Kill your credit card-less free trial. | Hacker News Reader

Kill your credit card-less free trial.

observationdeck.net

39 pointsrfelix212161 comments

This strikes me as dangerous advice that's very business centric instead of focusing on how customers think and act. Contrary to what every business owner wants to believe, their business isn't particularly amazing or magical and customers have no problems abandoning them during evaluations if there's the slightest conversion obstacle. Not letting customers try out your service before entering credit card details? Yeah, that's a major obstacle. Better check out the competitor's product that's easier to use.

Requiring credit card details is a good way to make people distrustful of you. In my case at least, I am extremely wary if I'm asked for them, because it always seems that the free trial is not actually free and the site owners want me to not come back but end up being billed unwittingly.

I agree wholeheartedly. Often I will auto-cancel subs immediately after sign up to make sure paying is a conscious choice, but will discover they've considered this as a possibility. So the 30 day free period is instantly ended when I cancel the auto billing part way through. That may be a gripe of my very own, however.

" Some would even sign up, promptly forget that they signed up, then never use the app again"

"...only 1% of them actually stick around and pay you for your product."

Those two quotes lead me to believe that you're not doing lifecycle emails. Would those people still forget after receiving four emails from you over the course of those 30 days with specific calls-to-action leading them back to the site? Would only 1% of them convert if you actually sent them an email asking them to come back to the site and enter their card details?

My experience has been a lot different to the author's. I see almost exactly 25% of my free trial customers converting to paid. And over half of those enter their card details on the day I send them an email asking them to.

So if you're not doing that, I could see you getting the results you were getting.

That said, it sounds like it would make for an interesting long-term A/B test. I'll try to get around to doing that, and if I do I'll be sure to post the results here.

Not everyone likes spamming their users.

This isn't spam, its marketing to people who opted into receiving communications from you in exchange for a free trial of your service. As long as you stop shortly after the trial period there is nothing wrong with it in my opinion.

I actually really enjoy a good series of marketing communications from a product I'm evaluating to 1) show me how others are using the product 2) point out a product benefit I might not have notice 3) to remind me to actually use the product during the trial

"I actually really enjoy a good series of marketing communications from a product I'm evaluating"

Said no user ever. As people interested in the business of software I think we're all a bit damaged when it comes to how much marketing nonsense we're ready to accept.

Well, at least one person said it! ;)

I signed up for a free trial of bigcommerce to compare it to another e-commerce platform. I signed up and just never got around to using it. And then bigcommerce sent me a customer service email titled "Need some inspiration?" with examples of how other companies are using the service, three days later "Hear from other successful online merchants" with case studies of successful bigcommerce entrepreneurs.

Finally I received "Your trial is almost over — give me a call to keep your store" from their customer service rep, I'm sure this too was automated but I did end up on a telephone call with their regional sales rep and I'm now a paying user.

So it absolutely does work and you can provide value to trial users if you schedule your emails and the content of your emails correctly.

I'll make a mental note that this might work for certain apps provided clear opt-out is given and the email frequency is reasonable.

Frequency is key. Possibly twice a week, but preferably once a week maximum.

> in exchange for a free trial

So I pay for the trial by being spammed? What kind of logic is that? You're telling your users "buy this product to get spammed even more!". Are you sure you want your users' money? They aren't, at this point. Marketing must always be opt-in.

> I actually really enjoy a good series of marketing

communications from a product I'm evaluating I suspect you are in a small minority.

You've completely glossed over the fact that sending those emails leads to people using the service. Are you suggesting that those users are angry and still handing over the card details?

Of course it would be much better if companies could only email those users who they knew were going to pay (and not bother the other percentage who wont). Until this magic exists, companies will send email because it works. Those who do not are holding themselves back.

I suspect there are still a few users out there that consider receiving four emails in exchange for a free month of a $50/month product to be a fair exchange. Or at least not unfair enough to get this upset about.

Over the years, I imagine I must have sent out something like twenty thousand of those mails for S3stat alone. Some people do indeed click the Unsubscribe link on those mails. I've never had anybody complain that they were "spam" though, since "hey, I hope you're enjoying the service, let me know if you have any questions!" doesn't really trip too many people's rage filter.

