Nine Cents a Minute to Nowhere: The ISPs Still Selling Dial-Up and the People Still Buying It

Somewhere in eastern Kentucky, a 74-year-old retired schoolteacher is checking her email over a 56K connection. She's been with the same ISP since 1998. The company has changed names twice, been sold once, and survived one near-bankruptcy, but it never stopped offering the plan she's on. It costs her $9.95 a month. She has no intention of switching.

This is not a story about people who don't know any better. This is a story about the economics, the stubbornness, and occasionally the genuine logic behind a corner of the internet that most of the tech world declared dead before the first iPhone ever shipped.

Who's Actually Keeping the Lights On

There are, depending on how loosely you define the term, somewhere between a dozen and two dozen ISPs in the United States that still list dial-up as an active, purchasable service. Not archived. Not legacy-only. Actually available if you call the number on the website and ask for it.

NetZero and EarthLink both technically still offer it, though their dial-up plans feel more like museum pieces kept behind glass than real products anyone is expected to buy. The more interesting cases are the regional operators — the ones running out of strip mall offices in places like Beckley, West Virginia, or Bemidji, Minnesota, who never made the strategic decision to kill dial-up so much as they just never got around to it.

Talk to the owners of these outfits and you hear a consistent story. Killing the service would cost them something. Not much, but something. The infrastructure is already paid off. The modems in the rack are ancient but functional. Maintaining a dial-up tier requires almost no ongoing investment once the original equipment is amortized. And the customers who are still on it? They're almost never calling support.

"Honestly, they're my easiest accounts," said one operator in rural Tennessee who asked not to be named because he wasn't sure his business partners would appreciate the publicity. "They pay on time, they don't complain, and they definitely don't call me at 11 p.m. because Netflix is buffering."

The Customer List Is Not What You'd Expect

Spend enough time poking around the dial-up holdout community and you start to notice the customer base breaks into a few surprisingly distinct groups.

The first is the genuinely underserved rural population — people in areas where broadband infrastructure simply doesn't reach. The FCC's maps have gotten better at acknowledging these gaps in recent years, but the coverage problem is real and ongoing. For a household where the nearest cable head-end is twelve miles away and fixed wireless is blocked by terrain, a phone line that connects to the internet is not nostalgia. It's infrastructure.

The second group is retirees on fixed incomes who established their internet habits in the late nineties and have no compelling reason to change them. They check email. Maybe they read the news. The connection is slow by modern standards but plenty fast for plain text, and the monthly cost is a fraction of what a broadband plan would run.

The third group is the one that makes this feel like a story worth writing for a site like this one: the enthusiasts. People who maintain active dial-up accounts the way someone might maintain a vintage car — not because it's the most efficient way to get from point A to point B, but because there's something in the ritual worth preserving. Some of them are running BBS software. Some of them are just chasing the sound of that handshake.

The Economics of a Dying Product That Won't Quite Die

From a pure business standpoint, dial-up is a remarkably low-cost service to offer if you already have the pieces in place. The phone lines belong to the local exchange carrier, not the ISP. The modems are long since depreciated. Bandwidth consumption from a 56K customer is so negligible by modern standards that it barely registers on network operations dashboards built to handle gigabit traffic.

The margin on a $9.95 monthly plan is thin but real, especially if you're running a few hundred of these accounts with essentially zero support overhead. One small ISP owner in the Midwest — who has about 300 active dial-up subscribers alongside a much larger DSL and fiber customer base — described it as "found money." The accounts exist. They bill automatically. The customers almost never churn because they're not being actively marketed to by competitors.

The risk is infrastructure decay. Modems fail. Analog phone networks are being quietly decommissioned in more parts of the country every year as carriers push customers toward VoIP and fiber. Some ISPs have already lost the ability to offer dial-up not because they chose to stop but because the underlying phone infrastructure that made it possible simply went away.

What They're Actually Using It For

This is the part that genuinely surprised us when we started making calls.

A surprising number of active dial-up users are running software that was designed for exactly this kind of connection. Terminal clients. BBS front-ends. Email programs that compress attachments and queue outgoing messages for batch transmission. In some cases, people have set up their home networks specifically so that one machine — usually something old enough to have a physical modem bay — handles dial-up duties while the rest of the house runs on broadband.

It's a deliberate separation. The slow machine is for slow things. Correspondence. Reading. Connecting to systems that were built for exactly this speed. There's a certain clarity to it that's hard to argue with if you've ever tried to read a long email on a phone while three apps are fighting for your attention in the background.

Others are using it for remote access to systems and equipment in locations where no other connectivity option exists. Agricultural monitoring setups. Weather stations. Remote sensors on equipment that was installed fifteen years ago and was never designed to talk to anything but a phone line.

The Handshake Nobody Wants to Be Last

What's strange about the dial-up holdout situation is that it doesn't feel like it's heading toward a dramatic conclusion. There's no scheduled sunset. No industry-wide announcement. It's more likely to end the way it's been ending for twenty years — one ISP at a time, one customer at a time, as phone infrastructure disappears and the last few modems in the rack finally give out with no replacement parts available.

The schoolteacher in Kentucky will probably keep her account until one of those things happens. The enthusiasts will probably spin up some kind of emulated analog network to keep the handshake alive even after the real one goes quiet. And the regional ISP operators will keep cashing those $9.95 checks until the math stops working.

It's not a revolution. It's not even really a movement. It's just a stubborn little corner of the network that never got the memo that it was supposed to be gone by now.

Somehow, that feels exactly right.