/
RSS Feed
In this episode of Startups For The Rest Of Us, Rob interviews Danielle Simpson and Arvid Kahl, co-founders of FeedbackPanda, a SaaS business they bootstrapped to $55k MRR with no outside funding and no employees. They sold directly to teachers, a price-sensitive market, and they used referral programs and word of mouth to create rapid growth. You will hear about the struggles, victories, highs, and lows of their startup journey. Arvid and Danielle give honest, powerful insight into what it was really like to manage their company just the two of them, and what ultimately led to their decision to sell their company for a life-changing amount of money.
* We are in a slightly different headspace in this episode, because we recorded this before the COVID-19 crisis. But we still wanted to share this episode, because we want you to benefit from this powerful conversation.
The finer points of the episode:
- 2:12 – What it was like to sell FeedbackPanda for a lifechanging sum of money
- 6:40 – Why they ultimately made the decision to sell their business
- 12:40 – How the perfect combination of luck and skill led to their business’ huge success
- 14:38 – What it was like selling to teachers, a price-sensitive market
- 18:44 – Using referral programs and word-of-mouth to generate extremely rapid growth
- 24:20 – Can their approach to growth be replicated in other industries?
- 29:11 – More about their decision not to hire anyone
- 33:49 – The biggest low point of their startup journey and how they overcame them
- 38:25 – When did they start thinking about selling their company?
- 40:34 – What is next for Arvid and Danielle?
Items mentioned in this episode:
Rob: A quick note before we dive into this episode of Startups for the Rest of Us. I recorded this interview a couple of months ago, and then due to the coronavirus and a couple of other factors, it’s just been delayed longer than I thought it would. We make no mention of Corona. We are in a little bit of a different headspace than we might be today, but the story the founders tell, the conversation is still extremely compelling. I just wanted you to have that in mind as you listen to this episode.
Welcome to this week’s episode of Startups for the Rest of Us. I’m your host, Rob Walling. Each week on the show you hear from startup founders who are ambitious, but some of them bootstrap, some self-fund, and others raise small amounts of money. The uniting factor is that they want to build great companies, they want to build great lives for themselves, and they don’t want to go down the traditional venture track of go big, go home, I have to build a unicorn or bust.
This week, we hear from Arvid Kahl and Danielle Simpson, the co-founders of FeedbackPanda. Arvid and Danielle did an attendee talk at MicroConf in Croatia just a few months back, and I wanted to invite them to the show because of how interesting their story is. Over 2 years, they bootstrap FeedbackPanda to $55,000 MRR with no employees. FeedbackPanda is student feedback for teachers. They’re selling into the teacher market. They had rapid growth via word of mouth and a clever referral program. They really caught a nice wave in language learning.
We dive into all kinds of things today. They eventually sold the company for a life-changing sum of money after two years. In this conversation, you’ll hear about Arvid and Danielle’s struggles, their victories, their failures, the anxiety, the highs, and the lows. We dig into a lot of pretty fascinating things. I hope you enjoy this conversation, and with that, let’s dive into the conversation with Arvid and Danielle of FeedbackPanda.
Arvid and Danielle, thank you so much for joining me on the show today.
Danielle: Thanks so much for having us.
Arvid: Yeah, hi.
Rob: Welcome to Startups for the Rest of Us. I am just so fascinated with your story—growing in two years as bootstrappers from $0 to $55,000 MRR is just a fascinating story. I want to start off almost with the part of a typical hero’s journey at looking at the end of that journey. I’d love to hear, what did it feel like when you saw the money from selling FeedbackPanda into your bank account, when you refreshed that and you all those zeroes?
Danielle: Oh my gosh. It was such a mix of relief of never having seen that amount of money before, just shocked to have that amount on one bank account, total just happiness. How about you?
Arvid: It was pretty strange, it’s quite a weird feeling. The weird part about it to me was how nothing really changed. You have all these stories you hear about lives being changed and it does change in a numerical way, but we were still in our apartment that we’ve been in before for the two years prior to building the business every single day. We were still sitting on the same chairs and the same couch. The only thing we did was really get out two glasses and some affordable champagne.
Danielle: In the morning.