Apologies if you've signed up for our trial and I sent you a thanks message. I didn't mean anything by it.

Almost everything I sign up for these days ends up sending me several emails reminding me that I have done so. Yes, I did sign up. No, I don't want to be reminded of it all the time.

It is becoming annoying enough that I avoid signing up for things in the first place, or at least give them an email address I can easily filter.

I'm only doing lifecycle emails on some of my products. Some of them were built way before the current lifecycle email craze, and I haven't gotten around to adding them on the others yet. On those older products, I do send emails asking them to come back and enter credit card details.

Great point, though! My larger point, I guess is about leveraging something that only more interested users will go through (credit card entry in this case) to let them filter themselves out early.

An interesting side effect of prompting credit card sign-ups is that the visitor is engaged in getting use out of the service immediately. They're running against a ticking clock (14 or 30 days) to get whatever service running.

Those who don't prompt for cards are under no immediate time pressure and therefore less likely to engage a month down the line when the specific problem that caused them to search for you in the first place is no longer as acute.

Exactly. This is basically what I was trying to say.

There is a strong argument that many people hear alarm bells when asked for a credit card for a free trial. Yet, there is a valid case for wanting to overcome people's inaction, where people just don't bother to sign up (some of whom would actively enjoy/benefit from the product if you could persuade them to look deeper).

I wonder if there is a model that goes someway towards addressing both of these:

1) One month free. No credit card required.

2) Another month free (with more 'credits' or whatever), if during that first month you supply your credit card details. You will automatically be renewed if you don't cancel, but are free to do so (and will get reminders).

In this way, people get to try the product without alarm bells. They get something in return for demonstrating their interest, by way of more free time and you giving them access to more 'credits' (monitor more servers, look at more profiles, whatever it is).

This helps overcome their inaction, but avoids the alarm bells. Not sure how well it'd work but would be interesting to see.

That's an interesting idea! I wonder how that would work. It's a bit more complexity on the front end, but it's intriguing.

Really interesting. I'd love to see some numbers from a site/product that actually A/B tested both options.

If I had to use my credit card to get a free trial with Buffer for example, I would have never tried it (and eventually converted), but I guess it depends on the product.

There's this report from Totango - http://www.totango.com/wp-content/uploads/2012/11/2012-SaaS-... (PDF)

Really great report! It's based on 100 SaaS companies, and concludes very differently from this blog post: Requiring credit card for signup results in 50% lower retained users after 90 days.

I'm not sure the conclusions are very different. Yes, in this report, you get more customers if you don't ask for credit-card upfront.

But, those 120 signups you get from 1000 trial signups, meaning, for each signed up customer you need to support over 8 trials (support here means your time and to a lesser extent, your service's resources).

For the 60 signups in the case where the credit card was asked upfront, there were 200 free trial signups, ie a bit over 3 trials per customer.

So which is better depends on how much money (or time) you invested in getting the 10000 interested visitors to site (prior to them signing up), and how much money (ie. time and other resources) you need to spend to get them to become paying customers.

Right. The conclusions actually seem to be the same. Another point: Having more customers (at any cost, haha) isn't always what you want.

Less customers who are actually happy to pay me money for my product > More customers who don't care about the product as much, or at all.

That's a good observation. I didn't consider the cost of converting signups to paying customers though, as in my experience, the cost is little to non (when talking the type of SaaS products in the report). But in a scenario where there is significant cost pr acquisition, then the point is very valid.

I think jasonkester below nailed it when stating if you need to interact individually with customers or in any other way have significant expense supporting trials, then that's really something to address. Again, at least in these kinds of products, obviously it does not apply to enterprise etc.

Conversion rate 100/200 = 50%(cc required) vs 200/1,000 = 20% (cc not required).

As OP mentioned in his post: > Also, as an independent developer, it’s also very important for me to minimize time spent dealing with users who will never give me money for my product.

As an indie dev, it's a big difference.

