Developers building a software business on our own terms.
Ben: John, I love your shirt. I have the same shirt. , that's why I love it.
Josh: What shirt is that?, I can't see. Your microphone is blocking it. Okay, cool. What company is that? Or what—what is the shirt?
Ben: It's the “creative process” shirt and I don't remember where I bought it now.
John: It might be, yeah. Cotton Bureau. I go there for like dev tees every like six months and just buy a bunch of new ones and throw them on. So yeah, that's probably where.
Ben: Yeah. It's almost as cool as your Flipper shirt, but for those who are listening and aren't right in front of the computer, the “creative process” shirt, it has a big old squiggle, and then it goes out to a straight line. And so it's basically like what the process is: you’re all over the place, all over the map, you're trying to figure things out. And then eventually you hone in and now you're ready to go and you're off and rocking.
Josh: Yeah. I really liked that. I want one too. I was going to say those are like some very tasteful scribbles, yeah.
Ben: Yeah. And Josh, I know you have high standards for shirts. You'll appreciate the blend of this particular shirt. It's a nice feeling shirt. So, yeah.
Josh: I do have extremely high standards for shirts.
John: I also have similar highs—high standards too, just so you know. And my dad was like, “Oh, I got this t shirt, but I bought a large and it doesn't quite fit me. Do you wear a large?” And I was like, “Yeah”. And he's like, “Okay, here, I'll give it to you”. And I was like, “Well, I'm not necessarily going to take it. Like, can you just, well, let me check. And it was next level. And I was like, “Okay, yep. Yeah, I'll take it. That's fine. Yep. We're good”. So same thing, not even a gift.
Ben: Nice. Well, uh, in case you haven't noticed, we have John Nunemaker again with us on the pod today. We're excited to have with us. Thanks for coming back, John.
John: Love it.
Ben: I think at this point we can just stop introducing you because you're a regular. Everybody knows your voice now, so.
Josh: Yeah, the third host.
Ben: The third host. Yes. And, Josh, you have a bit of news tomorrow, you have some little rugrats heading back to school, right? Your summer is almost over.
Josh: Uh, this—actually today, this morning was the first day of school. So we got the, whatever, got the chalkboards out and took some school pictures in the yard and sent them off. This is actually a perfect day to come back to the podcast because now the house is quiet for the first time this summer, so yeah, but they're excited and it's a third grade for my daughter and first grade for my son.
Ben: Now was he, was your son, was he doing full day kindergarten? Or was this the beginning of full day for him?
Josh: Uh, no, he was full day before. And same school. They're both really familiar with where they're going, and we did like the, whatever, back to school night last week, got to see their classrooms and meet the teachers and everything, they got to pick their desks.
Ben: Oh, that's cool.
Josh: Yeah.
Ben: So, our school system, we don't start until next week. We’re, like, the latest in the country, I think.
John: We are on week three. We had a Wednesday start to the week. So not, so this week is the third week, last week was the second at the week before that was just a Wednesday, Thursday, Friday, which is amazing for the first week. It's like just enough where they get tired, but then they can recover over the weekend.
And then the last full week was last week and it's first and fourth, almost identical as you. And so the fourth grader, my son, he's good to go. He's just like, yeah, I'm with my friends again, uh, you know, first graders, she is definitely tired, definitely tired, but it's fun.
Josh: Yeah. Our school does that too. Like, it's a lighter week this week. And then, uh, I think they have Friday off maybe. And then it starts full time next week.
Ben: Yeah, it's good to ease back into things. You know, when you come back from vacation, you're not a hundred percent, right? You gotta get back in the mode. So I think it's cool to do that with the kids too.
Josh: Yeah, I'm sure it's nice for the teachers as well because they probably have a lot of , extra, extra work to do, like for the first week, getting all the kids organized.
John: I thought you were going to go into like, it's nice to ease back into your podcast when you've been off for a month, like—
Josh: That's, yeah, for—
Ben: All right. You went there. All right.
Josh: Yep. You two recorded a podcast without me, yeah.
Ben: That was—
Josh: It has been a little while for me. Yep.
Ben: Yeah. it has been a few weeks.
Josh: Is this really this is the start of season six, because I feel like there's been a little bit of a reset over the summer and we don't really have a strategy for picking seasons. It's just like when we feel like it's a different season. I'm not sure if we're there yet, but yeah, at some point we will be.
Ben: I had the same thought this morning when I was setting up the recording session. I'm like, “Oh, is this, it kind of feels like a new season”. Maybe it's because we've been off for a month, so. Maybe, maybe this will be episode one, season six, we'll see.
Josh: Yeah. We'll see. I think the other options we might wait, because we normally take like the month of December off as a company and that's always a good opportunity cause you can kind of restart a fresh season for a new year. We'll see what we want to do.
Ben: But yeah, we took a little bit of a podcast break because, well, let's see, Josh is on vacation. And then we had summer slow down thing. And also we've been working really hard on getting alarms done for Insights. And it's taken a little longer than we expected. And so we're heads down on that. So yeah, it's kind of a confluence of events.
Josh: Yeah. Yeah. And when just, yeah, we've been trying to make the most of the diminished summertime that we have to work as well. So I've been, like, extra focused on trying to get that deep work in and all that.
Ben: Yeah, we definitely have a cyclical kind of thing happening when you're in the Pacific Northwest. The summers are hard to get a whole lot of work done because the sunshine and the good weather is so tempting. I know for me, like every day, I want to be out on my bike and not in front of the computer, so conversely, in the wintertime, it's much easier to get work done because like, ah, who wants to be outside? Like, it's dark at two in the afternoon or whatever, so it's like, “Ah, just hunker down and do some work”.
Josh: Yeah. It works. It works well up here. Like summers are very low—the low tide—be outside and vacations and all that, and the kids are out of school.
