Maybe the “pay $1 for 10 more goals” feature ($1 can be something else) should be brought back. So 0$/month users can still play around with adding new goals and they feel like they are sucked into a subscription.
Fully agree. Nor can Beeminder make you practice well twice a week; judgements of quality still fall within the honour-system part of recording datapoints.
If $2k is the implied budget, then I’m guessing we’re also not talking about ordinary once-a-week kind of lessons, but some kind of frequent practice-in-my-presence kind of deal. Still sounds like an underlying don’t-really-want-to-learn situation, though.
I think this is the aspect I keep coming back to – I try to track my goals via at least two metrics, and the more important the actual task the more value I get out of the throw-everything-against-the-wall-and-see-what-sticks approach. I have seven goals for studying for my comprehensive exams!
So “three beeminder goals” for me means “beemind only one thing you’re trying to do, but badly.” And I don’t know that the value of infinibee would be clear to someone whose “core beeminder” experience doesn’t include what I think of as actual beeminder. (I think capping at a higher number would address this, though?)
But then, I seem to have become a “power user” somewhere along the line, so maybe my sense of “how Beeminder is best used” is needlessly complicated.
Fascinating, I think you have successfully articulated something I’ve been thinking, albeit in a more nebulous way.
Relatedly, I was surprised at how motivating Stepbet was for me (TL;DR you risk $40 for 6 weeks). Around here getting two large pizzas delivered costs ~$60 so on paper $40 isn’t a ton of money. But the combination of autodata, community, and time-bounded gave me a lot of motivation for a brief time period.
For Beeminder by itself, I try really hard to be motivated by the higher pledge levels I’m not at. I only allow myself to step down 6 pledges per year, so I avoid derailing at $30 because I’m afraid of $210.
Arguably, sustainable revenue aside, the ideal beemindee is so financially constrained that they don’t dare ever derail and so morally upstanding that they’re never tempted to weasel out of the commitment.
Would it help if we had a feature that constrained step-down somehow?
On some of my goals, I allow myself to step-down the pledge after 90 days of good behaviour, and have sometimes wondered whether that would make for a good automated feature. Not least because I suspect that higher pledges make for more conservative roads and less overall awesomeness.
I haven’t really been tempted to abuse it. I track it with its own goal of course (which would be Do Less if it weren’t for all the Do Less headaches, so it’s Do More but I don’t cheat).
It doesn’t exactly make me want to leave, no, but it does make me want to stop creating new goals. I’m hugely hassle-avoidant. Dealing with the decision energy this will cost right when I’m in the middle of trying to make a new goal is more than I feel like dealing with.
Of course, if I don’t make new goals, what am I sticking around for? The “lifestyle” goals I basically never derail on, I guess. Which puts you back at not making money off me.
It’s fine as a conclusion, sure, in other words if money isn’t very tight in your life at the moment, then beeminder is not for you. That’s probably just plain correct. If you want to spend $5000 on learning to play the guitar, then beeminder quite simply isn’t the best tool for it.
But it sucks from beeminders perspective if the conclusion of wealthy users is that beeminder has nothing to offer them while the conclusion of poorer users is that beeminder is to expensive. Because that leads to a end-result where beeminder has few paying customers.
That’s why I asked who beeminder is for. The current model seems to be not-good for both wealthy and poor users; so what’s left then?
The poor users who don’t dare ever derail get good value from beeminder. But they don’t provide any -income- for beeminder. I was arguing from the perspective of beeminder. The entire point of the subscription-models is after all to provide an income to beeminder.
And the person who is STRONGLY motivated not to derail at $10, is probably also strongly disinclined to sign up for $16/month of costs, no?
I think you’re under-appreciating the beauty of the exponential pledge schedule! If you’re Richie Rich and it takes $2430 to motivate you to practice guitar, that means that you derailed at all the previous amounts – $0 + $5 + $10 + $30 + $90 + $270 + $810 – which adds up, not coincidentally, to exactly half of the $2430 you now have at stake. So you coughed up $1215 but now you’re at the ridiculous pledge of $2430 and that should keep you practicing for years without spending another dime.
For me $810 tends to be the point where I consider it out of the question to derail and diligently stay on the road indefinitely.