Arvid: Right, because that’s when we saw it due to the time difference between the States and Germany where we live. Just cheers and then that was it, right? Then we went right to the transition because that is also one of the bigger things we had to do at that point. It was a wonderful feeling. Also, I guess, just a couple hours later and a couple of days later, the feeling of dread and the feeling of, “What now?” came to the forefront of my mind at least because you just wonder, “Okay, now this is done,” and you have to shift your thinking. There was a lot of joy and a lot of slight- to medium-confusion.
Danielle: It kind of oscillated between all of those emotions for the next few months, I would say.
Rob: You had such a massive transition, and I hear this often from entrepreneurs who sell their companies, and I went through the same thing a couple of times. It makes sense that your response was elation followed by confusion followed by some day drinking, which is something I highly recommend. Once you sell your company, please, day drink a day or two. Just take a day off.
You know what’s fascinating is, Arvid, you just said, “Our lives didn’t change. We still were in this apartment sitting in the same chairs drinking affordable champagne.” What I find when I talk to makers is we don’t do it for the money, we do it for the freedom. We do it so that we can make and work on interesting projects, and build things that we want to build rather than buy a Maserati or buy a big house. I am curious. Did you guys buy anything interesting or cool with the money that maybe you had never indulged in before?
Danielle: We definitely took a very nice vacation. We spent ten days in South Africa, and it was amazing to be in the middle of the African bush and be in the middle of nature, and have something that we didn’t have for the past two years. The privilege of being fully present in those moments where we weren’t looking at Intercom while we’re also trying to look at an elephant or something like that. That was quite indulgent for us.
Arvid: That was the true luxury was having time to ourselves both as people who have been working 24/7 for two years and as a couple. Just to be able to be present with each other. No Maserati could buy us that kind of luxury as much as a vacation—just being away from it all. Material things, I don’t think so, and I think you’re absolutely right. Makers do it for the making, not for the raking. You don’t try to rake it and you just try to continue making stuff. That is what you want to do.
We were at an event, at a Berlin startup community just a couple of weeks prior to the whole due diligence phase, the negotiations, and all that stuff. There were people talking there about this post-economic state of mind, and that’s what you want to reach. You want to reach financial stability that allows you to make decisions that do not hinge on you being financially stable from the decisions you make in the business. It’s like you can make risky business choices because you know that personally, your finances are in order.
That’s one of the things that actually led us to consider selling the business is that all of our assets were bound in it. We had a great business, but it was also a lot to lose at the later point, later stage, in the second year of the business. That is one of those feelings that we had in addition to the normal dread of running a gigantic business with just two people.
Rob: Absolutely. Most entrepreneurs are just way under-diversified. Most entrepreneurs who are having success like you had literally had millions of dollars tied up into a small business, and they might have $100,000 in the public stock market. It’s completely under-diversified. I don’t think a lot of people think about that.
In the last episode, Ruben Gamez was on, and he brought up a really good point that I feel like we should hammer home is you can’t run a business like we run and expect it to be here in 10 or 20 years. A lot of these businesses just don’t make it because of the massive shifts in Google, APIs, and protocols. How many SaaS companies do we know that really last that long? It’s a lot more difficult than it seems. I’m curious, can you talk about how much you sold the company for?
Arvid: We cannot talk about how much we sold the company for, but what we can say is the fact that we sold it for a life-changing amount of money. That’s why we got out that champagne because it was actually something to celebrate, but we can’t be specific about the numbers.
Rob: Of course. Most acquisitions, especially at these types of sizes, are under NDA. I’m going to do some loose math, don’t confirm or deny this. When I think of a company selling to a financial buyer, $55,000 MRR mostly profit because you had no employees at $660,000 ARR, 3x–4x multiple puts it somewhere in the $1.5–$2.5 million dollar range in my head. If it was a strategic buyer, it would be a little more. Just so listeners have a context, I have zero inside information, and again, I don’t want you to confirm or deny. That’s probably the kind of range we’re talking about if this was a market rate sell.
Arvid: Interesting.
Rob: You like that?
Arvid: Yeah. Great. We won’t confirm or deny anything, but definitely an interesting insight. Interesting also into the math because that’s what most people are really interested in, we were also interested in trying to figure out how much our business might be worth. There were numbers—multipliers—from something below one to something above ten. People were talking about SaaS, content, e-commerce and all these different kinds of things.