John: Oh yeah. I think we do the same thing. It could just be like, mentality/business size also. Because like Box Out similar, like December, I'm usually MIA. I'm just—I’m gone. And I mean, the same thing in the summer, we try and plan. big, long trips and things like that. It's not that way for everyone.
Because it's kind of like tax season for sales. So the sales side is really busy in the summer because that's when all the renewals are July 1st and that kind of timeframe, but for on the dev side, we don't want to break stuff, right? When people are wanting to renew. So, we just scoot out and go on trips anyway. So summer always ends up slower and like same young kids, they're running around. It's good to spend time with them and, , all that kind of stuff. So.
Josh: Yeah. I gotta—I want to try that RV life you got going on over there. That's pretty cool. I saw your pictures and it looked like you guys had just a blast and how many, you were gone for like—
John: Yeah, this was, yeah, this was a month. So it's like 30 days. Yep. But let,
Josh: That was like, yeah, back in—
John: June to mid July, basically. Normally it's like beginning of June, like right when the kids got out of school, like we leave like the next week, but we just, for various reasons, couldn't this year. but we've talked about maybe Oregon, Washington next year. So just throwing that out there, okay.
Ben: Yeah, bring it.
Josh: maybe we'll have a live—
John: Okay, now we need to do it just for that. I mean, I feel like that. Yeah. Yep. That would be a great idea. But the problem is we have to do six weeks. So, it had to be like at least six, maybe eight. Because it's like we've done almost all the way to the West Coast and in six weeks. And it was a hustle. I mean, it was a hustle to get around. And we're like, we're all the way out there. We probably should hit California too. And so then, it's really a hustle. So I don't know, we may just be gone all next summer. We'll find out.
Ben: Well, once, once you get out here, all the way out to the Pacific Northwest, then, you know, the next level is, take it on the road to Iceland, right? You gotta, like, cargo lift that bad boy and just—
John: That would be cool. Yeah.
Josh: That'd be cool. Wonder if you could fit that on a ferry, like on the, whatever, the Puget Sound ferry to Victoria.
John: We've ferried our camper. Yeah. And there's a couple of places where, like in Michigan, and some other where you—you basically, the only way to get across is to ferry it. So you know, pull in and on this massive boat and you have a massive camper on it. It's weird, but it works.
Josh: That's cool. Sounds like fun. Yeah.
Ben: Yeah, we have some pretty good ferries over here, so you have no shortage of options
John: We don't have very many. We have like, I think Ludington maybe in Michigan, it's like kind of halfway up on the west side of Michigan. There's a— like a ferry there, I think, but yeah.
Josh: Yeah, we like the ferry to Canada, you can take it from Seattle or a few of the other like coastal—there’s a few ports that you can get on it and then it goes, yeah, to various places. And then also the San Juan islands, and other parts of the Puget sound. It's just, it's a really nice area.
Ben: We’ll plan on in, next summer, we’ll see you here. But, between now and then, you're gonna be raking in the big bucks because you just overhauled your pricing on Flipper.
John: Just about done. We preemptively posted all the articles and stuff just to— let’s start with that and just see what the reaction is. And then we'll go to the next version, which is, okay, now we can start letting customers switch to it. And then the next version, which is it’s all live and, you know, slowly in a way to, like, get it out and get feedback. So it's not just like this shocking thing, but we definitely over engineered it. That's for sure. And I'm okay with that, because again, pricing is just one of those things that's always sensitive and scary.
And so having the control to be able to like, put anybody on price, any pricing version for like a little period of time and stuff. Just—it makes you feel warm and fuzzy as a developer. So.
Josh: Yeah, I was reading the two blog posts—so you posted a blog post about this on the Flipper blog and we'll put these in the show notes, and then Garrett did a post on his blog and I was reading yours. And then I read his and I loved both of them, but yours dove into the why are you were choosing the specific, pricing and then he was much more focused on the implementation, like even code examples.
And I appreciated that as well, because yeah, we might steal some of that. I don't know—I was going to say Ben and I were just doing our shaping session for the next few cycles—actually like the rest of the year. We realize there's two cycles left in this year for us, if we do that, like, MIA in December, we're back to pricing ourselves and we're thinking about that as well. This was timely and I'm sure it'll be helpful. But yeah, it seems like Garrett's got some pretty flexible modeling going in your internal.
John: It's definitely been painful. Because going totally changing from either like metered or usage or like seat to fixed plan, it's very different. And then all your language and like—what I love about Garrett is he cares about all the verbiage. And so it has to be like, “Well, this is seat limited. This is not seat limited. This is seat limited with a max. This is seat limited without a max.”
And all those kinds of like subtle things that, like, I'm way too lazy to do. So, there's a reason he wrote the tech side and I wrote the business side. Because I was like, hand wavy, here's what I want to do. And then Garrett went hard on it for a while and actually wrote all the hard stuff. But I think it’s—I don't know. I think it's in a good place. And I can feel like—it was very intentional to split off business and other. And we also very intentionally, and it's fascinating, but the programmer side is what took off, like Garrett, you know, he's a stinker.
He doesn't have analytics on his blog, which would—I mean, that drives me insane. I can't live without analytics. He was a front page of Packer news for quite a while and stuff like that, which is cool, I think, from the standpoint of like “When's the last time a Ruby post was on Hacker News for any period of time.”
And it was funny as he submitted it like the day of and nothing happened. And I submitted it the next day and it went to the front page. So, I mean, who knows? And I didn't know he had submitted it. I was just like, “Hey, I'm going to submit it”. And I submitted it. And he's like, I already did. And he's like, “Well, it doesn't matter”. And then it popped off over there. So no comments. So he—he assumes everything was correct. And no one had any issues. if you get no comments.
Ben: It’s perfect
Josh: Of course.Yeah, it was like functionally perfect.