If your product requires interacting with individual customers during their 30 day trial, that would seem to be something you'd want to work on changing immediately.

The sorts of questions that people frequently ask in the first month, tend to be things that would fit in a FAQ. If you find yourself sending the same paragraph long email more than once, it's probably worth finding a UX tweak to the product to ensure you never have to send that mail again.

For my products, for every 100 trial customers, I'll exchange emails with maybe 5 of them. That mail is usually very short, pointing to the piece of documentation that explains the issue. Only in rare cases is any actual involvement required on my part.

That's baked into the product intentionally.

I'd like to see a "PubMed" of case studies and findings like that.

X-posted on FS: http://firespotting.com/item?id=1144

What you're basically saying to prospective new users is this:

"I don't trust you enough to spend bandwidth and server-resources on you without getting your ID and credit card info. On the other hand, I expect you trust me with your personal identification and credit card information right off the bat."

I've had to deal with credit-card fraud in the past, so I simply don't deal with companies that do this. I suppose the question is whether or not you're saving enough on costs to offset losing clients who are cautious with their CC and ID info.

When I started off, I wanted every user who came to my site to convert to a paying user. Over time, I realized that I actually didn't want for every person who showed up to become a paying user, because I spend more time dealing with issues that they bring with them than writing new code or adding features.

After you've been running businesses for awhile, you start to realize that the goal should be to get a specific type of customer (not every customer is a good customer for you, some types are a bigger drain on support resources than others), and the easiest way I've found to get those is to require a credit card for the trial. That might mean that I have less users overall, but I end up with the ones that I want.

I'm also not talking about saving on literal costs, necessarily, It's more about opportunity cost. I could be dealing with a ton of free users, or I could be dealing with a self-selected group of paid users and work on new projects. As a developer, a couple of hours here and there could potentially turn into a new product.

To me, this more or less sounds like "I'm having trouble converting, so let's focus on the really, really, really interested users". In some occations this might be the right choice to make (the service you're providing have very low margins or it's a very complex service, so overhead from support becomes costly, or time is very limited), but in general it seems like very bad recommendation.

More likely it's your value proposition that is weak – either your customers don't understand what (value and/or service) you're providing, or your onboarding is just not good enough.

If you fail to tell your visitor what you're actually offering (before they signup) you might end up with a lot of users that sign up, just to learn what service you're actually providing. This is obviously much more likely if you're not requiring a credit card at signup. Try to have people understand WHY they're signing up before they actually do. They should be convinced that you're the right choice before entering their email.

Second: Your onboarding process is not over when the user has created their account. Guide your users. Give them examples, feedback and inspiration. Still having trouble getting started? Email them a week later and ask if you can help. Keep giving them reasons to come back and eventually convert.

Here's the thing: when there are a lot of services that do something similar, I sign up to the free trials of lots of them at once to evaluate them. If you require a credit card, I probably won't include your program in the initial evaluation.

So yes, if you require a credit card, you're getting fewer people with a higher likelihood of conversion, but you're also missing out on a lot of people who are doing comparisons of services. How many of those people you convert is likely to be a function of how good you are compared to the other options.

If you are offering something unique or have amazing word of mouth, this probably won't matter. If on the other hand you think your product is one of the best solutions to the problem out there, you're probably losing more customers than you want by demanding credit card details.

In the case of the OP, doing something that on the face of it anyone who needed that service could hack together a script to do roughly the same thing within a relatively short period of time, I would have thought that it would be very important to remove as many barriers to trying the thing as possible (but yes, there's no need to let the free user have access to checking more than 2 sites).

For website monitoring, a big part of the reason that people buy is trust, not necessarily ease of signing up. When you're running a commodity business at scale, you'll also find that more and more users will sign up for free accounts repeatedly over time instead of paying you money, which is yet another reason to require credit cards for free trials.

When you're doing most things yourself, like me, the goal is to get a specific type of customer (not every customer is a good customer for you, some types are a bigger drain on support resources than others), and the easiest way I've found to get those is to require a credit card for the trial. That might mean that I have less users overall, but I end up with the ones that I want.