There’s a lot of information in the market, there’s a lot of misinformation in the market, and every transaction seems to be extremely unique when it comes to the actual numbers that the multipliers—whichever ones are chosen—actually work on. As you said, we had essentially just two of us as employees and that’s founder-employees. It’s a different math than if we had four employees but the founders didn’t work in the company. All these kinds of things just go into a very complicated piece of mathematics. Ranges are usually the best you can do.
Rob: Yeah, and that’s the thing is you’re in a unique situation because most companies doing $55,000 MRR have multiple employees, have a team of three, four, five people, which would make it much less profitable than what I’m expecting FeedbackPanda was. Sales multiples also depend a lot so much on growth rate. A company that’s declining versus flat versus growing—5% month-over-month, 20% month-over-month. That’s where this gets very complicated and there is no one formula. It’s just negotiation at that point.
I am curious, if you had held out longer, if you waited another 6, 12, 18 months, do you feel like you could have sold for more? Were you continuing to grow and do you feel like you could have made more money having held out?
Danielle: Absolutely. There were channels that we hadn’t yet activated. FeedbackPanda is in a quite price-sensitive niche, but we regularly had customers proclaiming on Facebook that they would be willing to pay more. We just didn’t flip that switch. If you look at our previous members as well, we had a steady growth rate for our entire lifetime. We weren’t really slowing down.
Arvid: We were piggybacking on the growth of the Chinese education market, too. That one is still growing. FeedbackPanda’s customers being recruited by Chinese online English teaching schools, we were selling to mostly North American teachers that were hired by Chinese companies. There’s a lot going on in China when it comes to online teaching, which is the presence of the internet in even more rural areas and all these kinds of things.
That growth didn’t stall. There was no plateau in sight. There was a lot of competition on-site, which again if you look at Chinese businesses, it is a pretty good sign of a healthy industry because as long as there’s competition, there’s no clear winner and there’s growth. Our growth trajectory was very much aligned with the Chinese education system growth trajectory and that one was still and is still growing.
Rob: That’s great when you can catch a wave like that. Was that a bit of luck or did you guys see that that wave was coming and hop on it?
Danielle: It was both. We got lucky that we were in a position to see the wave coming and that we were in that position ready to build something. I saw the problem and we just hopped right in.
Arvid: Yeah, I think so. The lucky part was that Danielle injured her leg at some point. As lucky as this can be because that made you have to work from home, so we needed to find something to actually do. As an opera singer, you can’t really work from home that much. It’s not easy to sing in an apartment, but it is easy to teach. That worked. These Chinese companies just came up at that point. They’ve been around for (I would guess) three, four, or five months at that point, not much longer than that.
We were just at the beginning of that wave and we recognized it because people were there to see it. That’s an important point is we just fell into the industry and then we saw it for what it was. As much as most teachers are entrepreneurial, they’re not software tech kind of entrepreneurial. They’re entrepreneurial in their own ways at building content businesses and brands like influencer brands, but not necessarily a SaaS that could scale almost infinitely in the branch or in the niche itself.
Rob: I have this mental model. I like the way you said, “Well, there was some luck, but also some skill,” I will say. I have this mental model; I should probably write a blog post about it. It’s that success comes down to these three factors: luck, skill, and hard work. In the case of the two of you, you’ve already talked about the hard work. You said you work 24/7 in essence for two years. The skill that the two of you have built up over the years applies to that to a certain degree, as well as being able to forecast and notice the Chinese market taking off, and then that little bit of luck.
I’ve seen startups have varying degrees of this. You’ll see someone get extremely lucky that they happen to come across an idea, and then they really don’t need that much skill or hard work to get it done, but it’s very, very rare, and you can’t control luck. I like looking at these really meticulous, repeatable startups where most of us are willing to do the hard work, and it gets building up skills over time, and then hoping for that bit of luck that pushes you over the edge but not counting on 80%–90% of it to be luck.
Arvid: Yeah, you have a lot of overlap with opportunity surface theory at this point, because skills and hard work build the opportunity surface, then you just need something to actually strike it, and that would be the luck component.
