For some reason our most popular IFTTT recipe involves sending spam Youtube links as datapoints to Beeminder so that Beeminder will link to them or something. It’s pretty annoying.
The most drastic possible solution to that is to require a payment method to sign up for Beeminder. There are actually more and more reasons that we’re taking that idea more and more seriously (even though there are much more direct solutions to the IFTTT/Youtube spam problem) so I wanted to run it by you all before we get too tempted to just up and try it.
In fact, let’s hypothetically consider the most extreme version: We tell everyone with a Beeminder account but no payment method that they have 30 days to add one or we’re culling their account.
(To be clear, this is just about adding a credit card (or, if one must, PayPal), not premium plans.)
That would be bad for me personally
That would be bad for people I know personally
That would be bad for my ability to evangelize Beeminder
I predict that would be a horrible idea
I’m all for it, beeminding without a way to be stung is dumb
If it makes Beeminder more sustainable, it’s worth it
“Would be bad for my ability to evangelize Beeminder” - I know I wouldn’t have started using Beeminder if it asked me up-front for a credit card, so I therefore wouldn’t be able recommend it the same way.
“If it makes Beeminder more sustainable, it’s worth it” That said, I don’t recall clearly how I moved from free -> paid; (I assume some combination of having a long enough unbroken ‘chain’ of success that I felt like it was productive, and really liking your blog posts/etc and wanting to support the service (if only so I wouldn’t have to code something similar (meaning worse, but “functional-enough”) myself ))
I note (at time of writing) that everyone wants Beeminder to be sustainable, even the one who also predicted it would be a horrible idea )
“beeminding without a way to be stung is dumb” - I do get this, though again I always want to “dip my toes in” before getting into heavy commitment with new services.
That said, what about this random idea: Allow a no-credit-card user to sign up if they link their twitter account, with authorization to tweet-shame them if they derail – it might be a way to be stung without requiring a CC
They’d still have to provide a CC to unlock the task, of course, they’d just be able to bypass the initial sign-up requirement)
I object on principle to any hurdle to demoing a website. Websites which require you to register before demoing the product really irritate me. Unfortunately this is 99% of websites. So I’m constantly irritated. So maybe not a useful opinion.
I actually changed my mind on this. I was going to say “beeminding without a way to be stung is dumb”, but I think it would be nice for someone starting out to have a $0 pledge goal to see if commitment devices are right for them.
I think it would drastically increase the number of people who abandon the site as they’re signing up.
I would say a much simpler solution is to just set a limit for automated data points for people who haven’t added a credit card. Or even just setting restrictions for people who don’t have a credit card added, e.g., if you don’t add a credit card, anything resembling a URL would be automatically removed from datapoint comments.
Since the only viable way to use Beeminder for spam is putting URLs in datapoint comments, just blocking those for individuals without a credit card would probably totally solve all spam issues.
I rarely join anything that demands a credit card to set up an account. I like to join, take a look round, see how other people are using a service, and then get into it if it sounds like it works for me. Even the kind of free trial where you have to give your credit card details first kind of skeeves me out. Sure, it says I can cancel any time but I worry about forgetting or the site having secret ways of making it hard to cancel. I’m way more likely to jump in if I get a chance to poke around first. Otherwise I probably wouldn’t even have looked at Beeminder. So I would definitely worry about maintaining user sign ups and the ability of existing users to evangelize – rare are the situations offline where you would feel comfy handing over your credit card as insurance to enter a shop where you might not want to buy anything, but once you’re in all sorts of unforeseen items can tempt you.
I would make a partial vote for horrible idea… not certain, but yeah. it definitely has an impact on the tone of things. Especially if what if you get a bunch of backlash after you announce it, and then you decide not to do it… well, then you’ve just issued an empty threat, basically, which is not the sort of thing you want on your track record.
But of course, if it does in fact makes beeminder more sustainable, it’s worth it.