Ben: Yeah, it's like Venus rising from the sea, right? It's just fully formed
John: That's, you know, if you don't get a comment, you know, that you're right. I mean—it’s just the most—that’s the most guaranteed way. Someone commenting and saying you're right, you can't even trust that. But you know, if you get no comments, you know you're right. So—so yeah.
Ben: Yeah. That's a mic drop on Hacker News.
John: He was kind of like, you know, what he was like—should we post this on the Flipper blog? And I was like, “Here's my gut sense”. This is coming back to a marketing thing. And I'm curious what you all think about this as well. But I was like, I don't think that posts like that do well on company blogs. Blogs—like I get the goal for SEO and the goal for these other things.
But like, if you're both in it, if you’re—you know—both owners, whatever, and you're in it for the long run, like it's better—honestly, I think it's better these days just to build up a personal brand and, you know, send people, to your company, than to actually do it on the company site. Just the stuff that I put on the company site and nothing happens.
I mean, it's just nothing because people see a company name and they're just like, “Bahhh, I'm not even going to read that. They're just going to talk about how great their thing is, even if it's totally unrelated”. Whereas when I put stuff on my personal site, , then it will do great. , I've even had posts rejected that were, like, all I did was like—If I had to put this on my personal site, it would have done great. And I've submitted to whatever, Lobsters or something like that. It gets rejected. Because they're like, Oh, it's a company,
Josh: Yeah, we've run into that on Reddit and put—yeah, because we do a lot of, like, developer content marketing stuff. But I felt the same tension when I write—personally write something about the business or that's maybe like not a direct company announcement, but yeah, it's, I don't know. I've seen like the other thing companies will do if they're large enough, I guess, is have a separate engineering blog, which I've seen a few recently, but you gotta be like publishing a lot of engineering posts, which is something we actually want to do more at Honeybadger, but I'm not sure if it would make sense for us to have a separate blog.
John: So I told him, I was like, just put it on your site. Because it will do better there. And if it does better, that's better for us. More people will click the links, you know? And so he's been doing that with a few of them recently. Like anything that's, it's related to Flipper, but, and honestly, that's what I did back in the day on Rails tips.
It's what I did on ordered lists. It's what—so I'm like, that thing actually works. I just forgot that it worked and then just was like, well, I'll just put it on the Flipper blog because of whatever. But I'm realizing, I think that's better just to keep that for company related things or business related things, stuff like that.
And then focus more of the engineering stuff or things that would be like, for lack of a better word, it's cringe, but like lead gen, you know, like things that are like your goal is you want them to read this and then be like, I should go use Flipper to do this. For those kinds of posts, it's way better just on a personal site.
So that's what he did. And I think it has worked well. And we did the same thing with—he just posted a associated objects post. And it's also great. And both of those things, I think, Garrett’s really good at writing that stuff up and doing the figures and the explanations and all that. He goes through the details like a lot more thoroughly than I would.
I lose steam and it was like, “Ah, I'm gonna ship it, hit a button”, but yeah, the business side was definitely like, that was like—I was just—I think everybody right now is SAAS priced out. Like just, you think about even if you have a company that's doing well—it’s like there's stuff that you enjoy paying for and there's stuff that like you love the product, but you feel annoyed every month. And that's what just, what caused this is I'm like—Flippers young. It's new. It's not established like you all, or like Box Out, like my other software. So, like, we can play around with this stuff and see what works and what doesn't. But I was like, I just don't like per seat pricing.
I just don't like it at all. When I'm using software like that, and I go to add a person, I have to think every time and be like, is it really worth adding them to this product or not, or is it really worth sending all of this data to this system? If I'm metered on it and the metering is like really trying to find that point of extraction where that's the most the customer will pay.
Like, I just—I don't like it. I love the, you know, 37 signals. I was looking at the BaseCamp pricing and I was like, screw it. That's what we're doing. I just—I liked it a lot. They do per seat, but there's a max. And I was like, that’s—I like that, and again, it's different, you know, they're like, there's only so many things a person can click and do and type a message and all that.
So that makes their product easier to put a maximum on. Flipper, Honeybadger, people will send all the errors in the history of the world, or they’ll pull the crap out of your system or send a bunch of telemetry data. There's things that can get it a little more out of scope.
Josh: We're much more tied to costs, like the costs of usage in Basecamp does not necessarily equate to exponential increase in what it costs to have that customer, which, I'm incredibly jealous of. I would love to have that type of app, but our, I guess our skillset is just accepting all this data throughput.
John: Yeah, I love that they're like—they could go for maximum extraction. They have the social proof and all the other things, and they don’t—they're just like, “Look for 300 bucks, we can do it and it's worth it to us. And it's worth it to you. So like, let's just, let's chop it off there.”
So that's the idea. And also we had started doing like the pro gem, where people can install a gem that people understand, like the Sidekiq model and stuff like that. And yeah. And the main reason is that, like, for compliance or adding a new vendor is the thing that's sometimes tricky at a bigger company and adding a new gem is very easy, especially if it's a thousand bucks a year or something like that.
So we had a couple of customers come around that. So I was like, okay, yeah, let's just make something and see what happens. So we did. And then all of a sudden I was like, dang, I just got a thousand bucks. And I'm like, that's five seats for the whole year all up front. And I was like, that's the other thing that per seat pricing is not real great at is volume and annual kind of stuff.
And I was like, I don't like that either. I like just paying a fixed cost as a business owner. I love paying—like once I know I love a piece of software and I'm going to keep using it, I love paying like a thousand to 3000 bucks and just forgetting about it. you get a couple months free, you get maybe some other benefits.
I think there's peace of mind in that. And it makes me like that product even more, and I get that with Honeybadger a lot. Like it's very rare that we go over our error quota and then have to upgrade for a month or something like that. It doesn't really happen. And so I feel like I feel more joy with that than other products where maybe I go over every month. And then I'm like, “Oh, not again. now I've got to spend an hour trying to cut this a little bit”. Or, that, that was the driving force was like, what's like a fair price. And then also what do I get annoyed by? And how can I just make that not a thing for any customers that I have? And so that's what drove it, yeah.
