When should you give up or pivot?
How do you know when to cut your losses or to pivot to something else? This week, Colleen learned that people are more excited about how she's building her service than the actual service, and Michele talks about a product that came back from the dead.
Michele Hansen
Welcome to the Software Social podcast where we invite you to join our weekly conversation about what's going on in our businesses. I'm Michele Hansen.
Colleen Schnettler
And I'm Colleen Schnettler.
Michele Hansen
How's it going, Colleen?
Colleen Schnettler
I'm doing well. It's been a pretty exciting week for me on a tech front. So I haven't had as much time this week. I'm really trying to commit to talking to more people. I mentioned that.
Michele Hansen
Yeah!
Colleen Schnettler
And it's kind of funny, because I've been talking to more people and they don't really care about my image management solution.
Michele Hansen
Oh :(
Colleen Schnettler
Well, it's okay. This this kind of interesting thing has happened, where they care at there. They don't care about that, but they're really excited about the work I'm doing with AWS and Arc. I'm, I'm moving my tech over to go from kind of a monolithic Rails app to like AWS, Lambda serverless functions and everyone is really interested in that?
Michele Hansen
Really?
Colleen Schnettler
Yeah, it's kind of been a funny thing where I've wanted to talk to people about image management. And they want to talk to me about how I'm using lambda functions to do manipulations on images.
Michele Hansen
Really...
Colleen Schnettler
And so, yeah, and so I there's a few things I'm trying to be wary of here. One is, I feel like it's not a good idea to use new tech on a new idea. Because you're, it's like a double uphill battle. You have to both build something, ship it and learn some new tech. But this AWS stuff is really cool. I love learning it and the architecture with it really is better. It just feels cleaner. I'm I'm not a huge microservices person, but this particular, what I'm building really lends itself well to these lambda functions. So that's been really fun. But it has been an interesting thing to find out that people are more interested in that. So as you know, one of the things I struggle with is jumping too quickly from idea to idea. But I'm seeing this opportunity where everyone wants to learn how to write lambda functions in the cloud with Ruby. And I'm like, Oh! Well, that's interesting.
Michele Hansen
Did you course business at one point? I forget.
Colleen Schnettler
I thought about it. I know that would be, that would be no, I thought about it. And I started doing video courses. I think I did like three or four. And it just got, I mean, it's kind of a slog to do that kind of thing, especially if it's not, it was I was building something that was more geared towards beginners. And so it was pretty boring for me to actually make them because you have to really go back and I felt like there was a lot of beginner course content out there already. So I never I never really gave that a good shot. But this is an interesting thing, because I am really when you hear these people who are successful, they're like, Oh, I was interested in this thing. And I blogged about it everyday for a year. And then boom, I'm the expert in this thing. And I have a product and this thing. I don't know, it's just I don't want to be jumping, right? This is I need to really focus, which is one of the things I struggle with. But I'm really tempted, I'm really tempted to like, get more into this AWS in the cloud serverless functions stuff. That seems to be more interesting to my developer friends.
Michele Hansen
Maybe it's not jumping if you just hang out on a particular stepping stone for a little bit. Maybe there's like a mini course here or something or even if it's just a blog post and people get excited about that. But maybe there's, you know, a little a little stop along the way here.
Colleen Schnettler
So I'd really like to get the image thing done. Because I think as we've been talking every week, I feel like I'm just so close.
Michele Hansen
And you need it. Like for your sanity. On the projects you work on.
Colleen Schnettler
Yes, I need it, I'm going to use it on my clients. And I just want to finish something like I just want to finish a side project. That's just where I am. It's like graveyard of abandoned side projects. So I want to finish it, and I'll use it. And I think it'll make my life easier. So that's great. But I wonder this is something I'm kind of like putting in the back of my mind and kind of tabling as people are interested in this, they want to learn more, they want to learn, you know, how you can use it. And the other thing, there's, I feel like serverless is huge, and there's so much opportunity for growth. And I don't know if there is like, I know, AWS has a marketplace, but I don't know anything about it. But so for example, this one serverless function I have all you do is if your images in my bucket, you can just send query parameters for the size you want of the image and it returns the image already resized and optimized and things like that. Wouldn't it be cool if you could have like a suite of little serverless functions that you could somehow bundle together? Like a whole bunch, I don't know, you'd have like, I don't know what it would look like you'd have like five different API endpoints you as my customer could hit that would handle all of your different image manipulation, like, or I could have one for video and one for images. I don't know. I feel like there might be something there. And I don't know what that looks like. But it's definitely like tucked away in the back of my mind, like what I'm working on. Now. I want to finish first. That's important. And it's so close. But I think after that might be a good time to start blogging about all this stuff I have learned in the AWS ecosystem. And once I don't forget it, but also, you know, to see if people are interested in that kind of information as well.
Michele Hansen
It sounds like they are and like, I can't tell you how many projects and products come out of setting out to do something else. Like just earlier this week, Adam Wathan did a huge thread on the development of Tailwind. I don't know if you saw it. And he mentioned how they were working on another side project, but like the CSS was driving him nuts. And so they just he just made that. And then he open sourced it. And people were super psyched about it. And they were really surprised by that. And then it like led to this whole thing that is Tailwind. I mean, that happened with Geocodio too. To to like we set out to make an app that told you grocery store and coffee shop opening hours and ran into problems with geocoding. And you know, we got some good, good traction and some excited users on the grocery store app. But man, when you find a problem that developers have, and you solve it for them, people get really, really excited about that.
Colleen Schnettler
Yeah. So when you guys launched Geocodio did you do that through some kind of existing marketplace, or were you just out in the free world?
Michele Hansen
Hacker News. That was the place to launch in January 2014. Product Hunt like wasn't even a thing. Yeah. And we did a ton of, you know, posting on like StackOverflow and Reddit.