And oh yeah, I think that evangelizing something is made harder by there being a larger wall up. I don’t evangelize beeminder that much directly/personally, and if I did then my pitch would probably be strong enough that the person would be down to put in their payment info, but I would imagine this would dramatically reduce the number of people who sign up casually to check it out… probably including people referred from Complice.
I guess I’ll further add that the ability to force people to pay/provide credit card on signup probably depends to a large degree to the extent to which the person can be convinced that they want the thing before they’ve tried it. Of course, in Beeminder’s case, part of “tried it” means having money on the line, so there’s a bit of a catch-22 there perhaps
But as shanaqui says, getting a chance to even poke around something unfamiliar before entering a credit card makes tons of sense to me.
The example for why this is coming up now, seems like an argument for making IFTTT premium, or at least dependent on a CC. I very quickly went for a premium account back when, so I’m probably more positive in general to giving up my CC, though.
I certainly wouldn’t keep such a requirement from recommending Beeminder, and I think most people accepts that the requirement of a CC or not depends on what the site is actually for. The trust issue is a different one though, but could possibly be alleviated through stuff like TrustPilot or similar? Although I tend to trust such sites less …
Whether you do it or not, the youtube spam thing (a common ransom of popularity) is a bad reason to motivate it.
The “most extreme version” sounds to me like a reasonable corollary of the idea that’s controversial in itself.
I’d be a lot more concerned about upcoming users being turned down by having to put up a credit card to signup. I’d be turned down. Think about what you expect an app that asks for your credit card info first thing on signup to do (hint: it involves illicit operations).
But then, you’re targeting a very niche demographic, and that population might be just crazy enough to tolerate it. And you have everglowing reviews / immense social credit in the rationality community, so you’re not just the next Nigerian prince on the block, at least for those people.
So overall, I’m torn on whether it’s a good idea or not. In your shoes I’d A-B test it (like, asap). As I’ve surely already mentioned.
There are certainly people who would be put off by the credit card requirement. I probably would have been one of them, though if I was sure I could do my first few graphs with $0 on the line I might still have tried it out, given the strong testimonials I’d seen.
The question, I think, is how many of the people who get scared away would eventually have become established beeminder users, versus “people who were never going to beemind anyway.” I don’t know how to know that! The whole thing gets into whether Beeminder is a niche product trying accurately signal what it is and scare off people outside its niche, or whether it’s a broadly appealing one trying to lure people in to learn the ropes.
I do agree, though, that “avoiding spam” is a bad reason to add the credit card wall.
I’ll echo the sentiments of many here: on the one hand, beeminding without a way to be stung defeats the purpose. On the other hand, putting a credit card down with unlimited authorization requires a lot of trust:
Trust that the humans running the service are not evil scammers
Trust that the service is well built (it functions as the designers intended and isn’t going to mistakenly charge my card)
Trust that the service is understandable (it functions the way I think it does/should … i.e. I’m not going to be mistakenly charged by user error)
…and add to that a learning curve where I may not know the right things to track, or how.
So I also would recommend against requiring a CC up front; maybe new users could get a first n derailments without a payment method before the goal gets paused in some way. (Not exactly a paywall; is it a “stingwall”?)
Echoing other comments here – the messaging would have to be pretty on point. Otherwise as a user, all I read is “we’re strangers and we need your credit card but don’t worry, we’re not going to do anything with it.” I don’t trust strange websites with my email address, half the time, much less putative access to my finances.
That said, you’re deeper in the product than I am; maybe that barrier to entry nets you a more useful sort of user. And maybe you could handle the messaging in a way that provides confidence. E.g. “If you don’t want to give us a CC, link your Venmo/Paypal account. That puts skin in the game for you, and even if we’re compromised by The Borg we won’t be able to access your finances without your permission. Remember, even though we’re requiring a financial link to you, Beeminder will never charge you for goals without a financial sting, and of course will never charge you unless you derail.”