I see. I'd assumed that you were competing primarily on ease of use and pleasant UX, which if that were the case, you'd want potential customers to see what you are doing.

If it's actually about trust, it sounds like word of mouth is actually more important to you. It just seems a shame that you'd be extremely unlikely to get my business (assuming I were looking for such a service) simply because I wouldn't be able to evaluate your service without handing over data that I'm loathe to spread around, particularly for services I might not actually want.

[By the way, another model that might work for website checking as well as restricting the number of sites is delaying the notifications on the free account after the first week. Anyone who really wants the service almost certainly needs the notification immediately. It also gives you another excuse to email them 'by the way, your premium notification speed add on is expiring, upgrade to a full account to keep your instant notifications....]

A bit annoying article. The phenomenon that article describes is not about credit cards, but instead of the psychology behind actively unsubscribing vs passively unsubscribing at the end of a free trial.

Most people has a tendency to inaction. If you give users two services of equal worth, but in one you need to unsubscribe to get rid of it and the other the default action of doing nothing results in unsubscribe, more people are going to end up paying the first one because they forgot or did not care enough to go and unsubscribe. Scammers who sends false bills use the same concept, as they simply hope some people will just automatically pay and not think or even read what the bill is.

In the end, using a credit card trial where the user need to actively unsubscribe at the end of the trial will give you more users who dislike your product or simply isnt using it at all. However, hopefully its enough of a barrier so people who wish to have a "viral" product can't use it.

You can fix this by not automatically charging at the end of a trial, and sending an email instead to ask if they still want to keep the account, and closing the account or charging them based on their action at that time. I didn't say anything about automatically charging anyone necessarily. This is more about using credit card entry as a barrier to entry that gets you more of the type of users that you want (ones who are interested in your product and likely to pay you).

Also, if someone gets charged at the end of a trial because your code is set to auto-bill them and they forgot, it's not like you can't refund them. Not doing so would do more harm than good. You only want to bill people who are actually using your product.

You definitely shouldn't use this as a tactic if you want to have a viral product. This is specifically about web apps.

You can make a lot of money if you bill people who are no longer interested in your product and forgot about it a month ago. Sometimes they'll pay you for years without realizing that you provide no useful or necessary service of any kind!

Maybe I'm just not ruthless enough for business, but this does not seem like a great way to make money. I'd rather make less through honest customers who like my product, then screw over those who don't.

Easy fix here: Send receipts every time you bill someone.

It would seem that no one has addressed the idea that maybe the product doesn't hold enough value to the user to continue after a trial, I really don't think these sorts of metrics are applicable unless both the business model and the product are virually idential to his.

There are too many factors to consider, what is the product? being the most important, how is it sold to me, what is the closing process?

What this post says to me is: we want our users credit card details because there is a likelyhood they will forget to cancel their trial and we will at least screw them out of one month subs.

Until now I've always charged a very low trial fee. Say, $1 for the first 30 days of service then auto bill after the 30 day $1 trial.

Never thought it was a good idea to request credit card for free trial, though I haven't tested it.

Now I am testing a free trial that is action based, so the first 3 actions are free, but anything beyond that requires a full subscription. The reason is because the product I'm selling the primary concern potential customers have is "how would I use this" so we give them time have an a-ha moment but not enough to get a tremendous amount of use from it.

For those requiring a credit card for trials, how are you tracking metrics/conversions? Since the conversion happens when the trial ends, which is an event disconnected from your customer's browsers session, none of the JavaScript/cookie-based 3rd party software will be able to be capture that event. Some do have a backend API you can call if you can identify the user (e.g., KissMetrics & MixPanel both have this), but others do not (e.g., AdWords and Optimizely).

I'm writing an app specifically to help with this. You can sign up on the bottom of the original blog post. :)

Thanks. I read the article, but completely blocked out that blue box. There's so much hawking of eBooks and such that I skip over that stuff now.

I would be more likely to sign up for full price. "Free trial - just add credit card" is a massive alarm bell for me.

