[idworks1]

Anecdata (only because I don't have access to the data anymore). Customer satisfaction is so much higher when they get a one click unsubscribe. In fact, when the friction is so low, the customer is likely to start the subscription back.

I say this as someone who worked in customer service automation. The worst customer satisfaction score with lowest rate of re-subscription is from companies that make it hell to unsubscribe.

I've seen customers send messages like "Cancel and refund immediately!" Since our response was ai driven, we cancel and refund no questions asked in less then a minute (we do fraud check in the background). Many times you get a response back from the customer apologizing for their tone and praising the product. Some of them restart the subscription a cycle or two later.

When you make it hard to cancel, you lose customers on the long term. Make it easy, in fact make it friendly. Unless you are selling a shady product, there is no reason to believe customers won't come back.

Edit: typo

[hirsin]

Speaking to this myself - I was blown away by Masterclass letting me do this, after I forgot it would renew. I responded to their "hey you've been charged" email with something like "I forgot about this and didn't mean to renew". In perhaps ten minutes I had a cancelation notice and refund.

Way more likely to get their stuff again. Bonus points for not sending the bill from some faceless "no reply" address.

[bertil]

For what it’s worth: I’ve been considering trying out Masterclass because their content seem great, but worried that something will come up and I’ll be locked into a yearly contract or something, so I never tried.

Your comment made me confident that I should.

[ChicagoBoy11]

I think it really depends on what you want to get out of it. There's content that is just a treat and fun to watch, but the idea of it being a "class" is rather disingenuous. Like, are you really going to become a composer after watching 6 hours of content? Or a rock climber?

But the stuff that is set up where you can have a chance at succeeding by practicing what you are seeing really does work. I was gifted a subscription and because of it watched the Thomas Keller series of cooking. I had never cooked before, but I have tried just about every recipe of the series a few times over, and no joke I've been able to produce truly REALLY fine dining experiences at home. Those lessons, for instance, are super didactic, build cooking up from first principles, show you all the steps, etc. It is fascinating content.

And then, for instance, watching some of the other stuff, like rock climbing, for example, can also just be fun. So, all in all, I think depending on what your profile is and what your expectations of it are it can actually be worth it.

[philistine]

If Masterclass were scummy with their cancellation policy, OP would not be able to see for themselves. A frictionless policy allows people to experiment with services. My wife and I subscribe to one streaming service at a time because of that.

[TheTrotters]

I had a great experience with Audible. I switched from Audible UK to Audible US but forgot to cancel the former. After six months I noticed I had 6 credits I didn't want. I sent one short e-mail explaining what happened and within 15-30 minutes I got a response that said they'll refund everything.

[Wowfunhappy]

I’m glad you had a good experience, but IMO the way Audible works is a mess of dark patterns to begin with. If you stop paying for one month you loose all your credits, so when I’m not reading as much I end up in a bind where I can’t cancel, but if I don’t cancel I get deeper in a hole.

You can pause a subscription, but only once every couple years and only for 90 days, and the option is hidden.

[tashoecraft]

I will say that maybe they’ve updated things, because I went to cancel and it said “you have x amount of credits, would you like to pause for y time in order to keep them”. In no way was it hidden. I hate the credit system and the fact that they go away, but pausing was encouraged

[Wowfunhappy]

Right, it doesn't show the option until they basically think you're about to cancel anyway.

[myaccount80]

Once I unsubscribed from Audible I started receiving email every 1-2 weeks with 60-70% discounts for couple of months. So I’ve been subscribing/unsubscribing to keep getting nice discount

[sreitshamer]

A “you’re about to be charged” email would be friendlier and give the customer a feeling of more control. That’s what we do. (We of course will also refund the charge if you email us after it happens.)

[CaptainJustin]

I love that idea. Can you imagine we had a standard around this? For example: Monthly payments - notification could go out >24 hours before the transaction will occur.

[hirsin]

They sent those too FWIW, I had just ignored them.

[mitchdoogle]

Email is a two-way medium. No reply addresses are a terrible dark pattern

[Joeri]

There is this common perception of companies as if they are entirely rational organizations, and every policy that we don’t like exists because it is profitable and benefits the company at the expense of the customer. But sometimes bad policies are just bad, they benefit no one, and they exist for dumb reasons. Maybe call to unsubscribe is one of those policies.

[somethingwitty1]

An opposite statement can be said with the same amount of authority though: There is a common perception that companies only create policies we don't like through accidents and unforeseeable outcomes, not by specifically crafting policies to benefit the company. But sometimes bad policies are malicious and designed to maximize profits, even at the expense of long-term profits and customer retention. Maybe call to unsubscribe is one of those policies.

As someone that has worked (briefly) for a company that operated in this fashion (and being a partial owner of one that the CEO tried to shift to this model...we got the board together and fired him), it is not an accidentally bad policy. It is actively discussed as a way to squeeze out an extra pay cycle (and often more) of payments. In recorded meetings or audited channels (such as email) or even PR releases, you are guided to discuss it as a "personal touch with the customer" and to help "lost customers" resolve the issues rather than cancel. You even try to convince your employees/engineers that is the reason. But when it is face-to-face conversations, the discussions are around the dollars and squeezing out as many pay cycles as you can. I know I was being a bit cheeky with my first paragraph, but this is definitely not one of those "whoops, we didn't think this through" kind of policies. If it were, the policy would have changed without the FTC or laws being needed.

[MathMonkeyMan]

There is a third option.

1. "Whoops, we didn't think this through."

2. This makes us more money in the end, that's why it's so pervasive.

3. It's difficult to correlate "making more money in the end" with our cancellation policy, so we make a measurement or otherwise tell ourselves a story consistent with (2), even though (2)'s conclusion doesn't truly follow.