Colleen Schnettler
And how did you know when you were ready to put it out there for other people? Had you guys been using it for your grocery store app for a while?
Michele Hansen
Yeah, I think at least six months or so.
Colleen Schnettler
That's quite a while.
Michele Hansen
Yeah. And it took a good friend of ours to be like, hey, like, you should just slap a paywall in front of this and see if other people will pay for it. And then at least you don't have to pay the server costs for it. And like maybe enough people will use it that that server is effectively free for you. And we're like, oh, yeah, that would be amazing. It was we -- were making three or $400 a month in ad revenue off of the app. And so we thought that was an amazing, huge success. I remember we had these conversations about how like, this is what's making money. We should focus on this, like, this geocoding thing is a distraction. I mean, I can't tell me how many of those conversations we had over the years. It's like this is the thing is making money. And this new thing is shiny and a distraction. And we need to make sure we focus on things making money. And this is, you know, the only thing that really ended up being a thing. Yeah, so many things come out of, you have another idea and you run into this foundational problem. And it turns out that everybody's more excited about that.
Colleen Schnettler
I and I've always felt like there's something in the AWS ecosystem.
Michele Hansen
It's so complicated.
Colleen Schnettler
Yes, it is so unnecessarily complicated.
Michele Hansen
Like, I don't know why we pay what we do for AWS. Like I literally like basically, there's no way for me to figure out why we are being charged what we are, and there's like third party websites you have to go to to even know what their pricing is? It's insane.
Colleen Schnettler
Yes, I agree. And so it feels like it's a huge that's always -- that has also been a pain. For me like this whole, this whole images, of course, but also all of the AWS stuff, so people are seem to be more excited, I have to say, like just talking to people and not that I've talked to tons of people, but I'm just trying to engage more in the communities I'm already a part of. And the image thing, as we've discussed, for most people, it's not frequent, it's painful, but they set it up once and they're done. Most people don't consult, so they don't see a lot of different clients. So they don't have to set this up over and over and over.
Michele Hansen
If you work for a big company, like you said, right, you said then you're done.
Colleen Schnettler
Yeah, exactly. But the AWS stuff is always evolving, and people seem really interested in the serverless stuff. And maybe that's just because it's a cool new hotness, but it seems like it's gonna work really well for me, and so, still focused on finishing what I'm working on, but I could see maybe getting -- writing a little bit about some of the struggles I've had with that and what I've learned with that once this image thing is complete.
Michele Hansen
Yeah, why not? Sounds like there might be something there.
Colleen Schnettler
Also, as a side note, our podcast has launched since we had our last episode. And so my non technical friends, my non technical friends have listened to the first episode, and they're so excited. And they say to me, Colleen, what is your idea? And when I tell them, they are so disappointed, like this huge. They're like, expecting me to be like, Oh, I'm gonna build the new Facebook or cure world hunger. And I'm like, Oh, I'm gonna help you put images on the internet. They're not impressed. It's been funny. It's been a funny response. So anyway, I think my goals for this week, I'm really, I'm really deep in the technical stuff. And I just want to get like I said, I'm just trying to get that done. So I think that's my goal for this week and just to keep engaging with people and talking to people.
Michele Hansen
Yeah, that's what you gotta do. Talk to people. See if your idea has interest.
Colleen Schnettler
Yeah, definitely. So what has been going on this week with Geocodio?
Michele Hansen
Gosh, well, you know, today was actually a really good day. I got two landing pages out which...
Colleen Schnettler
Nice!
Michele Hansen
Is it kind of work I enjoy. It's kind of weird though, because it defined scope. So it's good for if I want something I can knock out in an hour or so. But I don't actually find out the results of it for like, another month. So it's kind of funny to see Oh, wow, that landing page I wrote three months ago is actually doing really well. Like I should do more of those.
Colleen Schnettler
That's because it takes that long for the SEO to get back to you.
Michele Hansen
Yeah, it just takes time.
Colleen Schnettler
Yeah, so when you're doing a landing page, what do you use for that?
Michele Hansen
Uh, we just have a very basic Netlify CMS. And it's just, you know, it's just text and some screenshots. It's nothing fancy. But hey, that was just today, so I wasn't really thinking about. So I was just I just like enjoyed my work today. It was just kind of fun.
So what I was really thinking about this week. You know how a couple weeks ago we talked about projects that failed, products that failed, right? Like, we were like, Oh, we should do that another episode. So what actually ended up thinking about this week was products, like almost failed and then came back from the dead, but they didn't really die. And I don't know if that makes any sense whatsoever. I feel like I should back up here.
Colleen Schnettler
So maybe I'm a little confused.
Michele Hansen
So okay, let me. The high level where I got to what I've been thinking about is, recently I saw this Wall Street Journal article talking about how companies after the pandemic hit in March, they pulled back their software and r&d spending to the tune of 27% on an annualized basis. So software, r&d, etc spending is negative 27% since March, and I used to work in investing and one of the things I absorbed from that was that you should, you know, compare yourself to the market, right? So, you know, any, you know, portfolio manager or a fund you look at, it's always what what is their benchmark, right, like the S&P 500, or whatnot. And so I looked at that, and I was like, Oh, interesting, like, looks like a benchmark. I wonder how we did over the same time period just for accountability. You know, I always look at our numbers every day every week, but it's only once or twice a year that I really step back and look at like growth rates and what our customer portfolio looks like and think about goals for diversifying into different customer areas. So which is maybe a topic for another podcast, too. So anyway, I was inspired to run the numbers this week on our growth rates so far for January through the end of July of this year. And I gotta say, like, I love being knee deep in a spreadsheet, like it has been, it has been kind of like refreshing and kind of fun to be able to really like focus and just nerd out on the data like that, like, I really love it.
