that they have. And I believe Verizon has a chunk that they're using for fixed wireless in the C-band, so we're seeing more suburban and urban users for sure with fixed wireless, which is making people, old-timers like me, scratch my head just a little bit for fixed wireless. But it's just—it's attractive to everyone, not just these rural households.

Reynolds: You guys should check out the articles on Cap IQ Pro. It's interesting, as John just talked about, the general perception—it was rural, mountainous, and then if you look at states four and five here, or New York and New Jersey, so again, this is suburban or urban as the case may be.

Fletcher: That's the head-scratcher. I really want to do a follow-up question with people in New Jersey. Like, in my head, I don't picture a lot of rural areas in New Jersey, so I'm very curious about where those households are for fixed wireless in New Jersey.

Reynolds: Alright, let's finish up here with SpaceX. John, perhaps doing an IPO maybe as soon as this month, but expect it to happen sometime this year.

Fletcher: Yeah, it's rumored to happen this year at some point, whether they do a reverse IPO through Tesla or do an official new IPO, it's going to be the biggest IPO in history. And I did some back-of-the-envelope math just looking at Starlink subscribers, and they're about—this year they should generate over $10 billion in revenue from recurring subscription revenue for broadband services alone, excluding what you pay for the antennas and that kind of stuff.

Reynolds: Okay. So it's a pretty big business for them already.

Reynolds: And I think they doubled their subscriber base last year, up to about 9.2 million. A lot of that growth coming from our nation?

Fletcher: Yeah, they doubled last year. So they did two things recently. They dropped that price, like I mentioned, and also they took off their waiting lists across the world. They're gone. They're saying you're no longer on the waiting list. You can use our service. I think this next year is going to be gangbusters for Starlink. I think they're going to have an even better year. But that back-of-the-envelope analysis I did, I was very conservative about their growth rates. And so they might be making a lot more than $10 billion in subscription revenue from broadband alone. And they also have a different revenue source — a lot of these smartphones are connecting to the satellites, I'm sure some of you have noticed. And so that's also another revenue stream for LEO satellite operators as well.

Reynolds: Alright. A lot to look at. A lot to follow. That concludes this episode of “MediaTalk.” I want to thank our guests, Brian and John, not only for keeping us abreast of key data points, but sharing your insights about some of the dynamics that are shaping the broadband industry here in the US. We appreciate it, guys.

Fletcher: Anytime, Mike.

Bacon: Yep. Pleasure to be here.

Reynolds: Alright, thanks, guys. And this is Mike Reynolds. Thanks to all of you for listening. We'll catch up soon on the next installment of “MediaTalk.”