You know that feeling when you're trying to give a company money to stop taking your money, and they just won't let you?
Daniel sent us a long one, and it's worth hearing in full. He writes: I made a small bit of progress on my to-do list today, which I've never been way more behind on. I try to look at my utilities once a year, although more often than not, it happens once every few. I mentioned before how I had two cell phone subscriptions for many years to have a business line and a personal line. I finally got around to setting up a nice Twilio setup on the business line, which gives me one spare SIM card on the device for data. I was trying to cancel my subscription to the line I don't need anymore with my phone provider. They have a self-service online menu, and it lets you do literally everything except cancel the subscription. I went through the menu again and again and became certain that I must be missing something, so I asked my trusty colleague, Claude, to go through the menus. It confirmed there was no option to even request a call-back to disconnect, let alone self-service. I called the service line and went round and round phone trees and also couldn't find any options.
He goes on — and this is where it gets interesting. Fortunately, there is an Israeli website called netek.co.il. This exists for precisely this reason. Namely, service providers are making it so impossible for people to disconnect their own lines that somebody needed to create a product whose sole purpose is to complete these forms on exhausted subscribers' behalf. It is a completely free service, as far as I can deduce. Somebody created an automation to gather enough information to act on behalf of customers and then send a legally formatted disconnection request. But think about that. It takes a third-party service to format a request that the provider is doing everything in its power to hide away from you.
He notes that not all Israeli providers do this — some have proper self-service — but enough have taken the view that the easiest way to artificially increase retention is to make it impossible for people to leave. And then he lands the punch: I'm a thirty-seven-year-old who works in tech. If I couldn't figure out a way to disconnect, what hope does an elderly person or somebody who doesn't even know how to use a computer stand? It's not just poor business practice. I actually think that it's a form of robbery.
The word lands harder than people expect, but I think he's right in the ethical sense even if it's not theft in the legal one. What he's describing — I've seen this from the inside, actually. When I was still practicing, I had to help elderly patients cancel things. Grown adults in their seventies and eighties, completely defeated by a phone tree. One woman brought me her cell phone bill and a handwritten log of fourteen calls she'd made to cancel a tablet data plan she never used. Fourteen.
Fourteen calls.
And she was still being billed. So when Daniel says robbery, I hear the clinical version of that — it's extracting money under false pretenses. The pretense being that you entered this relationship voluntarily and can leave it the same way. That turns out to be false, and the company designed it to be false.
So today we're going to unpack the cancellation industrial complex. Why companies build these mazes, how a free third-party service in Israel had to be created just to navigate them, and what the FTC just did about it.
Let's start by looking at the specific tactics, because Daniel's experience isn't random — it's a playbook. The self-service portal that does everything except the one thing you need. That's not an oversight. Somebody sat in a product meeting and decided which menu items to include, and cancel subscription was deliberately left out.
The menu he went through three times. Then asked an AI to go through it. The AI also couldn't find it. At that point you start doubting your own sanity, which I think is part of the design.
It absolutely is. The term for this is "retention-by-obstruction," and it's a subset of dark patterns — user interfaces intentionally crafted to manipulate behavior. The FTC has been cataloguing these for years. Hidden menu paths are just one flavor. You also get the phone tree that loops without a disconnect option, the requirement to call during business hours when you subscribed online at two in the morning, the demand that you explain why you're leaving before the system will proceed.
The retention offer that resets the whole process.
Right. You say you want to cancel, they transfer you to a "loyalty specialist" who offers you three months at half price, and if you say no, sometimes the system just... doesn't process the cancellation. You have to start over. Each of these steps is engineered. The research on this is pretty stark — every additional step in a cancellation flow drops completion by ten to twenty percent. Add five steps and you've lost more than half your would-be cancellers.
So walk me through the economics. What does a percentage point of prevented cancellation actually mean in dollars?
Take a mid-size provider with a million subscribers at, say, thirty dollars a month. That's thirty million in monthly revenue. If their natural churn rate is three percent per month, they're losing thirty thousand subscribers. If friction cuts that to two percent, they keep ten thousand extra subscribers. That's three hundred thousand dollars a month. Three point six million a year. Per percentage point. The incentive to hide the cancel button is enormous — it's literally millions of dollars in retained revenue, and the cost to implement the dark pattern is a few product meetings and some deliberately bad UX design.