This reminds me of topics in government policy, psychology, etc.

[jjoonathan]

The simplest bumbling incentive following often leads to exactly the same place as the most cynical machiavellian scheming.

[cma]

Also, a single rule about what monetizes best may not apply to all companies (pissing off high dollar investment clients over something like that?), so they may mostly all be optimizing it even if there are different choices.

[Cederfjard]

You’ve only really stated though that these policies are deliberate, which I think few people would have thought otherwise, not that they’re necessarily the best policies there can be. The question is if they’re actually better for the bottom line than the alternative (given the timeframe that the people who make and influence these decisions care about). Is ”squeezing out an extra pay cycle” or two possible missing the forest for the trees, if customers who were happy with the cancellation process are more likely to return, proselytize for you and so on? Not saying that’s the case, very open to being influenced either way if anyone has data to share.

[butwhywhyoh]

The OP was directly countering the point made by the GP:

"But sometimes bad policies are just bad, they benefit no one, and they exist for dumb reasons. Maybe call to unsubscribe is one of those policies"

No one stated anything about it being the "best policy it can be".

[Cederfjard]

OK, fair enough. I still read it in the context of the thread's original thesis, that frictionless cancellations increase customer satisfaction, and thus retention and profit in the long run. And I still don't see the comment they directly responded to saying that the policies aren't deliberate, only that they "exist for dumb reasons" and "benefit no one", which could still arguably be true if the alternative is both more profitable and serves people better.

But I hear you - taken more in isolation, and with better faith on my part, the comment makes sense. I'm still curious about actual data though.

[rudian]

I think “companies” have run out of the “benefit of the doubt” as far as I’m concerned. Using the web has become a pain not because companies don’t care about UX, but because they think popup X will increase profits — and it often does.

Certainly there’s a good amount of ignorance in every company, but many choices are purposeful.

[robbmorganf]

I think they often benefit one specific person in the company, who happens to be the decision maker, and don't hurt the overall company obviously enough that anyone stops them.

[Retric]

The third option of hurts the customer and company is also extremely frequent. This can be as low level as developers choosing tools to pad their resume, systematic based on internal metrics, or even very high level internal politics based on which policy ends up making someone or some group look good.

[stedolph]

Yes, this is the way. It doesn't makes sense.

[Griffinsauce]

One option supports customers who don't need your service/product _right now_ and the other doesn't.

It's incredibly naive to think that customers are one-and-done for any service. Value is related to context, they may move out of it and back in and your service will make sense again (unless you add costs through shitty unsub. friction that is...)

It's also the difference between an endorsement and a warning to potential referrals. You're not just burning off this one customer, you're hurting your chances with everyone they know.

[loh]

I always treat my customers/clients the way I would want to be treated, and it always seems to work out for the best. I've also given refunds to some customers without them asking for it first because I value their satisfaction more than I value a few dollars. There's no way to know for sure, but I suspect treating customers like this (like humans) increases retention and well-being, and not only for them, but for me (and my company) as well. It's called the golden rule for a reason. :)

[alsetmusic]

I had a problem with a large appliance. Guy came and looked at it and recommended a course of action that was free and I took it. The problem came back after a month. When he found out his advice didn’t hold, he came immediately to perform a repair because, as he put it, he didn’t want his reputation to be damaged in my eyes.

He offered what he thought was best. When that wasn’t it, he wasted no time getting to the next step. I tipped him more than I otherwise would because his concern over getting it right seemed genuine, and I care more about that than I do getting it right the first time without really caring one way or the other.

[derefr]

> Customer satisfaction is so much higher when they get a one click unsubscribe.

I mean, that's pretty tautological: unsatisfied customers who become not-customers, aren't part of your metric any more! The lower the barrier-to-exit is for your customers, the more survivorship bias there will be in any customer-satisfaction metrics you're trying to track.

When there's zero friction to quitting (incl. a high availability of alternatives to switch to), any time someone doesn't like your company, they'll just churn, and you'll never hear about it. So you'll think everything is great—despite the churn, customer numbers are growing, and all your non-churning customers love you!—even while there's this huge shadow-population of customers who resent your service for one reason or another, but who took that resentment right out the door with them, never bringing it to you to address.

(Not saying this is a bad way to do things, precisely; just that you have to be aware that it's the siutation you've put yourself in, especially when interpreting customer-success metrics.)

An interesting corollary to that effect, though, is that organizations that are effectively impossible to leave (e.g. utilities) will get a "true measure" of satisfaction, with no survivorship bias. If there's "marketing science" to be done on customer satisfaction, utilities are probably a great "spherical cow" simplified environment to study it in.

[ikr678]

You can have low friction exit pathways that still capture that resentment/feedback.

'We're sorry to see you go! Are you cancelling X because (price/UI/technical issues/moving/no longer need/etc)'

[firefoxd]

That's incorrect. A customer who unsubscribes is still a customer and is counted towards CSAT scores. Note, that for many services you can make purchases without being a subscriber.

Also, modern helpdesks are a lot like Mailchimp. As in they handle customer service from many different companies and can track customers across services and score them.

[noneeeed]

I wish more companies would get this. I don't continuously subscribe to Netflix or NowTV but knowing that I can unsubscribe with a couple of click is one of the reasons I'm happy to sign up from.ti!e to.time for something specific. If I feel li!e signing up is a.risk, I do t do it.

[jonny_eh]

> The worst customer satisfaction score with least lowest rate of re-subscription is from companies that make it hell to unsubscribe.

Does "least lowest" mean highest? Or did you mean "least/lowest"?

[idworks1]

oops, that was a typo. Fixed it.