Josh: Yeah. I liked that you started with “what do I appreciate”, because I think you're a lot like us in that regard, especially with Flipper, that you are the customer—you’re the first customer of your product. And so hopefully what you like, your customers will also appreciate. And it's cool that you have other businesses that are also customers, so you can actually get it from the other side, but yeah, it seemed like a good way to extend that.
John: It helps a little bit. There's just a few that—we use this image editor, I think it's called Pintura or something like that, they’re like a thousand bucks a year. And I'm like, I pay them immediately every single time, because I'm just like, “Yeah, okay. I love the value. I get way more value than a thousand dollars”—hopefully none of them listened to this podcast, and I'm like—they’re getting a grand, I'm getting a value at—both sides are actually really happy in this scenario. I think that's actually what you want.
You want both sides to be really happy. Like we don't have to be at this place where it's like, well, begrudgingly, I'm giving you my, whatever, but I don't want to, but I am, so I really liked that model. So, that's what I was trying to achieve was basically like, what's pricing that I would be happy paying for the functionality that I get from Flipper, number one. And then like number two, how do I make it easier for people to pay me up front, which is a good thing in a startup.
If you have something that’s not crushing it and it's just growing slowly, it's good to have that money up front because then you can use it to help grow it faster. So those are the—I think—the main summary points, but—
Ben: Yeah, I like what you said about the fatigue—the SAAS fatigue, because definitely people are feeling that I think; not only with the number of subscriptions they're paying for every month, but also just the past couple of years, especially in the tech, it's been kind of tough on people getting laid off and trying to cut costs and all this kind of thing.
And so you have what you experienced then is a bunch of your account manager giving you a call or an email saying, Hey, let me talk to you about your usage. And it's like, Oh, okay. You want to sell me more, right. And getting the screws turned on them, it's like, “Hey, give us more money.”
It's like, I don't really want to give you more money. And yeah, we've tried to avoid that kind of feeling with our pricing in Honeybadger. Like, with the insights thing, one of the keys was, “hey, we really like Splunk's approach to being able to analyze a bunch of stuff. We really hate how Splunk likes to bill people.”
Right? Because it's Datadog, right—it’s just like, “How much money can we take from your wallet?” And that's not a great feeling. And like you said, you don't really want to continue to have a relationship with someone that you feel like you're just being abused by on, on a monthly or yearly basis, right?
Josh: Yeah. I was going to say, it's interesting, because, like, for our business especially, we have this, heavy, like the usage base—like, we can't not charge for usage. We've tried for years and it didn't work out so well, but at the same time, I think you're dead on that. Maybe the key is not thinking about the pricing during your day to day usage of the product. I think that's the goal because that's like where for me, that's where it gets annoying, is when I have to have those decision or inflection points when I'm just using the product and I'm not in a mode of wanting to evaluate costs or whatever.
So I wonder if—if you are a usage-based pricing model, maybe the key is finding that—the middle point or the mid point of “I'm providing so much value for what you're paying me”, that you don’t—your mind doesn't go there when you're using the product, even if you are technically increasing your costs by, whatever, sending more errors or—or whatever it is. It is harder to do that when you have something like ,with seats where you're inviting a user, that's a frequent thing you do, using the product and it's kind of hard to avoid that.
Ben: Yeah.
Josh: But I wonder if that's the key is providing over the value.
Ben: Yeah, when I launched RailsKits back in the day and I did the SAAS RailsKit, I was like, “Okay, what do I price it at?” And my motivation was I want this to be no brainer pricing. I want someone to look at the price and they would say, yeah, totally. I get it. I'm going to buy that because it's worth X amount of dollars to me, and you're charging one quarter of that or whatever it is.
I just wanted people to not have that objection and be like, “Okay, yeah, that's a deal”. And I think we've tried to do the same thing with Honeybadger. Although, yeah, usage, having real costs, it’s different from a piece of software that you can just ship to someone they run it themselves, right? So yeah, there's some factors there, But that’s—that was my goal.
Josh: I'm trying to remember what the cost—what the price was. Because I actually—I remember, like, looking at your pricing for RailsKits and going through a little bit of the decision process. Like, I had considered buying it. I might've actually bought it at one point. I forget, but it was appealing to me at that point, I remember. And I was, like, not—at that point, I was not spending money on much software related things too much. I was probably fairly sensitive, but I do, yeah, remember.
John: Well, I gotta say as a new Insights customer, I really like how you did daily limits. Because I upgraded and then I like blew through my limit because I wasn't paying attention to when I was logging and I was like, “Oh”, and then I upgraded to like a big limit. I don't even remember what it was. And then I was like, “Oh, actually I don't even need this much”.
This is a daily limit. I thought it was a monthly limit, and once I realized that I was like, "Oh, this is awesome". Especially for this kind of stuff where you might really need it at a certain time, but you might not be looking at it most of the time, but you still want that data going in there—it’s okay if it chops off towards the end of the day, like most of the time or something, or it's just very practical, pragmatic. I liked that a lot for metered pricing. Like I—I think that's a really thoughtful thing rather than doing this really long period, just chop it down to like a day or, something like that. I think that's what it was.
Ben: Mhmm. Yep, yep.
Josh: It's a day. Yeah. That reminds me, Ben, we should do that for errors when we get, you know, when we address it.
Ben: Yeah, yeah. We—
Josh: We want—we wanted to do that.
Ben: We spent a lot of time thinking about Insights pricing and the daily thing came about because of, again, my frustration on being on the other end, being a buyer. And it's like, do I have to pay for so much extra just in case I need it? Or, on the other hand, do I have to get cut off on the third day of the month because I'm on a certain plan and now I've used it all up.