And the cost of getting caught, historically, has been zero.
Until very recently, there was no enforcement mechanism with teeth. Which brings us to netek.co.il. This is the part of Daniel's story that really got me. Someone looked at this broken system and built a free tool whose entire purpose is to send legally formatted disconnection requests on behalf of exhausted customers. It scrapes or knows the required fields for each Israeli provider's formal cancellation process, auto-fills them with the user's details, and fires off a demand that cites Israeli consumer protection law.
The fact that this exists at all is an indictment.
It really is. Think about the sequence. A customer tries the self-service portal — no cancel option. Tries the phone tree — loops endlessly. Eventually discovers, probably through word of mouth or a desperate Google search, a third-party website they've never heard of, and has to trust it with their personal details just to exercise a right they already have. That's not a market. That's a hostage situation with extra steps.
And Daniel's point about being a thirty-seven-year-old tech worker — if he needed Claude to confirm he wasn't missing the button, and then needed netek to actually send the request, the system has been designed to defeat everyone. It's not a skill issue. It's an accessibility and equity issue. Elderly users, non-Hebrew speakers, anyone with low digital literacy — they're effectively trapped.
I saw this in my practice constantly. Patients who were paying for services they'd tried to cancel months earlier. They'd given up. The company counts on exhaustion as a business model. And the most vulnerable customers are the most profitable ones in this equation, because they're the least likely to successfully navigate the maze.
So how does netek actually work under the hood? Is it just a form-filler, or is there more to it?
From what I can gather, it's essentially a civic tech project. It knows the specific data fields each Israeli provider requires for a valid disconnection request — subscriber ID, phone number, ID number, sometimes a specific form reference. It collects that from the user once, generates a legally formatted document that cites the relevant consumer protection statutes, and submits it through whatever channel the provider is legally obligated to accept. Email, fax, a web form — whatever the official path is. The key is that it formats the request in a way that's hard for the provider to reject or ignore, because it references the law explicitly.
And it's free.
Free. Which suggests it's grant-funded, donation-supported, or built by someone who was personally infuriated enough to donate their engineering time. It's the kind of thing that shouldn't need to exist, and its existence is the clearest evidence that the market has failed.
The market didn't fail. The market optimized for exactly this outcome. The failure is regulatory.
Fair. And that brings us to the other pole of this episode — what regulators are doing about it. Because October twenty twenty-four was a genuinely significant moment. The FTC announced its final "Click to Cancel" rule. The core requirement is simple: if you subscribed online, you must be able to cancel online. No phone call required. No visiting a physical location. The cancellation mechanism has to be as easy as the signup mechanism.
Which sounds obvious to the point of absurdity. Like, why did we need a rule for this?
Because companies fought it for decades. The FTC's rule also requires sellers to get explicit consent before charging for negative-option offers — that's the thing where you get a free trial and then it auto-converts to paid — and they have to send annual reminders for subscriptions that auto-renew. The rule was supposed to take effect in early twenty twenty-five, though there's been some legal pushback from industry groups. But the principle is now codified at the federal level in the U.S.
Compare that to Israel. Daniel mentions consumer protection law trailing behind Western standards. What's the actual gap?
Israel's Consumer Protection Law, from nineteen eighty-one — that's law fifty-seven forty-one for the Hebrew calendar folks — requires clear disclosure of cancellation terms. But it doesn't mandate a specific mechanism. For distance purchases, you get a fourteen-day cooling-off period. For ongoing services like phone lines, the provider can set its own cancellation process as long as it's "reasonable." And the bar for reasonable, in practice, is on the floor. Daniel's experience — hidden menu, looping phone tree, no self-service option — that apparently clears the reasonableness bar in Israel. Or at least, nobody's been penalized enough to change behavior.
So the law says "be reasonable," and the provider says "we put a phone number somewhere on page seven of the terms and conditions," and that's satisfied.