It depends a lot on the goal. I just called $810 my motivation point but that’s for things like writing blog posts (classic case of a “one more day won’t matter” slippery slope of doom). For my pushups goal, $5 or $10 is plenty because it’s too silly to pay really any money just to not do some damn pushups. If I just set alarms or something I’d never follow through because I wouldn’t feel like doing them at that moment. I always prefer “later” or “tomorrow”. But a small nudge from Beeminder suffices to make me just do them.
Maybe the real counterargument to your theory that Beeminder’s motivational schema is self-defeating as your wealth increases is just the case studies like PhD theses that wouldn’t have happened without it. We know of many massively high-value things that Beeminder made possible (not least of which being Beeminder).
Or maybe the most fundamental counterargument is that Beeminder is profitable, and unless these premium changes have generated a jump in short-term revenue by cannibalizing long-term revenue (we’ll learn that soon!) then we’ve now achieved what we hard-committed to for 2016 for Beeminder to survive into 2017 and beyond. Phew!
Per-goal fees for advanced features
I agree with the point raised about there being a contradiction between expecting Beeminder users to fork over say $16/month while also expecting them to be sufficiently averse to stings from derailing. As I floated here, maybe there should be an alternative to the $16/month plan where users pay a one-time fee for auto-retrorachet to be enabled for a goal.
Same with other advanced features like custom aggregation methods, and maybe the SMS bot. Weekends off and road editor should be folded into the $4/month plan, IMHO, because I don’t think they’re either confusing or revenue-impacting.
The $32/month plan can either stay at $32, or maybe lowered to the $16/month tier. Either way, I predict a very small subset of users who actually care enough about those features that they’re willing to pay 4-8x the base user fee.
Most active users should pay a subscription fee, but newbees should be able to get the full experience for 6 months
Where I disagree with a lot of people here is the idea that $4/month is “too high” or “unfair” or “not worth it”. I’m speaking up for the silent majority here, because I don’t want this transition to a revenue plan based on monthly subscription fees to be–pardon the pun–derailed, just because a small subset of users are particularly upset and vocal.
By all means, these upset users should be taken care of. Give them extra free goals, deep discounts, whatever makes them happy. But don’t change the basic policy (the # of free goals may need to be bumped up or combined with a time-limited trial of several months). Having special discount periods where users can get on an Infinibee plan for $2/month instead of $4/month may also be a good idea. The two keys here are a). get most active users on some sort of subscription plan, in order to ensure Beeminder’s long-term financial viability, and b). differentiate between different users’ willingness to pay so that everyone can get on Infinibee at a price they’re comfortable with.
I’m on record elsewhere as saying that I think the new limits on free plans are too strict, that newbees should be able to have unlimited goals (and goal types) for several months (4-6 months, IMHO), after which they can be given the option of either only having 3 active goals or to pony up and pay $2-$4/month.
However, I really don’t get the outrage over the basic idea of charging some amount of money for a subscription fee. Surely if you use Beeminder enough to care about these changes, it’s worth some amount of money to continue to have Beeminder around to use. Maybe this is $4/mo., maybe this is $2/mo., but some non-zero amount. To be really blunt, if you’re not getting $2/mo or $24/year of value from Beeminder (on top of the $ you spend on derailments), you’re using it wrong.
I think some of you may be looking at the situation with the wrong frame. The relevant comparison is not $2-$4/mo. vs. free, unlimited goals (the pre-Infinibee era), but $2-$4/mo. vs. not having Beeminder around anymore. If it helps, think of the period where you got Beeminder for free as a really long extended free trial or beta test. Or another way to look at it is that the $2-$4/mo. fee is a communal tax we pay for an important public good. Finally, to put the price in perspective: would you rather drink one less coffee/smoothie/soda, or stop using Beeminder?
Running A/B pricing tests is ethical
Finally, I know I’ve said that I think $4/mo. is fair and a “no-brainer” for me, but maybe that point is $2/mo for most people. This is an empirical question, and should be answered empirically. I know running an A/B test with pricing seems ethically shady, but I personally think it’s justifiable to let a subset of users get a discount (it may be different if $2/mo. was the original price point and now you’re charging some users $4). After all, special discounts have already been handed out left and right in this thread and in the comments section of the Infinibee blog post! There’s a strong norm for treating people equally, but while I think that’s a good general rule of thumb, this heuristic shouldn’t be slavishly followed.