And it was like, well, how can we avoid that kind of pain with Insights? And so let's try daily, because if you blow with your daily quota, it's like, ah, okay, you're probably fine with that because tomorrow's a new day, right? And it’ll be back to normal. Or you're like, well, I really need all that. And so you upgrade for that day and then the next day you can drop back down again, So I'm glad to hear that that’s worked for you.
John: I don't know what the right word would be, but a faster feedback loop for me as a customer, because if I have to wait 30 days or two weeks to get some estimate of like how much data I'm going to send to you, and then I sent way too much—now I need to go back in and need to cut some data out or I need to up my plan.
And so like the feedback loop is so far—I think probably from the business side, it's good because it hooks people in. Because now they're like, well, I'm already committed for this long, so, I'm just going to go in. But from the customer side, the daily is so much better because again, if you mess it up, you know, right away and you have a pretty good idea.
Like it, you know, within 24 hours, I'm like—yeah, I’m under my limit. I'm cool, yeah, I know exactly which plan I'm going to be on and I'm ready to commit to this. So, I think that’s is a really interesting, choice. I'm trying to think of anything else that meters daily, like it's not a common thing. So, I think it was really thoughtful and I gotta say, I like it personally, so.
Josh: Mhm. Awesome. Yeah. I hadn't thought as much about the—I mean, we thought about it a little bit, like, of how it helps figure out how much you're going to need, because that is—that’s like the problem. The other problem with our pricing model that we've struggled with is how do people know what they're going to—yeah—what they're going to use. and that's something I'm still iterating on and I'd love to have it like an actual, like, useful calculator of some kind that we could put on the pricing page that like helps people figure that out. But in reality, a lot of times you just don't know, so I think that's really a great feedback from you.
John: One random idea when you said that is, you could, again, throw a log rage in, get like a standard rails request, a bite size, and then apply that to, like, a rate just to be like, look, if you're a typical rails app that does 10 requests a second, and you are logging this type of information, this is what plan you're going to want to choose, that something like that might help, okay.
Josh: I thought, I, yeah.
Ben: Yeah.
Josh: I think so. Yeah. , I'd also like communicating the fact that this is low stakes to get it wrong. I think if we can bake that into our message, that this is what you should expect, but if you get it wrong, you've got a day, a worst case, and you can try again the next day, or whatever, yeah.
Ben: Yeah. I was talking to a customer who was just getting started and they were really concerned about the costs of Insights in particular. And they're like, “but I don't want to show up at the end of the month and have a 10, 000 bill from you guys”. And I'm like, “Well, that's not how I work.”
You can, you set your limit and then if you hit the limit, well, then you're done. Like, we're not going to charge you more than whatever you agreed to do. And if you decide, oh wait, I don't want to be done, well, then you can choose that you want to spend more and get more quota. And like, he was starting to feel better about that.
And he's like, okay, but what if like I add the fifth user? And then all of a sudden you're going to charge me 10,000 and I'm like, no, no, no, no, that's, that's not how I do it. We don't do per seat pricing, you know, and I'm thinking, I can imagine the various vendors that he's dealt with in the past, you know, and I can picture them, the names in my head about who has charged this ridiculous per seat pricing or who has given him the surprise bill at the end of the month because of the amount of data that he sent, you know?
And, so I think by the end of the call, he was like, Oh, perfect. Okay. I'm feeling a little bit better about that. Because, yeah, we're not here trying to take advantage of you. We're just trying to make sure that we make a profit and that you have a great product, you know?
Josh: Once you get the people , in the same room, , you can really convince them that, but I think like John's right that like people are just expecting to just be bled. And who wants to get on a call, and spend an hour talking, for some demo or something to be hit by a, whatever, 2, 000 bill the next month.
Ben: I mean, to your point earlier, like you don't need to maximize the amount of revenue you get from each customer, right? You can find a happy place where you're making a good profit and your customers are happy to pay you because they're getting good value.
John: I think when you have investors or when you have other things going on that, are pushing you to grow, grow, grow, that that's why you get in that scenario. Because then it goes all the way through, but like when you have whatever you want to call all of our businesses—I would call them sanely run. I don't, we don't have to call them lifestyle or anything like that, but like, calm, there we go. That's a good one. Yeah, I'll use calm. So like, when you have a calm company, I'm like, Box Out hasn't changed pricing in five years. Like, all our competition has, we're at the—at this point now, like we're not the highest price solution anymore.
And part of me is like, well, maybe we should up our prices just so people understand, like we do feel like we're better. But the other side of me is like, again, if you can get someone who's just this value that you're getting is so much more that you're just like, yeah, I'm just add to cart.
It's fine. Like, I’m going to do this, our biggest plan is like 2,400 a year, I think, or something like that, which is like, I mean, that's not that big. Like, there’s schools that, I mean, massive schools, Notre Dame, et cetera, that use us and some of them pay more for custom stuff and like, that's fine.
But like for most people, they can spend the same amount as somebody else and get the same result as like a top tier school or something like that. That's a huge deal for them and fits in their budget. So, it's just about having some way for them to get started. So like free trial or a low plan, and then some larger plan.
I think the one thing we've talked about, I don't know if you guys, I guess you kind of do this because you have meter pricing, but like we've talked about having like a monster plan, like a plan that like nobody would buy just because maybe some people will buy it and that might, and it will make sense for them.
It'll have the value in it and they have the budget and they just didn't know that they could put the budget towards that, but just so people understand, like, look, we're not the cheapest. We just happened to be a very well run company and we're all doing fantastic. We don't need to charge you this crazy amount because we don't have growth estimates that are getting set by a board that's occupied by other people, it's just us.
I've been thinking about that a lot. And that's what I've been thinking about with Flipper too, is—I’m like the same thing. I'm like, I just want something that's fair. The only two costs that can increase on that are basically the meterable stuff, which would be how often do they pull to get synchronizing the data back and forth and how often do they push the telemetry data to us?