Rob: The niche of selling to teachers, I’ve heard some horror stories about it, to be honest about price sensitivity, and needing a lot of support, is that correct? Was it a tough space to be in, in terms of just needing the volume of customers? Your price points—at least today—are $15 a month, so by MicroConf’s B2B SaaS standards, that it’s a pretty low price. I would expect high churn and price sensitivity. Is that accurate?
Danielle: Price sensitivity, 100%, but high churn, not at all. The teachers in this niche were already pre-selected, so they were somewhat technical. They knew how to integrate a SaaS into their workflow much more than had we gone to a brick-and-mortar school, tried to then convince the school board or the principal, had to go and train teachers how to use the software. The teachers already teaching in this niche were very self-directed. They were having a lot of trouble solving this problem that we ended up solving for them.
The high churn was not even an issue there, but the price point, we actually started lower. We started at $5 a month understanding that this audience might be price-sensitive. Then when they were converting so quickly, and seeing the value so quickly, then we doubled the price to $10 a month, got rid of the $5 a month plan altogether, and then we ran it at that price point.
Arvid: It was $10 for a year, and then we decided to increase it again by 50% to what is now $15 monthly. We released a referral system at the same time so that people could get the cheaper price, which was the actual price—$10 a month at the point—if they were referred or were referring, so people could go back to that price, so it didn’t cost them that much.
There were a couple of mistakes that we made there when it came to grandfathering the plans. We didn’t really ever set a horizon on that so it was an infinite or indefinite grandfathering of $10 a month plan if you had it before—I don’t know—December 31, 2018, then we would keep that forever, which also kept retention high (I guess) because people didn’t want to lose their plan, that kind of level.
Danielle: That goes back to what Arvid was talking about at the beginning with this post-economic state of mind. We felt we had to grandfather all of those teachers in because we were more worried about losing them as customers than the potential gain of having to pay 50% more.
Arvid: Yeah. We didn’t really want to experiment much with pricing because we really felt that if that is too much and we lose our customers, the whole business is going to implode. That’s the kind of fear that I had at that point, which is why referral system with this really, really high 30% coupon or something for life. Lots of the decisions we made came not from a place of optimism but of severe pessimism of potential future.
Rob: Do you think those decisions were correct? Would you make those again?
Arvid: No. They were correct for us and the business at the time, but if we were doing it again from a different financial security perspective or knowing that it wasn’t as crazy, sure.
Danielle: Can I just interject here, too? Had we been isolated in our different roles, if we weren’t the CTO, CEO, and the people talking directly to our customers on the support desk—hearing the sometimes quite sad stories about people who couldn’t make a $10 a month payment—we really empathized with our customers. I know Arvid has a huge heart and would often gift subscriptions to people. That was more of just feeling that was the right thing to do. For us, it wasn’t a business decision, it’s more of philanthropy.
Arvid: Yeah. There was a lot of emotion.
Rob: I can see that. I’m curious, I want to find out, why did this grow so quickly? Was it just the referral program? You’ve talked about word of mouth being the main driver.
Arvid: The referral system only came in a year after we actually already had all this growth.
Danielle: The Chinese schools that were hiring all of these teachers were doubling the teachers that they had contracted. It was like a gig. It was like Uber for online teaching. They hire these American teachers, and they had 10,000 teachers when I started in 2017. They doubled by December of 2018, so they had 20,000 teachers. When we sold in July of 2019, they were at 75,000 teachers. This was one company alone. We were really benefiting from this hiring rush of not only this one company that’s just one company that was the largest, but several other Chinese English companies coming up.
Arvid: We sold to the teachers directly. We were benefiting just from the sheer number of them being recruited because if we had tried to sell to the schools, it would have been a whole different thing. It would have been much more complicated, and there’s bureaucracy, and then there’s the whole thing with getting into the Chinese market, which is it is technically impossible without having somebody in there to do the work with you, so we just decided against that.
Partnering is also hard, so by selling directly in like a B2BC selling to the individual agents with a budget, which the teachers were, we could leverage the growth of that market without anything in the middle. The one great benefit of teachers is that teachers love sharing, and teachers organize in groups. Essentially, they organize as tribes even.
The teacher community around these English as a second language online schools, organize at Facebook groups. They are in Instagram communities. There is a lot of sharing, a lot of communicating going on. If you can get into these communities as a part of them—as Danielle did because she was a teacher at that point—then you see what people are talking about, what they were sharing with each other, and you can become part of the conversation, become part of the actual people trying to solve their problems.