Right. And enforcement is the other half. Israel has a Consumer Protection Authority, but it's under-resourced and tends to go after the most egregious cases — outright fraud, dangerous products. A telecom provider making cancellation annoying but technically possible? That's not where the enforcement energy goes. Compare that to the EU's Consumer Rights Directive, which requires cancellation to be possible by the same medium as signup — very similar to the FTC rule. Australia's Telecommunications Consumer Protections Code requires providers to have a "simple, transparent, and efficient" cancellation process. California passed SB three thirteen in twenty twenty-three, which requires clear cancellation instructions and annual reminders for auto-renewing subscriptions.
California's been ahead on this for a while. Their auto-renewal law keeps getting tightened.
And each tightening is a response to companies finding new loopholes. The original California law said you had to provide a cancellation mechanism. Companies provided one — buried in a forty-page PDF. So the legislature came back and said no, it has to be clear and conspicuous. Companies made it clear but required a phone call during business hours for an online subscription. So the legislature came back again. It's an arms race.
Which raises the question of why Israel lags. It's not like the country lacks technical talent — netek itself proves otherwise.
I think it's a few things. Smaller market, so there's less regulatory attention from global watchdogs. The telecom sector is an oligopoly — three major players, and they tend to move in lockstep on practices like this. And there's a cultural dimension. Israeli consumer protection has historically been more reactive than prescriptive. The attitude is "we'll figure it out" rather than "let's write a rule that covers every scenario." That works fine when the problem is a faulty toaster. It fails completely when the problem is a systematically engineered dark pattern designed by well-paid UX researchers.
The netek workaround is a symptom of regulatory failure. It's citizens patching a hole the government should have filled.
And to Daniel's point about robbery — let's sit with that for a minute. Legally, it's not theft. You agreed to the terms of service, and buried in paragraph forty-seven it probably says cancellations must be requested by phone during business hours or whatever. But ethically, the company made a promise: sign up here, manage your account here, cancel when you want. The self-service portal is the embodiment of that promise. When the cancel button is missing, the promise was false. They're extracting money under a pretense they knowingly broke. The FTC's rule essentially codifies that this practice is deceptive and unfair. It's not saying it's inconvenient — it's saying it's illegal.
There's a distinction worth making here. Some companies make cancellation hard through incompetence — bad UX, understaffed call centers, legacy systems that don't talk to each other. That's a different problem. What Daniel's describing is intentional. Someone decided the cancel button wouldn't be there.
And that's the line the FTC rule draws. If you built a self-service portal for everything else, you can build it for cancellation. The technical capability exists. You chose not to deploy it. That choice is now, in the U.S. at least, a violation.
So given this patchwork of regulation and workarounds, what can someone actually do if they're stuck in a cancellation maze right now?
First, check if a service like netek exists in your country. These are popping up in multiple jurisdictions — civic tech projects that automate the legal cancellation request. In the U.S., there are a few. Some are free, some charge a small fee. The key thing they provide is the legally formatted language that signals to the provider that you know your rights and you're documenting the attempt.
Second option: regulatory complaint portals. The FTC has an online complaint system. Israel's Consumer Protection Authority has one too, though as we've discussed, enforcement is spotty. But a complaint creates a paper trail, and enough complaints about the same provider can eventually trigger investigation.
Third, and this is the nuclear option — credit card dispute. If you've attempted to cancel in good faith and the company continues to charge you, you can dispute those charges as unauthorized. Document your cancellation attempts. Save the emails, note the dates and times of phone calls, screenshot the missing cancel button. Your credit card issuer will ask for evidence, and a clear paper trail usually wins the dispute.
The credit card networks themselves have been getting stricter about this. Visa and Mastercard updated their merchant rules a couple years back to require clearer cancellation processes for recurring billing. They have a financial incentive too — chargebacks cost them money.
But all of these are workarounds. The real fix has to come from regulation with teeth. The FTC rule is a good start. The EU directive is good. But enforcement is everything. A rule without penalties is a suggestion. The FTC can seek civil penalties now, but the amounts need to be large enough that compliance is cheaper than the fine.