And if those two things are the two things, well, we can get around that by just putting limits in place. Like you're talking about, that's a very easy problem to solve to say like, look, you can only pull us every 60 seconds, and if you want updates that are faster than that, just use a web hook.
That's really easy, we have that already. You'll get instantaneous updates, it's the right way to do it. don't just set all your thousand servers sitting there every 10 seconds hammering us, like that doesn't make sense. And the same thing with telemetry, we can set limits on how much telemetry we receive.
But even that, the biggest thing is, like, not receiving the request. It's more like the crunching of the data. I was looking recently, like, we've been storing all the telemetry data, like for all time, we haven't shut any of it down or actually enforced any limits.
We store it for free people, paid people all the same. And so, again, I'm lazy. So we were like, okay, I looked at the table and it's like, okay, the bill is starting to get to like five, 600 bucks a month, which not a big deal, except, for this app and the size, I was like, that's kind of more than it should be.
It should probably be in the 100 to 200 range. And I was like, okay, we're getting charged based on storage. And all of our raw data is that like 300 to 500 gigabytes or something like that. And I'm like, it’s just raw data we don't need—we’ve compressed it all into hourly aggregates.
And it's like the raw data is like 300 some 500 some gigabytes. The compressed data is like 200 meg. At that point, you're like, well, if you just delete all the raw data, cause you don't need it anymore or put it in cheaper storage, like S3 or some other file system, whatever, then you have immediately dramatically reduced the cost of it, and therefore the cost that, you have to incur or pass on to the customer.
And so I'm like, these are all these problems are very solvable problems. And I would rather have be forced to have to solve them by having a lower price, that's like a great value than to just be like, I’m going to be lazy and just pass on more costs so that I don't have to, you know, truncate data or aggregate or think about stuff or learn Click House or whatever, so.
Ben: Yeah. I think what you said about limits is key and Josh kind of alluded to this earlier. Like, when we started out, we didn't really have limits on some areas that drove a lot of our costs. And we had to learn the hard way that limits are a good thing and charging for usage is a good thing.
But over time we've learned that whatever new feature we deploy, it's got to have a limit. we never deploy something that's unlimited and maybe the limit is high, and at a point where we think no one will touch it, right? Or it's low, and we've already planned to give people a way to buy more.
For example, like our, you were talking about polling, like, we have a limit on the number of poll requests you can make to our API. I can’t remember what it is right now. It's whatever per minute, right? Or per hour, whatever. And for 99.9 percent of our user population, it's fine. They never notice.
They use the API and, whatever, but for, like, literally three customers, they've been like, “Hey, this limit is pretty low. We're trying to do this and we can't do it because your API is not allowing us to read data fast enough”. And I'm like, okay, we can bump you up, but you know, if we didn't have that limit, then like we would, we don't even know, like, we could be spending tons of resources on serving someone who's pulling every second for no good reason.
So, yeah, that's key, I think to help keep your costs down and keep your pricing reasonable. Yeah.
Josh: Lately we've been trending towards starting with low—the low end for limits. Because even if you're not going to charge for more, like, you can always expand it later. And it's like, people are getting something new. You can write another blog post about it and be like, we're increasing the limit for everyone, whatever. Merry Christmas.
John: Agreed. We even had some of those discussions on, like, how we break down some of the pricing and we're like, ah, I don't know. Because we ended up the tiers, the fixed tiers will have, like, seat limits just because it is additional cost to us with more developers because we do personal environment synchronization.
And so, like, when they're working on their laptop, they're making requests to us and stuff like that. From that standpoint, we're like, okay, here is some factor of cost. So I think we, I don't remember what we ended up doing. Maybe it was like 10, 20, 50, or something, everybody was currently like, I don't think we have anybody who has 50 seats and we might have one or two customers who have 50 seats, but most of the people are lower than that.
And so we're like, it's, that's a future problem. And if people, you know, hit that limit, like. It's really easy. they'll say, “Hey, I hit this limi”t. And it's kind of like that AWS style. AWS has hidden limits all over the place. And then you talk to a rep and you get a price and you get it worked out.
And so I feel like that model of like, here's a low limit and then you can customize it and raise it up. It makes a lot of sense. And we went back and forth on the number of seats and we're like, you know what? I was like, let's just start low. Same as what you guys were just talking about.
Let's start low. And we can always raise it. That's fine. If we find that we've done something wrong, but when you lower it, that's, people don't like that as much. and I wouldn't like that as much. And I think the only other thing that we did, is we incentivized yearly plans. I was like, look, I want people to pay yearly. So I'm like, if you want the—let’s say it's one, two and three—if you want the middle retention for analytics, you can pay yearly and then we'll give you a higher retention for that. And I think that's cool.
Josh: Yeah, I thought that I—yeah, I wrote that down when I saw that. I was—I thought that was really cool. Because, you know, it's standard to offer like a discount, like two months off or something, but I hadn't seen like actually including extras in the annual, and it makes total sense like that. That's a great incentive and I'm sure it'll get more takers for annual.
John: That's the hope. I don't know. That might've been a shower thought. I don't know. I was just like—all of a sudden I was like, what if we like—because again, I sold, like, one program and right as soon as that happened, it's like “today I learned”. I was like, I just got all the seat payments basically upfront.
How can I do this more? It's like Mikey likes it. And I was like, uh, okay. And then we were right then we were talking about all the pricing stuff. And I was like, Garrett, What if we up the retention periods for, like, analytics? Like, we'll store—instead of a day, we'll store a week or a month, if you buy yearly, because I'm like, at that point, they're committing to us.
I'm happy to give them more. Like, they've made a commitment. It's fair on both ends. And obviously we do two months, like Box Out, that's three months free because we're like, you know what? That's a lot of schools. Schools have three months off. We don't want them to think about canceling and re-upping and all that kind of stuff. We'll just give them three months off. So they don't even think about it during the summer, when they're not using it.