Okay, so where this gets to like dead products. So, you know, I calculate over our growth rates and everything, okay, really got me thinking about this was, somewhere we've seen a lot of growth this year is in our HIPAA compliant product. So, you've been to the doctor's office, you've probably signed a form about like, data and how they can, you know, access information and there's like HIPAA Privacy Practices and stuff like that. You've probably signed this form with every new doctor visit you've ever gone on?
Colleen Schnettler
Yes.
Michele Hansen
Yes. So this is this is basically, you know, GDPR for healthcare data in the US. And so two years ago, yeah, August of 2018, we launched a HIPAA compliant version of Geocodio thinking that it was basically the same product we were doing but a new market and there was a lot of potential because there weren't really any good HIPAA compliant geocoding options available. And so for example, if a doctor's office wanted to put their patients on a map, the coordinates are considered personally identifiable information under HIPAA and so they can't just use an off-the-shelf service for it. And so kind of like 2017 actually, right after I'd gone full time, we were thinking about like big projects to chew on. And it was either, do we go worldwide, or do we do HIPAA, we had gotten requests for it for years. And we decided you know, we're small team the world is big complicated place it's probably more straightforward for us to go into a new market but we already have the product all built out and everything at least from a you know, a UI and functional perspective and we had some interest. So we spent about a year rebuilding the product from the ground up working with a security consultants like we, like you know, it looks the same to the to the customer but it it everything under the hood is different, like data centers are different because we have to have special data processing agreements with them and all this stuff. So anyway, so I spent, you know, months and months and months on this, from the customer side of things, you know, gathering interest in it. Running surveys, figuring out pricing, like generating interest, all of this stuff.
So we're feeling pretty good about it, we launch it and August of 2018. And after like a month, no one signs up for it.
Colleen Schnettler
Oh, wow.
Michele Hansen
And this is a problem. Not only have we spent a year of our time on this and, and hours, like an hours of work. So the way our service works is we basically for the most part, we have two separate ways of accessing Geocodio. We have this public cluster, which if we have a pay as you go plan, the requests go there. And then we have unlimited customers who each have their own dedicated server. And they can basically run as much as they want on that for a flat rate. And it's also we pay a flat rate for that, so it works out for everyone. And what we really wanted to do is launch a pay as you go option for HIPAA and make it like super accessible and easy for people to do this. But that requires us to be paying for this public cluster. And we didn't have any customers.
Colleen Schnettler
Oh, I see.
Michele Hansen
And what is the first rule of bootstrapping? Always make money. You always have to be cashflow positive, you're not burning somebody else's money. You are burning your own money if you are losing it. And so we're like, oh, god, like what happened? And, you know, I ran some testing on the homepage, because I was like, What happened here? Like, like, where did we go? Like, we talked to all these people. We've done these surveys, like, where, you know, we had years of demand from people, like what happened? And so it turned out that, you know, I'd go through the website with everything, everything was fine. Okay, what would your next step be if you wanted to sign up? Like, oh, I would go talk to my legal department. And then it was like, Oh, no. So this is what I had underestimated, was that the sales cycle is not, somebody just signs up and they and they get going. It's it has to go through security reviews, it has to go through legal reviews. It has to go through lots of phone calls. And so I didn't realize this at the time, but it like it can take us a year to 18 months sometimes to sign on HIPAA customers.
Colleen Schnettler
Wow. That's a long time.
Michele Hansen
And when normally our average time from someone signs up to add a credit card is like less than 12 hours.
Colleen Schnettler Right? Wow.
Michele Hansen
But the median time to first add a credit card is less than 12 hours. So that sales cycle is really long. And this is something that can kill you. And I think this is something that the founders of Binomial have talked about quite a bit too is like sales cycle can be killer, especially for a small company. So anyway, we were like well, alright, so we had basically we had to shut down the pay as you go option for the product. We didn't have any customers on the Unlimited side, so the product was basically dead a month after we launched it because we couldn't burn money on these servers.
Colleen Schnettler
Sure that makes sense.
Michele Hansen
Yeah. So that's two years ago. And, but we kept it open sort of as an option, someone reached out as an unlimited option, which, you know, we only pay for the infrastructure, if somebody is actually using the product is actually, you know, has a subscription. So that works out for us. So, so, you know, it's sort of been very, very slowly growing over the past two years, but in March, we had a big uptick in interest in it. Understandably, there's a lot more interest in health related data. And so when I was crunching the numbers this week, one of my first thoughts was, what percentage of this is HIPAA related? Like and what would our growth numbers look like, if we didn't have that product? So do you want to dive into some numbers with me?
Colleen Schnettler
Yeah, let's do it. Yeah. Okay.
Michele Hansen
So compared to January to end of July 2019. Our growth this year so far was 59%
Colleen Schnettler
59%?!
Michele Hansen
Revenue growth, that's revenue growth came.
Colleen Schnettler
Okay.
Michele Hansen
Bear in mind, this is not, I'm not talking about net income, right. I'm not subtracting costs from it, and it's more expensive for us to run, you know, HIPAA, but this is just sort of...
Colleen Schnettler
Sure. That seems great, though.
Michele Hansen
Yeah, right. I was like, No, that's break really weirdly good. And I almost feel guilty that we're like, doing well and everything is terrible. So I was like, okay, that's so that's really good. Like we're comparing your benchmark is negative 27% like that's... That's good. So that's I was like, what would that be like without HIPAA? Like, like, has the pandemic really like and like how much has that contributed to this growth? Like, because we had one customer who signed up who -- whipped through everything and they got started in a day, which never happened before. They were like, We need this right now. Like get us set up, like somehow everything got approved, like overnight and it was it was that... you ever hear people talking about, you know, you have product market fit when people are literally taking the product out of your hands.
Colleen Schnettler
Yeah.
Michele Hansen
I was like, wow, okay, I've never experienced it quite like before.
Okay, so if you take out the HIPAA customers we have gotten since March, our growth for January to end of July is 41%.