That’s where we did our problem validation, that’s where we found all these problems that people had. We saw how many people of these teachers had the same problems, were they critical, were they just nuances or annoyances. That’s also where we started our marketing, and where we did all of our marketing, which wasn’t really too much. It boiled down to a number of comments on some Facebook posts, that’s what we did because the tribe was so strong that once we actually put our solution in there, and people recognized it as a product that they could really use, they started sharing it for us.
The referral system we put in place because we were just raising our prices and we want to make it still affordable to the people that were currently purchasing the subscription to our software. The actual growth comes from the tribal structure of the teacher community and from us finding the water cooler—finding the place where they were hanging out.
Danielle: I would say it was lucky that they even knew that they had a problem. Teachers are so used to doing the extra work for zero pay. We really did get lucky that they recognized that this shouldn’t be the case and that there could be options for them.
Arvid: When you think about it, there’s this whole thing about prospect awareness funnel by Eugene Schwartz when it comes to like I’m completely unaware, and problem-aware, solution-aware, product-aware until they are fully aware of your product. We were lucky that people were already at the problem-awareness and often solution-awareness stage because, while we didn’t have competition in the actual industry, there were competitive alternatives.
People were using Excel sheets and Word documents, or they were already starting to share their templates that they used for their feedback on Google Docs or in certain groups on Facebook. We already saw people solving the problem in a rickety way, everyone for themselves, and we just really consolidated this into a centralized system that also had the strong network effect, which mirrored the internal structure of the tribe that we had found on Facebook.
In FeedbackPanda, sharing was built into the process. People would write templates for their feedback, and allow them to generate it easily, and then they could share it to what we call the cloud. In the cloud, people could find templates by other teachers, use them, modify them, and share them back.
There was this whole community building that happened around the actual data in our product, which had this gigantic network effect pulling in even more people because once they signed up, there were literally hundreds of thousands of templates for them to pick to make their job much faster.
Rob: It’s fascinating to hear the whole tribal structure that you’re talking about because I have not heard of a bootstrap SaaS app that has grown as fast as you did by making some Facebook posts. Because you said, it’s a tribe, water cooler. That was the key, so this feels to me—at least looking across the apps that I know about that I have had fast growth—like a one-off. The uniqueness of the teachers and how it was like a confluence of events, (a) it’s a tribal structure, (b) you found the water cooler, and (c) it was growing so fast that people were just diving in. Do you agree with that or do you feel like your approach here could be reproduced in other spaces?
Danielle: I’m not sure if this makes it fall into the category of one-off or repeatable, but I think what we did once people actually got to FeedbackPanda, we talked directly to them, and we were selling them on our product. Arvid said there wasn’t much competition. There was another SaaS that tried to offer something similar to what we were offering and to my knowledge, they don’t have even a fraction of the teachers that signed up for FeedbackPanda.
I think that doing those things that don’t quite scale or you don’t see them scaling, talking to a teacher daily trying to get them to that moment of really understanding how the product works, and then turning those first conversations into knowledge base articles where teachers can find the information for themselves, or communicating in a way that is like a soundbite that they can take from that conversation, go back to the community, share the information, that communication style was actually how we activated the tribe for us. Those direct conversations using user-engage at the very beginning and then migrated to Intercom.
Arvid: We always try to be in direct contact with people. The tribe is one thing, but we did amplify it. We needed to amplify it, we needed to project a brand into the tribe that was both part of the tribe and solving something for them on a professional level. We were the teachers that helped other teachers. I think that’s the transferable component as well. It works in software development. If you sell a product to engineers as an engineer, and you’re part of their community, you have a reputation as being a good engineer—if you have that—that really helps because you sell it through your reputation, you sell it through some sense of authority, and having a following already.
You were talking about this in the episode with Ruben as well. If you don’t have an audience, if you don’t have a market, there is an audience somewhere, you just haven’t found it yet. I believe that there are tribal or at least community structures for every single thing, for every kind of thing we’re part of. I’ve been talking with a lot of consulting clients recently, and they are in specialty foods, they are in document processing for specific kinds of jobs, they are in many kinds of groups of people.