If you were genuinely going to charge everyone $4/mo. for a service, but charge half of them $2/mo. while you’re figuring out pricing, that seems perfectly fine ethically speaking. And if you do decide to lower the rate to $2/mo., you don’t need to reimburse the people who paid $4/mo. for the extra money they paid during those months (the $4 rate was an amount both parties agreed to as a fair price for Beeminder’s services). What would be ethically questionable is if you then lowered pricing for new users to $2/mo. while continuing to quietly charge existing users the $4 rate.
Since the dawn of time Beeminder has positioned itself as a product that is free unless you derail. Even today https://www.beeminder.com/overview says:
Keep all your datapoints on the road and Beeminder will always be free.
It’s the core philosophy of the service, and it’s what attracted me to it in the first place. The outrage at $4/mo is not about the money or about the value, it’s about that philosophy and the change to it.
I don’t know if I’d say it’s the “core philosophy” of the service, but I take your point that Beeminder has in the past positioned itself as a free product, where you only pay if you derail. (Though I will point out that even in the pre-Infinibee era, that statement wasn’t entirely true, as users had to pay a small fee for every 10 goals created after the first 10).
I guess I just personally don’t feel betrayed or outraged about this change in policy. And I think most users feel this way (but who really knows?). Like I said, I’m perfectly okay with anyone who has such strong feelings about this to get a deep discount or grandfathered in (unlimited goals, goal types, etc; no auto-retroratchet or other advanced features that have always required a premium plan). I do think that applying a grandfathering policy to the entire long-time user base would be a mistake though.
You got me thinking though…How would you feel about ads? (And no cheating and saying that you’d use an ad-blocker). I’d imagine that at least some goals could have very targeted ads, like ads for running shoes or diet supplements for step and weight loss goals. Personally, targeted ads creep me out, but maybe this is another option to explore.
Maybe that’s too strong of a phrase. I used it because I’ve always viewed it as a differentiator between Beeminder and other services. But you are right that my argument is at risk of becoming circular.
It’s not in the past – that’s a quote from the current website, live today. The blog also has many articles still referencing the old philosophy.
I don’t know how to answer the question then. I basically never see ads on the Internet because I never browse without an adblocker. Not trying to be a smartass, it’s just a fact.
For me it’s not so much the cost as the decision energy (hassle) involved. I’m happy to give money to folks doing work I appreciate - sometimes even if I don’t get anything in return! But when I’m asked for money midway through something I’m trying to achieve, I reflexively say no because I’m annoyed at the interruption. (See also: being prompted to sign up for a newsletter, or forced to watch an ad, or anything else that interrupts my train of thought). The value I place on avoiding hassle is pretty high - I’ve probably spent/forfeited thousands of dollars over the years in the name of convenience. I’m not actually sure the value I get from achieving my Beeminder goals would win by comparison.
That said, I’ve invested enough energy in this conversation at this point that I’m finally going to wander off and get a plan - as a task entirely separate from trying to create a goal. But if I hadn’t kept engaging with this conversation (and therefore had it at the top of my mind at a convenient time), I likely would have gotten stuck in a loop of “try to create goal, hit paywall, ragequit” until Beeminder stopped being useful to me.
@andrewlu, so much good stuff here. Thank you so much!
As to a/b pricing tests, we really wanted to but it felt too icky. But we came as close as possible. I feel like it tampers with the scientific integrity of the experiment to publicly describe the tests we’re running but I’d vaguely characterize it as one step less icky than what @andrewlu proposed. There’s no bucket of users paying less than the advertised prices. We do give out a lot of discounts but we always find a specific excuse to do so.
I also agree that price discrimination is fully ethical (despite how shady it probably sounds if you’re not familiar with that econ concept ) and I’m thinking about ways to do it more, like student discounts. I also like the idea of discounts to reward public testimonials. Or discounts that people can give out to friends. (Let me know – email@example.com – if you want to jump on any of that now.)
To @drtall and @andrewlu, regarding Tenet the First (that Beeminder be free forever, not a crippled or time-limited version, but real beeminding), I guess I convinced myself that that isn’t violated just because you can’t create an unlimited number of goals. Like if there were a limit on datapoints or a limit on how long you could beemind for free, that would contradict “Beeminder is free forever as long as you don’t derail”. Like imagine that something in the interface or server capacity had meant a limit of 3 goals from day one. The Beeminder in that universe is much worse, yes, but “Beeminder is free forever if you don’t derail” is just as true in that universe. Speaking of which, I’m wondering if 3 vs 7 free goals could make all the difference in the perception here.