Colleen Schnettler
Okay
Michele Hansen
So it dips but it doesn't go all the way. And then I was like, let's go back all the way.
Colleen Schnettler
Right.
Michele Hansen
So like we launched this product. You know, there are these issues with sales cycle where like, Oh, this is a colossal mistake and a waste of a year and and we are full of regret and angst about all that. So last year just for this was just the same time period, it's not full year. So our revenue would have been down 7% last year [ed: 7% lower]
Colleen Schnettler
without HIPAA?
Michele Hansen
Without it right. Without those numbers, I'm trying to like what happens if I pull out all of those customers basically, okay, in the data. So and then, so then if I took them all so if I took them all out entirely, then looking at 2020, our growth would only be 27%.
Colleen Schnettler
Wow.
Michele Hansen
So without this product that we thought was a colossal failure when we launched it, and was a waste of a year's worth of work. Actually, has made a huge like 32 point percentage difference in our growth this year.
Colleen Schnettler
That's amazing. So my first question for you, you have a background in economics, don't you?
Michele Hansen
Yes. How could you tell?
Colleen Schnettler
So isn't the first rule of economics...
Michele Hansen
The dismal science!
Colleen Schnettler
Isn't the first rule of economics, the sunk cost fallacy, like,
Michele Hansen
Oh, God, yeah.
Colleen Schnettler
Why did you guys walk away after after did not like what -- how did you make the decision once you were so far into it, to not walk away and to not give up on it?
Michele Hansen
Well, we basically did.
Colleen Schnettler
Right, okay. Right, because we shut down your public pay-as-you-go.
Michele Hansen
Yeah we shut down the public cluster. And if anyone wanted to sign up, they had to be on the unlimited level, which is still the case today. I mean, we maybe we could launch a pay as you go, but I'm just just given the sale cycle it's not... I feel no urgency to do that. I'd like to make it more accessible. But yeah, I mean, sunk cost fallacy is you've put a lot of time into something you know, the first rule when you're in a hole, stop digging.
Colleen Schnettler
Right.
Michele Hansen
So after a month, we pulled the plug on it.
Colleen Schnettler
Because it didn't cost you anything.
Michele Hansen
Yeah.
Colleen Schnettler
So so because it didn't cost you anything to have those two because you only pay for the private servers if you don't have the clients, right? Customers.
Michele Hansen
Exactly. Yeah.
Colleen Schnettler
Well, that's amazing how that worked out for you guys.
Michele Hansen
Yeah, it's just kind of like wild to see like something sometimes something fails. But, like timing is a huge part of that. Like, it reminded me how people talked about companies like Blue Apron and whatnot where I mean before I guess I haven't looked at them recently, but like before the pandemic I mean their expenses were out of control, like they're the unit economics on it, which is you know, the taking into account the what they spend on each customer in terms of the product but also in terms of marketing and then how much money they get back. were terrible. But now, things like Blue Apron or food delivery, right? Like people are seeking out those products much more frequently than they were before, like the need for those products has changed. And so instead of having to market all the time and give a lot of stuff away for free people are seeking them out and their unit economics are improving. And so it's like sometimes like the product is not necessarily a bad idea but the timing for it isn't right and it's a question of, if you can survive until the timing is right you know, it's kind of like that stock market quote what the market can stay irrational longer than you can stay liquid. I think I butchered that. But you know, so yeah, I don't know if you like the product like came back from the dead or what happened.
Colleen Schnettler
That's really interesting. Because it's so hard when you get started, or probably at any point in your business career to figure out when you've given something enough, because you hear so much, you hear so much advice that says you have to know when to walk away. Right? You have to know when this was a bust. Don't sink any more time in it. But then you hear stories like this where it took you guys two years to actually see a return of an entire year of work. If you didn't already have a successful business before that. You probably would have been even more upset when it didn't take off.
Michele Hansen
Oh, for sure. Yeah, I mean, I think the only thing that really kept it alive was that we could, you know, sort of throw it in the deep freeze for a while until some customers came along.
Colleen Schnettler
Well, that's great. That worked out.
Michele Hansen
Yeah, yeah. I mean, and you know, now I need to, you know, make sure I'm double checking all of my net income numbers. I'm sure they're, they're not going to be nearly as amazing because a big part of that is AWS probably.
Colleen Schnettler
Those jerks. <laughing>
Michele Hansen
It's wild to think about it. And we launched it basically exactly two years ago. And it just it took, it took a lot of time for that product to grow into its own.
Colleen Schnettler
And I think that's something that I struggle and I think I've mentioned this before, and I think other people struggle with is things take time. Just in general, like, things take time.
Michele Hansen
And that's okay.
Like it took us right, three and a half years to go full time on Geocodio -- just me.
Colleen Schnettler
Yeah.
Michele Hansen
But the key thing was is that we were not burning money through that time. Like we were always making money, even if it was only god -- I think our first month our margin was $10.
Colleen Schnettler
Wow.
Michele Hansen
Like, as long as you're not spending as long as you're not spending...I know, it's very impressive.
Dude, as long as you're not spending more money on it, then you're making and you're enjoying it.
Colleen Schnettler
Yeah.
Michele Hansen
...Then you can keep going. But if you're losing money, and it's your money. That's, that's, that's there's some tough, tough decisions to be made there. Yeah.
Well, I guess that's about it for this week. Thank you so much everyone for listening, by the way, as Colleen mentioned earlier that the podcast just went live this week, and we're pretty floored that people want to listen to us. I think me I'm speaking for myself, like, I'm pretty surprised that it's more than just a couple friends of ours listening. And so if you enjoy our show and want to help other people find it. We'd be incredibly grateful if you tweet about it, or can write a review for so it would mean so much to us.
Colleen Schnettler
Yeah, thank you so much everyone for listening and we would love to engage with you. So please feel free to reach out to our podcast Twitter handle which is @softwaresocpod, or either of us [@leenyburger, @mjwhansen]. We're both on Twitter and we'd love to hear what you think and any questions you have about the show.