In my experience, this type of free trial often means you want my credit card details so that you can make it as awkward as possible for me to cancel and maybe place dubious charges when I'm not looking.

That experience makes it difficult for me to sign up for a similar free trial.

I'm sorry that you feel that way. I definitely understand the feeling of being burned, but I've never had issues with something like this, as a user myself of services with a free trial that requires a credit card.

Things like Twitter and Hacker News, where you can broadcast your displeasure of a company's terrible tactics would make it difficult for a company to keep signing up new users if people are complaining about it. The only companies that I see getting away with stuff like this are huge companies who don't care about their customers anyway, like Tivo.

All of my apps have a prominent link in the user settings panel that lets a user cancel their account/billing with one click.

The biggest problem with requiring a credit card with a free trial is that unscrupulous people also use the same tactic. Rightly or wrongly, the pattern "requires credit card for trial == scam" is etched into some people's minds. In order to overcome this, the company would have to be well known and well respected, at least within their niche.

That is definitely true. Personally, I feel fairly secure in putting my credit card info into web apps for free trials because of protection offered by the credit card company. If I have problems with charges that I don't want on the card, I can call them up and reverse them.

I see how some people's knee jerk reaction upon seeing that a credit card is required for a free trial would be to just close the tab and find a competitor, though.

Like almost everything, there's no simple answer as to whether requiring a credit card to start a free trial will be net positive.

This is very dependent on your product, market segment, trial length, pricing, and many other variables. You need to test this out for yourself.

Amazon does this neat thing with Prime where if you stop right after a billing, if you haven't "used" the service they reimburse you fully. Was a pleasant surprise (Amazon's customer service is world-class)

Your advertised feed[1] seems to be missing. Returns an empty document.

[1] http://observationdeck.net/feed/

Sorry about that. It's fixed now. Apparently this happened: https://twitter.com/rfelix/status/341487193224798208

Does anyone know any good articles about trial vs. non-trial for premium subscriptions?

Just know that I won't be using your product. And most of my friends have no credit card either.

Besides not being able to sign up, I also wouldn't want to. From what I heard about how credit cards work (as opposed to debit cards, which I think is the right term for the cards that we commonly use here), I'd never sign up for a free trial and give the developer (and any potential hackers) the permission to withdraw arbitrary amounts from my credit card. I sign up for so many things and try out so many demos, it'd be a major security risk. I even prefer not to give out my real name anywhere, let alone bank account details.

But then again, if you think this system works for you, go ahead and use it. If it makes you richer, who am I to say you shouldn't.

You sound kind of angry. To be fair, I think the author was referring to all kinds of cards, not just credit cards. It's something of a catch-all.

Then you're not the target market. If you have no credit card, it's very, very hard as a global website (i.e. not localized to the Netherlands) to accept money from you. So if that's the case, then they don't even want you as a customer, because you have no way of paying them.Giving you a free trial is then counterproductive.

exactly.

It's actually really easy.... ( http://bitpay.com )

No it's not, that just moves the difficulty from merchant to client. Actually getting bitcoins is not easy, takes a long time, etc. It adds friction, support overhead, risk, latency, for very little pay-off.

We're talking about subscriptions here, so the customer would have to send bitcoins every month. He would have to first get bitcoins, then those few days later actually remember wanting to use the service, send the service the coins he paid for a few days back, wait half an hour, again. All that, for those maybe 1 out of 10K leads that would even know what bitcoin is.

I'm curious about how you would become a paying customer to a service if you don't like providing your details online. PayPal?

If you're never willing to pay then there is very little incentive for a business to want to support your needs.

In the UK, a credit card give you much more protection than a debit card and is safer to use online. http://www.money.co.uk/article/1003471-how-section-75-of-the...

That I'm reluctant to give out personal details doesn't mean that I won't do it when something is good enough that I want to pay for it. Paypal is a perfectly fine payment option... as long as that doesn't require a credit card. I don't know how, but somehow Grooveshark made their Paypal button in a way that only accounts with credit cards are accepted. Totally useless.

Reply on news.ycombinator.com