Ben: Yep.
Josh: That's cool. Yeah. I'd almost want to frame that. Like I like pitch it as like, you get the summer for free, like, schools out schools out on pricing or what, or something, you know?
John: Yeah, we okay. I'm gonna steal that. I'm gonna write that done.
Ben: Love it.
Josh: Yeah, please do. Yeah. Glad I can contribute to something. Because I'm actually writing notes in our shaping pitch for our pricing, because we'll think about some of this. But yeah, I think this is good. I did want to say, just one—an extra benefit when you start with the low limits is that it also forces the people who care get in touch with you, like you'll hear from them if they're hitting a limit.
And oftentimes those conversations wouldn't happen if you just started with a super high limit that no one is—everyone’s going to like, obviously. So, sometimes you get those conversations and then it helps you figure out like, where should this limit actually be?
Ben: Yeah. And another bit with the starting low and going high later versus the other way around. I think , that hits into the loss aversion thing. It's a economic thing where people like much more sensitive to losing 10 than they are to gaining 10, or whatever it is. So yeah, yeah. Definitely start low.
John: We do that also with, like, actors—like the number of actors you could add to an individual feature way back in the day, I set the limit of like 100. I was like, I'm just gonna pick a number, set it to a hundred. This is like my personal goal. And then we'd have people who are like, “I'm trying to import and I've got like 5,000 actors”, so then we basically just made it so we could up the limit for certain people and stuff.
And, like, over time we just have naturally upped the limit. So, we had to do that less often. But it's just like little things like that, especially where it's like something where someone might shoot themselves in the foot or cause problems. Like you can just put a low limit in and be like, look, just trust me.
You're probably not going to want to go over it. And if you go over it, you're going to want to talk to me, so you understand why you're going over it. So that, and that’s—it's happened. I mean, we've had people reach out and they're like, “Oh, we want to go, we want to add more actors”. And then I'm like, “Hey, human, nice to meet you.”
And you get to, like, talk to them and that's really fun. They get a good touch point. Cause most of the time, again, developers are trying to avoid talking to you because they think they're going to get squeezed, you know?
Ben: Right.
Josh: Having those points, like where you force someone into a conversation, especially with developers, I think is just essential. We can't get people on the phone any other way, unless they want something from us. And so, you know, create more of those.
John: Yeah. We actually had a bug one time that like made it, I can't remember what it was, but like, so people had to reach out. Because they're like, oh, like something is just really not—and we're like, We should just do this every month. Add this bug every month just to like start meeting the customers.
Because it wasn't terrible. It was just like, just enough that people were like, “Yeah, I need to let them know about this”. We were really tempted about, like, bug driven development for like a little while just, and it was like, no, we can't do that. But maybe once in a while we'll see. So yeah, that's a good point.
Ben: So I think that the takeaway for this episode is if you want to be a company's favorite customer, just call them up out of the blue and say, “Hey, I want to chat to your dev team and talk about the product”. They will love you forever.
John: Oh, at least on DevTools. Because DevTool is so hard. Like on other—you know, again, it's totally different on Box Out because everything we do there, it's magical to the end user. They're just like, I don't even comprehend how that can happen. So, there, it’s less, I would say it's less where we're like, yes, please talk to us.
Because we get a lot of people who are like, well, I'm sharing a team password with everyone. And it's annoying for this reason. And we're like, just don't do that. We let you add as many people as you want. “Oh, really? How do I do that?” You know, it's like, so it's, there's definitely, know, your audience, so like if your audience is devs, you should definitely break stuff once in a while and force them to converse with you and have low limits. Yeah.
Ben: Make him talk to you. Yeah. Well, and as we all know, devs are the best customers.
Josh: They’re like a totally different species of customer.
Ben: But the best one, yeah.
John: Like, when you make them happy, you know, you've really done something well, yeah.
Ben: Well—
Josh: It's a, yeah. It’s like a really hard fish to catch. Like, it's like, you know, this is not, this is not like some trout. This is like, I don't know. What's the rare. Yeah. Something rare. I did want to ask, so, least favorite pricing of all time. Do you all have any—does anyone come to mind or someone that you're to throw under the bus?
Ben: I have a least least favorite pricing. I'm not going to, I'm not going to say the name, but you can guess who it is because it's someone that we all know, but a company that prices not only per user, like per seat, but also has different kinds of users, like they charge one rate for someone that gets to log in and see a chart, but then they charge a different rate for someone that logs in and sees like more charts, I bet, or whatever it is, like a tier based on level of access
Josh: Mhm.
John: Well, that mixed with like metered usage is just like, I feel like it's gutting. Yeah. I mean, I—again, like I've used—I’ll say the name because I don't care, but I've used New Relic forever and I loved them and they put me on a free plan, like way back in the day. Like, I think it was Lou himself put me on a free plan just because I was writing up and having graphs and Rails tips and stuff.
And I was on that plan forever. And then they're like, “Hey, you need to pay”. And I was like, no problem. I've had a free plan for so long. This is fair. And this is right. And then it was like, I'd like to add a second user. Oh, that'll be like a thousand dollars. And I was like, what? I'm like, uh, okay. I mean, that's the kind of stuff, like when it's you're talking about tiered.
And then also the data coming in is also getting to—it’s just, when you add all those things up, and then you're like, you're just, yeah, it's super frustrating. That's, I think that was, I want to say five or six months ago, maybe a little more now when I went through that.
Okay. I—this is actually—we’re going to change Flipper’s pricing because that was kind of like [the straw that broke the camel's back. I was just like, no, I'm not gonna, I don't want anyone else to use my software and feel this ever. So we're going to find a way. Because we talked about maybe doing analytics as a separate thing based on how much data is coming in, in addition to the per seat, because that's what other people do and I was just like, no, like, I—yeah. So anyways.