Welcome to the Software Social podcast where we invite you to join our weekly conversation about what's going on in our businesses. I'm Michele Hansen.
Colleen Schnettler
And I'm Colleen Schnettler.
Michele Hansen
How's it going, Colleen?
Colleen Schnettler
I'm doing well. It's been a pretty exciting week for me on a tech front. So I haven't had as much time this week. I'm really trying to commit to talking to more people. I mentioned that.
Michele Hansen
Yeah!
Colleen Schnettler
And it's kind of funny, because I've been talking to more people and they don't really care about my image management solution.
Michele Hansen
Oh :(
Colleen Schnettler
Well, it's okay. This this kind of interesting thing has happened, where they care at there. They don't care about that, but they're really excited about the work I'm doing with AWS and Arc. I'm, I'm moving my tech over to go from kind of a monolithic Rails app to like AWS, Lambda serverless functions and everyone is really interested in that?
Michele Hansen
Really?
Colleen Schnettler
Yeah, it's kind of been a funny thing where I've wanted to talk to people about image management. And they want to talk to me about how I'm using lambda functions to do manipulations on images.
Michele Hansen
Really...
Colleen Schnettler
And so, yeah, and so I there's a few things I'm trying to be wary of here. One is, I feel like it's not a good idea to use new tech on a new idea. Because you're, it's like a double uphill battle. You have to both build something, ship it and learn some new tech. But this AWS stuff is really cool. I love learning it and the architecture with it really is better. It just feels cleaner. I'm I'm not a huge microservices person, but this particular, what I'm building really lends itself well to these lambda functions. So that's been really fun. But it has been an interesting thing to find out that people are more interested in that. So as you know, one of the things I struggle with is jumping too quickly from idea to idea. But I'm seeing this opportunity where everyone wants to learn how to write lambda functions in the cloud with Ruby. And I'm like, Oh! Well, that's interesting.
Michele Hansen
Did you course business at one point? I forget.
Colleen Schnettler
I thought about it. I know that would be, that would be no, I thought about it. And I started doing video courses. I think I did like three or four. And it just got, I mean, it's kind of a slog to do that kind of thing, especially if it's not, it was I was building something that was more geared towards beginners. And so it was pretty boring for me to actually make them because you have to really go back and I felt like there was a lot of beginner course content out there already. So I never I never really gave that a good shot. But this is an interesting thing, because I am really when you hear these people who are successful, they're like, Oh, I was interested in this thing. And I blogged about it everyday for a year. And then boom, I'm the expert in this thing. And I have a product and this thing. I don't know, it's just I don't want to be jumping, right? This is I need to really focus, which is one of the things I struggle with. But I'm really tempted, I'm really tempted to like, get more into this AWS in the cloud serverless functions stuff. That seems to be more interesting to my developer friends.
Michele Hansen
Maybe it's not jumping if you just hang out on a particular stepping stone for a little bit. Maybe there's like a mini course here or something or even if it's just a blog post and people get excited about that. But maybe there's, you know, a little a little stop along the way here.
Colleen Schnettler
So I'd really like to get the image thing done. Because I think as we've been talking every week, I feel like I'm just so close.
Michele Hansen
And you need it. Like for your sanity. On the projects you work on.
Colleen Schnettler
Yes, I need it, I'm going to use it on my clients. And I just want to finish something like I just want to finish a side project. That's just where I am. It's like graveyard of abandoned side projects. So I want to finish it, and I'll use it. And I think it'll make my life easier. So that's great. But I wonder this is something I'm kind of like putting in the back of my mind and kind of tabling as people are interested in this, they want to learn more, they want to learn, you know, how you can use it. And the other thing, there's, I feel like serverless is huge, and there's so much opportunity for growth. And I don't know if there is like, I know, AWS has a marketplace, but I don't know anything about it. But so for example, this one serverless function I have all you do is if your images in my bucket, you can just send query parameters for the size you want of the image and it returns the image already resized and optimized and things like that. Wouldn't it be cool if you could have like a suite of little serverless functions that you could somehow bundle together? Like a whole bunch, I don't know, you'd have like, I don't know what it would look like you'd have like five different API endpoints you as my customer could hit that would handle all of your different image manipulation, like, or I could have one for video and one for images. I don't know. I feel like there might be something there. And I don't know what that looks like. But it's definitely like tucked away in the back of my mind, like what I'm working on. Now. I want to finish first. That's important. And it's so close. But I think after that might be a good time to start blogging about all this stuff I have learned in the AWS ecosystem. And once I don't forget it, but also, you know, to see if people are interested in that kind of information as well.
Michele Hansen
It sounds like they are and like, I can't tell you how many projects and products come out of setting out to do something else. Like just earlier this week, Adam Wathan did a huge thread on the development of Tailwind. I don't know if you saw it. And he mentioned how they were working on another side project, but like the CSS was driving him nuts. And so they just he just made that. And then he open sourced it. And people were super psyched about it. And they were really surprised by that. And then it like led to this whole thing that is Tailwind. I mean, that happened with Geocodio too. To to like we set out to make an app that told you grocery store and coffee shop opening hours and ran into problems with geocoding. And you know, we got some good, good traction and some excited users on the grocery store app. But man, when you find a problem that developers have, and you solve it for them, people get really, really excited about that.
Colleen Schnettler
Yeah. So when you guys launched Geocodio did you do that through some kind of existing marketplace, or were you just out in the free world?
Michele Hansen
Hacker News. That was the place to launch in January 2014. Product Hunt like wasn't even a thing. Yeah. And we did a ton of, you know, posting on like StackOverflow and Reddit.
Colleen Schnettler
And how did you know when you were ready to put it out there for other people? Had you guys been using it for your grocery store app for a while?