In my case, Beeminder was one of several productivity apps I was experimenting with. I didn’t necessarily want to Beemind; I just wanted to get several habits on track. If BM had required a CC up front I just would have punted it to the bottom of the list, and worked my way through the free apps first. So the gamble there is that the other productivity apps/sites out there would be dissatisfying enough to make me keep searching for the Beeminder Solution, but not so frustrating that I burned out on productivity apps altogether. Slash, that the productivity app community would keep recommending this app until I went for it.
I don’t know if my organic conversion to premium member is the happy, sustainable path you’re looking for, but it was nice for me. I started with a couple habits, derailed a couple times, kept having ideas for things to beemind, consumed all your content marketing, and then ran into that limit of test habits right about the time I was ready to convert to premium. I forget how long that took, but it was months. Maybe a year. Way less traumatic as a user, but again, I don’t know if that’s SaaS revenue for you soon enough.
Anyway, I voted that CC would make it harder to evangelize, since my current line is “try it, you’ll like it, and it only costs $$ if you want it to.” But also voted that if this idea makes BM demonstrably more sustainable, then hecks yeah, do it.
Thanks so much for the ideas and insights, everyone! Let me dive in with replies:
Definitely understood that we’d lose some people, but now that you know we’re legit and everything, why have reservations about recommending us?
Good thinking but two counterarguments: (1) We have to keep the service we’re offering as focused as possible. I’ll wait for the laughter to subside. But, seriously, it’s worth striving for when possible and generalizing what being stung means is, therefore, something we should avoid. (2) Spammers can create spam Twitter accounts.
To clarify, that wouldn’t change. This is just about having the credit card from the start.
We do have that already. If you derail without a credit card on file the goal freezes till you add one. You can’t even create more than 1 or 2 goals (we’re a/b testing between 1 and 2 – and finding no difference) without adding a credit card.
Doesn’t that line still work? Granted, it’s not as reassuring as “you don’t even have to enter a credit card until the first time you go off track”.
I agree with everyone pointing out that spam signups aren’t a big factor here. It would be nice to not have all that messing up our analytics though!
And more on why I still like this idea despite some good counterarguments:
Adding a credit card is a tiny ask. There do exist people who will flee at that request. Most of them would’ve never paid anyway. Some would’ve and that’s a real loss. But, I conjecture, it’s counterbalanced by the people who will go ahead and add their credit card and start taking Beeminder more seriously.
Part of this argument is that when you’ve derailed is a horrible time to ask for your credit card because you’re highly liable to stick your head in the sand and ignore all our emails forever. Especially, as is commonly the case, if you’re uncertain about whether adding a credit card means getting immediately charged for past derailments. (Yes, you ought to know that those were $0 derailments, but you don’t, because you weren’t paying much attention, because, as I mentioned, you weren’t taking Beeminder seriously without having had a credit card entered!)
Disclaimer: I haven’t read most posts in the thread. Danny requested I go ahead and add my $0.02 cents here (my words, not his), so:
I’m all for requiring payment on signup. Without data, can’t know for sure, but I would wager that your new user signup metrics will drop somewhat, but value per user will increase such that there’s a net financial gain for Beeminder. That’s legitimately good for users who do sign up over the long run, and new user engagement (pray forgive the buzz phrase… it seems to fit well) should also go up.
That said, I strongly think you should “grandparent in” existing users who’ve yet to add their card; don’t apply this change to them. It would antagonize people who’re still trying out the service, by violating expectations you’ve already set with them. I don’t see any upside for Beeminder with this class of users. Extreme, indeed.
So: I’d vote down the “most extreme” hypothetical, but I would absolutely start requiring a credit card upon signup for new users.
Trying to put this mathematically, my theory is: [value lost due to people taking Beeminder seriously but spooked by violated expectations] (group 1) >> [value gained by people prompted to put up their card] (group 2). This ignores people who flee (a group 3), since there’s no value to account for there. The gains from group 2 will be one-time, short term, and likely to be realized by Beeminder eventually, anyway, whereas the value lost from group 1 is only lost if you force the question.