 Self-plagiarizing a footnote from an old Messy Matters post: Price discrimination means charging more to people who are willing to pay more. The trick is that people who are willing to pay more don’t freely admit this. Airlines manage to discriminate by charging more if your trip doesn’t span a Saturday (since that means it’s likely a business trip and business travelers tend to be willing to pay more). Grocery stores do it with coupons (by your willingness to cut out coupons you reveal yourself to be more price sensitive). Companies often discriminate by offering two tiers of service (like Windows Professional Edition vs Windows Home Edition, or whatever). Student/senior discounts are another classic form of price discrimination (the assumption being that students and seniors tend to be more price sensitive).
It might be worthwhile to differentiate between the different kinds of comments here, to get a sense of who discusses what from which perspective:
Let’s start with the business perspective: Some people are trying to figure what would make beeminder viable, what could be tried to make users like or at least tolerate the pricing changes and what beeminder can or can’t do ethically. This takes two forms: an insider perspective and outsider perspective. The insiders are the people who are running beeminder. The outsiders are simply users.
The users perspective: Some people are describing the situation for the users. Some of those comments are individual reports and some of them are trying to abstract from their immediate case to a general picture of who the beeminder user is (how the user is constructed) and how this pricing change acts on that user. Regardless of the style of comment “time of being with the service” is an important variable even though it has very different outcomes in the types of comments.
Generally speaking though, what people trying to figure out is how a stable pricing system could look like that seems:
a) viable (so the service sticks around)
b) stays true to comments/slogans made earlier by beeminder
c) convincing on its own
d) reliable (in terms of generating revenue and in terms of being predictable for the user)
This “search for the perfect pricing scheme” is either directly or indirectly articulated in all of these comments so far and it seems to me, that proposed solutions can be divided into those who think that beeminder has the general outlook of the pricing right (that is: have a free tier/trial period and premium) or that they don’t (scratch premium completely and ask for support payments instead, make payments for features goal based, etc.). It seems to me that mostly people are suggesting in a “yes and” way: Most want/are (begrudgingly) fine with to keep premium around (in one form or another) but they also want $change. This $change can for example be a different payment feature yet not implemented or a pricing change.
Now, what is actually happening around this discussion is the following: Participation in the discussion (no matter how you are participating) is rewarded: free goals, steep discounts, etc. are seemingly happening all the time to those people who are engaged participants in this discussion. Here’s the kicker and I’m not sure @dreev is aware of this: The actual pricing scheme for beeminder seems to be very much unstable and is much more individual then what is getting discussed here and that not only in the sense that this is implied in trying to figure out pricing but also in a very general sense of approach to doing business. Take as a different example the pricing scheme of headspace and notice that there is no discussion about pricing taking place, no obvious discounts (apart from paying for longer amounts of time) are being advertised and so on. I’m not saying that beeminder should do it like these folks do, but I’m saying we’re seemingly all discussing (as the ideal pricing scheme) something like headspace’s scheme whereas beeminder’s general approach to doing business is much more open and as a consequence of that much less straight forward as well.
In short: the opacity of pricing (and the problems of reliability for users and the company) is a consequence of the approach to business that beeminder follows. Discussing an “ideal pricing scheme” is not taking into account this fundamental difference that sets beeminder apart from other businesses. Beeminders more democratic approach to pricing means also, that a stable pricing scheme without differences in “class” between users will not come (we already have grandfathered in users, users on a discounted plan, users who pay in full, etc. etc.) and therefore should be embraced.
I think everything you wrote is true, perhaps minus the consideration of new users:
My fear is that new users, fresh to the website aren’t privy to the personalized discounts. If the Beeminder pricing scheme only meets the 4 goals you outlined when personalized discounts are applied, then the website cannot grow, which means it is in fact not viable.
So the question, then, is whether the personalized discounts are merely addressing “time of being with the service” issues or addressing fundamental flaws. I don’t know the answer to this question, but I think it is the question to be asked.
(copying my response from the daily beemail)
Yes, 4$ per month is a very small amount for a continuous user and beyond well deserved for the service you get back. Yes, the “beeminder should be free” rule is slightly disturbed by the limit on goals. Still we have to face that above all the service needs to continue to exist. In order to make this change more smooth I think you need to give people more options. Some people don’t like subscriptions so give them ways to pay for one time improvements to what they can do. I already posted in the thread about bringing back the “$1 for 10 goals” feature. The numbers can be different of course. One time payments can also act a gateway drug to a subscription…