Michele Hansen
Yeah, I think at least six months or so.
Colleen Schnettler
That's quite a while.
Michele Hansen
Yeah. And it took a good friend of ours to be like, hey, like, you should just slap a paywall in front of this and see if other people will pay for it. And then at least you don't have to pay the server costs for it. And like maybe enough people will use it that that server is effectively free for you. And we're like, oh, yeah, that would be amazing. It was we -- were making three or $400 a month in ad revenue off of the app. And so we thought that was an amazing, huge success. I remember we had these conversations about how like, this is what's making money. We should focus on this, like, this geocoding thing is a distraction. I mean, I can't tell me how many of those conversations we had over the years. It's like this is the thing is making money. And this new thing is shiny and a distraction. And we need to make sure we focus on things making money. And this is, you know, the only thing that really ended up being a thing. Yeah, so many things come out of, you have another idea and you run into this foundational problem. And it turns out that everybody's more excited about that.
Colleen Schnettler
I and I've always felt like there's something in the AWS ecosystem.
Michele Hansen
It's so complicated.
Colleen Schnettler
Yes, it is so unnecessarily complicated.
Michele Hansen
Like, I don't know why we pay what we do for AWS. Like I literally like basically, there's no way for me to figure out why we are being charged what we are, and there's like third party websites you have to go to to even know what their pricing is? It's insane.
Colleen Schnettler
Yes, I agree. And so it feels like it's a huge that's always -- that has also been a pain. For me like this whole, this whole images, of course, but also all of the AWS stuff, so people are seem to be more excited, I have to say, like just talking to people and not that I've talked to tons of people, but I'm just trying to engage more in the communities I'm already a part of. And the image thing, as we've discussed, for most people, it's not frequent, it's painful, but they set it up once and they're done. Most people don't consult, so they don't see a lot of different clients. So they don't have to set this up over and over and over.
Michele Hansen
If you work for a big company, like you said, right, you said then you're done.
Colleen Schnettler
Yeah, exactly. But the AWS stuff is always evolving, and people seem really interested in the serverless stuff. And maybe that's just because it's a cool new hotness, but it seems like it's gonna work really well for me, and so, still focused on finishing what I'm working on, but I could see maybe getting -- writing a little bit about some of the struggles I've had with that and what I've learned with that once this image thing is complete.
Michele Hansen
Yeah, why not? Sounds like there might be something there.
Colleen Schnettler
Also, as a side note, our podcast has launched since we had our last episode. And so my non technical friends, my non technical friends have listened to the first episode, and they're so excited. And they say to me, Colleen, what is your idea? And when I tell them, they are so disappointed, like this huge. They're like, expecting me to be like, Oh, I'm gonna build the new Facebook or cure world hunger. And I'm like, Oh, I'm gonna help you put images on the internet. They're not impressed. It's been funny. It's been a funny response. So anyway, I think my goals for this week, I'm really, I'm really deep in the technical stuff. And I just want to get like I said, I'm just trying to get that done. So I think that's my goal for this week and just to keep engaging with people and talking to people.
Michele Hansen
Yeah, that's what you gotta do. Talk to people. See if your idea has interest.
Colleen Schnettler
Yeah, definitely. So what has been going on this week with Geocodio?
Michele Hansen
Gosh, well, you know, today was actually a really good day. I got two landing pages out which...
Colleen Schnettler
Nice!
Michele Hansen
Is it kind of work I enjoy. It's kind of weird though, because it defined scope. So it's good for if I want something I can knock out in an hour or so. But I don't actually find out the results of it for like, another month. So it's kind of funny to see Oh, wow, that landing page I wrote three months ago is actually doing really well. Like I should do more of those.
Colleen Schnettler
That's because it takes that long for the SEO to get back to you.
Michele Hansen
Yeah, it just takes time.
Colleen Schnettler
Yeah, so when you're doing a landing page, what do you use for that?
Michele Hansen
Uh, we just have a very basic Netlify CMS. And it's just, you know, it's just text and some screenshots. It's nothing fancy. But hey, that was just today, so I wasn't really thinking about. So I was just I just like enjoyed my work today. It was just kind of fun.
So what I was really thinking about this week. You know how a couple weeks ago we talked about projects that failed, products that failed, right? Like, we were like, Oh, we should do that another episode. So what actually ended up thinking about this week was products, like almost failed and then came back from the dead, but they didn't really die. And I don't know if that makes any sense whatsoever. I feel like I should back up here.
Colleen Schnettler
So maybe I'm a little confused.
Michele Hansen
So okay, let me. The high level where I got to what I've been thinking about is, recently I saw this Wall Street Journal article talking about how companies after the pandemic hit in March, they pulled back their software and r&d spending to the tune of 27% on an annualized basis. So software, r&d, etc spending is negative 27% since March, and I used to work in investing and one of the things I absorbed from that was that you should, you know, compare yourself to the market, right? So, you know, any, you know, portfolio manager or a fund you look at, it's always what what is their benchmark, right, like the S&P 500, or whatnot. And so I looked at that, and I was like, Oh, interesting, like, looks like a benchmark. I wonder how we did over the same time period just for accountability. You know, I always look at our numbers every day every week, but it's only once or twice a year that I really step back and look at like growth rates and what our customer portfolio looks like and think about goals for diversifying into different customer areas. So which is maybe a topic for another podcast, too. So anyway, I was inspired to run the numbers this week on our growth rates so far for January through the end of July of this year. And I gotta say, like, I love being knee deep in a spreadsheet, like it has been, it has been kind of like refreshing and kind of fun to be able to really like focus and just nerd out on the data like that, like, I really love it.
Okay, so where this gets to like dead products. So, you know, I calculate over our growth rates and everything, okay, really got me thinking about this was, somewhere we've seen a lot of growth this year is in our HIPAA compliant product. So, you've been to the doctor's office, you've probably signed a form about like, data and how they can, you know, access information and there's like HIPAA Privacy Practices and stuff like that. You've probably signed this form with every new doctor visit you've ever gone on?