Ben: Mine was also New Relic.
John: I had a feeling it was, so I thought I'll just take the blame off you. It's my fault. New Relic, you can come after me or just change your pricing and then we'll talk good things. Yeah.
Josh: So John is also going with New Relic.
Ben: Yeah.
Josh: Okay. So for me, yeah, New Relic is a good one, but Netlify—I will also say the name—we use Netlify. I like a lot about Netlify. but the thing I don't like is their pricing and we have even left and come back, like, grudgingly, because we like the product so much, or it works so well for us otherwise.
And we've had to like, basically hack our usage a little bit around the pricing, but their pricing model is seat based, like many of their competitors. They're not unique in that but they also add a component of Git contributors are added to the seats, but they have it—so basically, if you have, say you have a GitHub repository, that's like building—so in our case, we have a middleman site that, like, builds our website and our blog. And we're using just, like, Markdown blog posts or whatever. Anyone who has a Netlify account that can get into the UI and use it directly as a seat.
But then also anyone who contributes—like, anyone who opens a pull request and gets—it gets merged basically is a seat for that month. Which, I appreciate that it's not, like, just added as a seat forever, or they're not charging for your entire GitHub organization, but, it feels like the next level down of like, how can we like just capture, everyone who is, like, doing anything touching this repo.
And in our case, the real slap in the face was when we got billed for our bot user on GitHub that we have that like, just, you know, makes a commit occasionally that, like, updates version numbers, for example, because we display the latest SDK versions on our docs site, for example. We’re getting charged, like whatever, like 25 a month for a bot that is just updating like a JSON file once a month.
And yeah, it's just—it’s like super frustrating. And then we have like other people who like write blog posts and might submit a PR, so every time like anyone who's not already on the team submits a PR for a blog post, we get charged for them too. That's my rant.
Ben: It's pretty galling that their business is serving a static webpage and they charge per user, who puts that webpage up on their servers. That’s just ridiculous.
Josh: Just not, yeah, it's not tied to reality.
John: And you have to believe that like somewhere, someone did just—did not expect this use case. That's what I like to think. And if we just reached out, maybe they would be like, Oh geez. Okay. Well, we'll just give you like whatever, a hundred dollar credit every month or something to fix this problem.
It reminds me of like the Netflix sharing thing, and the proliferation of that. And so like, it sounds like that's maybe why they added it. It was like, well, they just have one person in there to like approve things, but they've got a hundred people over here in army that’s doing their stuff.
But I'm like—for us, I'm like—even though it's on Flipper, I'm like, we have a few people who like do a shared one to get under the, like to get one seat. And, I literally reach out and I'm like, just add your whole team and I'll charge you 20 bucks. I don't care, but please just use the product, right? I don't care if that's all you want to pay.
I'll just do that just because—but don't use the product wrong. Because then you're just feeling pain. And you're going to associate that pain with it. And that's another reason why I'm just like, okay, we're just going to do fixed plans. You know, it's fine.
Ben: Yep. Totally.
Josh: Yeah. Well I did reach out to Netlify, and as I recall, they did not help me with any of this. So, um, so yeah, I guess Netlify fix your pricing.
John: Did you build a headless CMS or did you like use one that's out there? I'm just curious.
Josh: I'm actually using a Statamic, which is a Laravel based CMS. And it is awesome. I got to say , Rails doesn't have anything that compares that I've seen. And it’s—I wish it did, because I do like Rails apps, but I am on my first layer, like production Laravel app is in—is running, so, which we're now using with Insights, which is cool.
So we have something to monitor. But yeah, it's cool. It's like all flat file by default still. So it’s—it’s the familiar, like whatever, it's a little bit like Jekyll, except it has a CMS on top of it and, it can do, like, hybrid style. You can serve an actual, like, dynamic website with it because it's Laravel or you can generate it does like static generation and you can deploy it that way.
John: Because I pronounce everything wrong. And so like in my head, it's always been statomatic. and yeah.
Josh: That's same here, actually it's that Statamic was, like, the one I had to, like, correct, but I'm pretty sure that's not right. Do you know what Addison says and how Addison reads it, Ben? Satanic. So, wow, it is a strange name which is, I don't know if it, that helps or hurts it. Maybe it's the, one of those, like, things where it's quirky enough that it's like, I mean, we're talking about it right now for like 10 minutes on a podcast. So who's the evil genius here?
Ben: That makes me think that our next podcast episode just needs to be all about product naming. You know, we talked about pricing today.
John: Yeah.
Ben: Product naming.
Josh: Yeah. Because we're geniuses at product naming, obviously, with Honeybadger.
John: I just watched Despicable Me 4 and there's a honey badger in it. And I thought of you for like a good 10 minute period while watching the show last Friday night with the kids. So.
Ben: That's awesome.
Josh: That's very cool.
Ben: Cool. Well, any more wise words you want to drop on us, John, about pricing?
John: No, I mean, just make it so that it works for both sides, you know, I feel like that's people just, and honestly people, again, overthink it. They just don't over—I was overthinking it for a long time. Finally, I was like, screw it. Let's just do a couple fixed tiers. Three tiers are so easy. People like looking at, you know, am I small, medium or large?
Some people are going to buy large cause they want the best. Some people are gonna buy small cause they want to get started with the cheapest and most people are going to go in the middle because, well, they don't want to buy the cheapest or the most expensive. So I just think it's easy to process. Don't make people think about your pricing. So just make it simple. And if it's simple, then they won't think about it a ton. That's where I landed.
Ben: Love it.
Josh: Love it. Cool. Well, this has been FounderQuest. find us at FounderQuest podcast-dot-com, go rate us and review us and go check out John's new pricing when it drops and sign up for sign up for Honeybadger. We'll catch you next time.