If you're a developer or a product manager listening to this — audit your own company's cancellation flow. Run what I'd call the grandma test. Can someone with low tech literacy cancel in under three minutes without making a phone call? If the answer is no, you're part of the problem. You may not have designed the dark pattern, but you're maintaining it.
The three-minute threshold is actually a useful benchmark. If cancellation takes longer than signup, something is wrong. Signup for most services is what — name, email, credit card, click. Thirty seconds. If cancellation takes ten times that, the asymmetry is the point.
Here's the thing I keep coming back to. The netek model — a free, automated, legally-informed cancellation service — is a fascinating example of civic tech filling a regulatory gap. Could this be replicated globally? Imagine a service that knows the cancellation laws and required formats for every jurisdiction, and just... sends the right letter. You enter your details once, it fires off the legally appropriate demand in the right language with the right citations.
There are startups circling this idea. The challenge is that the legal formats vary by country, by provider type, sometimes by provider. In Israel, netek works because there are only a handful of major telecom providers and the legal requirements, while weak, are at least standardized enough to template. Scaling that to fifty countries and hundreds of provider types is a real engineering problem. But it's solvable. The question is whether it's a business or a public good.
Probably both. Charge the people who can pay, make it free for everyone else. The people who need it most are the ones least able to navigate the maze themselves.
Let me steelman the counterargument here, because I think there's a genuine one. The objection would be: companies have a right to try to retain customers. A cancellation call is a chance to fix a problem, offer a better plan, address a complaint. Forcing a one-click cancel means losing that opportunity, and some customers who would have stayed with a better offer will leave unnecessarily. That's not nothing.
That's fair as far as it goes. The retention conversation has legitimate value in some cases. But the objection collapses when the retention mechanism is indistinguishable from a trap. A cancel button that triggers a "wait, here's an offer" screen is fine — as long as there's also a "no thanks, really cancel" button right next to it. The problem isn't the retention attempt. It's the asymmetry. If I can sign up at two in the morning in thirty seconds, I should be able to cancel at two in the morning in thirty seconds. The retention offer can be part of that flow. It can't be the entire flow.
And the data bears this out. When companies are forced to make cancellation easy, churn does tick up slightly in the short term. But it also forces them to compete on actual service quality rather than on who has the stickiest trap. The long-term effect is a healthier market.
Where is this all heading? The FTC rule is a big deal, but it's already facing legal challenges from industry groups arguing it exceeds the agency's authority. The current Supreme Court has been skeptical of broad regulatory power. It's not clear the rule survives in its current form.
Which is why I think the real action is going to be at the state level in the U.S., and at the EU level internationally. California's model has already been copied by several other states. The EU's directive is in force and being strengthened. Even if the federal rule gets watered down, the direction of travel is clear. Subscriptions are becoming the dominant business model for everything — cars, coffee, software, fitness, even clothing. As that happens, the right to cancel easily becomes a fundamental consumer right, not a nice-to-have.
The arms race continues, though. Even with the FTC rule, companies will find new dark patterns. Maybe the cancel button works, but it requires a ten-minute survey first. Maybe it works, but only after you watch a video about all the features you'll miss. The creativity of people trying to prevent cancellation is impressive in the worst way.
The survey thing is already happening. I've seen flows where you click cancel, and you get a multi-page exit survey that "helps us improve our service." Each page has a tiny "skip" link and a large "stay subscribed" button. It's the same pattern, just moved one layer deeper.
The fundamental fix has to be a simple rule: the cancel action must be at least as prominent, at least as fast, and at least as easy as the subscribe action. No qualifiers. No exceptions for "complex products." If your subscribe button is a big red button on the homepage, your cancel button better be a big red button in account settings.
That's exactly what the FTC rule says, in regulatory language. Whether it sticks — we'll see. But the norm is shifting. Daniel's experience, and netek's existence, and the FTC rule, are all part of the same story. The cancellation industrial complex had a good run. The lights are coming on.
If you've got a weird prompt about a system that's broken in a fascinating way, send it to us. The website is my weird prompts dot com. We might just build an episode around it.
Thanks to our producer Hilbert Flumingtop for keeping this show running. This has been My Weird Prompts.
We'll be back soon. Don't let them hide the button.