Colleen Schnettler
Yes.
Michele Hansen
Yes. So this is this is basically, you know, GDPR for healthcare data in the US. And so two years ago, yeah, August of 2018, we launched a HIPAA compliant version of Geocodio thinking that it was basically the same product we were doing but a new market and there was a lot of potential because there weren't really any good HIPAA compliant geocoding options available. And so for example, if a doctor's office wanted to put their patients on a map, the coordinates are considered personally identifiable information under HIPAA and so they can't just use an off-the-shelf service for it. And so kind of like 2017 actually, right after I'd gone full time, we were thinking about like big projects to chew on. And it was either, do we go worldwide, or do we do HIPAA, we had gotten requests for it for years. And we decided you know, we're small team the world is big complicated place it's probably more straightforward for us to go into a new market but we already have the product all built out and everything at least from a you know, a UI and functional perspective and we had some interest. So we spent about a year rebuilding the product from the ground up working with a security consultants like we, like you know, it looks the same to the to the customer but it it everything under the hood is different, like data centers are different because we have to have special data processing agreements with them and all this stuff. So anyway, so I spent, you know, months and months and months on this, from the customer side of things, you know, gathering interest in it. Running surveys, figuring out pricing, like generating interest, all of this stuff.
So we're feeling pretty good about it, we launch it and August of 2018. And after like a month, no one signs up for it.
Colleen Schnettler
Oh, wow.
Michele Hansen
And this is a problem. Not only have we spent a year of our time on this and, and hours, like an hours of work. So the way our service works is we basically for the most part, we have two separate ways of accessing Geocodio. We have this public cluster, which if we have a pay as you go plan, the requests go there. And then we have unlimited customers who each have their own dedicated server. And they can basically run as much as they want on that for a flat rate. And it's also we pay a flat rate for that, so it works out for everyone. And what we really wanted to do is launch a pay as you go option for HIPAA and make it like super accessible and easy for people to do this. But that requires us to be paying for this public cluster. And we didn't have any customers.
Colleen Schnettler
Oh, I see.
Michele Hansen
And what is the first rule of bootstrapping? Always make money. You always have to be cashflow positive, you're not burning somebody else's money. You are burning your own money if you are losing it. And so we're like, oh, god, like what happened? And, you know, I ran some testing on the homepage, because I was like, What happened here? Like, like, where did we go? Like, we talked to all these people. We've done these surveys, like, where, you know, we had years of demand from people, like what happened? And so it turned out that, you know, I'd go through the website with everything, everything was fine. Okay, what would your next step be if you wanted to sign up? Like, oh, I would go talk to my legal department. And then it was like, Oh, no. So this is what I had underestimated, was that the sales cycle is not, somebody just signs up and they and they get going. It's it has to go through security reviews, it has to go through legal reviews. It has to go through lots of phone calls. And so I didn't realize this at the time, but it like it can take us a year to 18 months sometimes to sign on HIPAA customers.
Colleen Schnettler
Wow. That's a long time.
Michele Hansen
And when normally our average time from someone signs up to add a credit card is like less than 12 hours.
Colleen Schnettler Right? Wow.
Michele Hansen
But the median time to first add a credit card is less than 12 hours. So that sales cycle is really long. And this is something that can kill you. And I think this is something that the founders of Binomial have talked about quite a bit too is like sales cycle can be killer, especially for a small company. So anyway, we were like well, alright, so we had basically we had to shut down the pay as you go option for the product. We didn't have any customers on the Unlimited side, so the product was basically dead a month after we launched it because we couldn't burn money on these servers.
Colleen Schnettler
Sure that makes sense.
Michele Hansen
Yeah. So that's two years ago. And, but we kept it open sort of as an option, someone reached out as an unlimited option, which, you know, we only pay for the infrastructure, if somebody is actually using the product is actually, you know, has a subscription. So that works out for us. So, so, you know, it's sort of been very, very slowly growing over the past two years, but in March, we had a big uptick in interest in it. Understandably, there's a lot more interest in health related data. And so when I was crunching the numbers this week, one of my first thoughts was, what percentage of this is HIPAA related? Like and what would our growth numbers look like, if we didn't have that product? So do you want to dive into some numbers with me?
Colleen Schnettler
Yeah, let's do it. Yeah. Okay.
Michele Hansen
So compared to January to end of July 2019. Our growth this year so far was 59%
Colleen Schnettler
59%?!
Michele Hansen
Revenue growth, that's revenue growth came.
Colleen Schnettler
Okay.
Michele Hansen
Bear in mind, this is not, I'm not talking about net income, right. I'm not subtracting costs from it, and it's more expensive for us to run, you know, HIPAA, but this is just sort of...
Colleen Schnettler
Sure. That seems great, though.
Michele Hansen
Yeah, right. I was like, No, that's break really weirdly good. And I almost feel guilty that we're like, doing well and everything is terrible. So I was like, okay, that's so that's really good. Like we're comparing your benchmark is negative 27% like that's... That's good. So that's I was like, what would that be like without HIPAA? Like, like, has the pandemic really like and like how much has that contributed to this growth? Like, because we had one customer who signed up who -- whipped through everything and they got started in a day, which never happened before. They were like, We need this right now. Like get us set up, like somehow everything got approved, like overnight and it was it was that... you ever hear people talking about, you know, you have product market fit when people are literally taking the product out of your hands.
Colleen Schnettler
Yeah.
Michele Hansen
I was like, wow, okay, I've never experienced it quite like before.
Okay, so if you take out the HIPAA customers we have gotten since March, our growth for January to end of July is 41%.
Colleen Schnettler
Okay
Michele Hansen
So it dips but it doesn't go all the way. And then I was like, let's go back all the way.
Colleen Schnettler
Right.
Michele Hansen
So like we launched this product. You know, there are these issues with sales cycle where like, Oh, this is a colossal mistake and a waste of a year and and we are full of regret and angst about all that. So last year just for this was just the same time period, it's not full year. So our revenue would have been down 7% last year [ed: 7% lower]
Colleen Schnettler
without HIPAA?
Michele Hansen
Without it right. Without those numbers, I'm trying to like what happens if I pull out all of those customers basically, okay, in the data. So and then, so then if I took them all so if I took them all out entirely, then looking at 2020, our growth would only be 27%.
Colleen Schnettler
Wow.
Michele Hansen
So without this product that we thought was a colossal failure when we launched it, and was a waste of a year's worth of work. Actually, has made a huge like 32 point percentage difference in our growth this year.
Colleen Schnettler
That's amazing. So my first question for you, you have a background in economics, don't you?
Michele Hansen
Yes. How could you tell?
Colleen Schnettler
So isn't the first rule of economics...
Michele Hansen
The dismal science!
Colleen Schnettler
Isn't the first rule of economics, the sunk cost fallacy, like,
Michele Hansen
Oh, God, yeah.
Colleen Schnettler
Why did you guys walk away after after did not like what -- how did you make the decision once you were so far into it, to not walk away and to not give up on it?
Michele Hansen
Well, we basically did.
Colleen Schnettler
Right, okay. Right, because we shut down your public pay-as-you-go.
Michele Hansen
Yeah we shut down the public cluster. And if anyone wanted to sign up, they had to be on the unlimited level, which is still the case today. I mean, we maybe we could launch a pay as you go, but I'm just just given the sale cycle it's not... I feel no urgency to do that. I'd like to make it more accessible. But yeah, I mean, sunk cost fallacy is you've put a lot of time into something you know, the first rule when you're in a hole, stop digging.
Colleen Schnettler
Right.
Michele Hansen
So after a month, we pulled the plug on it.
Colleen Schnettler
Because it didn't cost you anything.
Michele Hansen
Yeah.
Colleen Schnettler
So so because it didn't cost you anything to have those two because you only pay for the private servers if you don't have the clients, right? Customers.
Michele Hansen
Exactly. Yeah.
Colleen Schnettler
Well, that's amazing how that worked out for you guys.
Michele Hansen
Yeah, it's just kind of like wild to see like something sometimes something fails. But, like timing is a huge part of that. Like, it reminded me how people talked about companies like Blue Apron and whatnot where I mean before I guess I haven't looked at them recently, but like before the pandemic I mean their expenses were out of control, like they're the unit economics on it, which is you know, the taking into account the what they spend on each customer in terms of the product but also in terms of marketing and then how much money they get back. were terrible. But now, things like Blue Apron or food delivery, right? Like people are seeking out those products much more frequently than they were before, like the need for those products has changed. And so instead of having to market all the time and give a lot of stuff away for free people are seeking them out and their unit economics are improving. And so it's like sometimes like the product is not necessarily a bad idea but the timing for it isn't right and it's a question of, if you can survive until the timing is right you know, it's kind of like that stock market quote what the market can stay irrational longer than you can stay liquid. I think I butchered that. But you know, so yeah, I don't know if you like the product like came back from the dead or what happened.
Colleen Schnettler
That's really interesting. Because it's so hard when you get started, or probably at any point in your business career to figure out when you've given something enough, because you hear so much, you hear so much advice that says you have to know when to walk away. Right? You have to know when this was a bust. Don't sink any more time in it. But then you hear stories like this where it took you guys two years to actually see a return of an entire year of work. If you didn't already have a successful business before that. You probably would have been even more upset when it didn't take off.
Michele Hansen
Oh, for sure. Yeah, I mean, I think the only thing that really kept it alive was that we could, you know, sort of throw it in the deep freeze for a while until some customers came along.
Colleen Schnettler
Well, that's great. That worked out.
Michele Hansen
Yeah, yeah. I mean, and you know, now I need to, you know, make sure I'm double checking all of my net income numbers. I'm sure they're, they're not going to be nearly as amazing because a big part of that is AWS probably.
Colleen Schnettler
Those jerks. <laughing>
Michele Hansen
It's wild to think about it. And we launched it basically exactly two years ago. And it just it took, it took a lot of time for that product to grow into its own.
Colleen Schnettler
And I think that's something that I struggle and I think I've mentioned this before, and I think other people struggle with is things take time. Just in general, like, things take time.
Michele Hansen
And that's okay.
Like it took us right, three and a half years to go full time on Geocodio -- just me.
Colleen Schnettler
Yeah.
Michele Hansen
But the key thing was is that we were not burning money through that time. Like we were always making money, even if it was only god -- I think our first month our margin was $10.
Colleen Schnettler
Wow.
Michele Hansen
Like, as long as you're not spending as long as you're not spending...I know, it's very impressive.
Dude, as long as you're not spending more money on it, then you're making and you're enjoying it.
Colleen Schnettler
Yeah.
Michele Hansen
...Then you can keep going. But if you're losing money, and it's your money. That's, that's, that's there's some tough, tough decisions to be made there. Yeah.
Well, I guess that's about it for this week. Thank you so much everyone for listening, by the way, as Colleen mentioned earlier that the podcast just went live this week, and we're pretty floored that people want to listen to us. I think me I'm speaking for myself, like, I'm pretty surprised that it's more than just a couple friends of ours listening. And so if you enjoy our show and want to help other people find it. We'd be incredibly grateful if you tweet about it, or can write a review for so it would mean so much to us.
Colleen Schnettler
Yeah, thank you so much everyone for listening and we would love to engage with you. So please feel free to reach out to our podcast Twitter handle which is @softwaresocpod, or either of us [@leenyburger, @mjwhansen]. We're both on Twitter and we'd love to hear what you think and any questions you have about the show.