Whenever I talk to them, after a while, we figure out there is a niche community. There is a group congregating somewhere. There is a meet-up for this in this city. There are three people that go there at all times. There is this online bulletin board where people that look like it’s from the 80s, but that’s usually a good sign that shows that there’s a community that has been around for a long, long time hidden away in this one niche location. You just really have to find it, and become part of it, and not do the marketing way where you start trying to push your product from the beginning. You just become part of it because you want to be part of it.
You want to talk to these people, you want to help them out, and then eventually, you can put your product in there because it’s a natural progression from being a part of the community to wanting to help the community.
Danielle: There was a lot of integrity that we built through our brand. Something I didn’t expect to do was release a product that wasn’t quite perfect or finished. Anybody who knows me knows that I like to get things as close to perfect as possible before revealing it to anybody, but this would have been detrimental to FeedbackPanda. Arvid quite nicely balances me out in this regard.
We released the product, then we got a lot of great feedback from the teachers in the community. They thought that he was a wizard because they would give him their feedback, and then in a day or two, he would release the feature that they had wanted. I’m not saying that you should build every feature that somebody asks you for. We really looked through all of those examples, what really deserved to be a part of the product, but even that was just reinforcing this dependability on FeedbackPanda.
Rob: It’s great to be makers who can ship fast in a space where all the software sucks because you don’t have to be that good to be really, really impressive. That’s cool. I love anecdotes like that. We used to ship stuff in the early days of Drip the same day. Someone would send a supportive comment, “How come it doesn’t do this?” I would literally build it two hours later, send them an email. Those were some of the funnest ones.
I’m curious. You’re growing this company, it’s obviously moving very quickly, why did you not hire someone to help? Was there just not enough work that you needed help, or was there some other reason for that decision?
Arvid: There’s a short and a long answer. The short one is I’m an idiot, and Danielle could tell you why.
Danielle: I would say that be careful when you get advice from books and podcasts without really critically examining whether they apply to your situation. We were both guilty of taking this advice, “Hire late, hire late, hire late,” to heart. What we made now in retrospect can say, we should have built a roadmap that said, “Okay, when we get to this goal, when we reach this amount of support tickets, or whether it’s tied to MRR, or to the amount of work that you have, we could have said this is when we’re going to hire.” We didn’t do that because we didn’t expect to grow that quickly. We thought we had more time.
Arvid: We didn’t really set goals much. The one and only goal that I set for myself was, “I want to grow this business to the unimaginable number of $50,000 MRR,” and that’s going to be impossible so might just as well set it as a goal. Once we hit that, “Yeah, what now?” Even before, once we hit, $10,000, “Oh, this works,” or $20,000, “Okay, now we can pay ourselves.” Then $30,000, $40,000, “Oh, this is still growing.”
There was never a moment of scheduled reflection. I’ve been writing about this recently, I just remembered. The thing I called is continuous validation, but not just of a product but actually of your business. Are you still in it for the same reasons? Are you still fine with what you’re doing? These kinds of things, had we done this—and we did it sometimes, but by far not enough—we would have understood—at least a year in—that it would be fine to hire a customer service person.
You just alluded to it (I guess) that there wasn’t enough to do for a full 40-hour a week position. To me, in my complete absence of knowledge and experience, if you hire somebody, it has to be a full-time position. That was what I thought. Of course, you don’t. Of course, you can have somebody do this for the two hours a day that most people are asking questions, and then do the rest later, or early, or whatever. To me, in the middle of it and always being interrupted by conversations on Intercom at random hours, it felt like, “Might just as well do it myself until we have somebody to take it all away from me completely,” which meant that I blocked all of this.
I thought, “I could do it. I can take care of it.” Yeah, it caused me a lot of sanity, caused a lot of anxiety that I had to learn how to deal with. In retrospect, once we sold, we had to hire our replacement so there was no way around that. When we did that I figured out, “Oh, this is actually easy and enjoyable and I should have done this immediately.” That’s one of the big learnings for me.
Rob: It’s a good point, and you’ve touched on two problems really that a lot of founders face, especially in the bootstrap space. One, we have the savior complex where we feel like we need to do everything, and it is hard to delegate, or you have a mental roadblock that won’t let you hire someone part-time, you just don’t think about it, or you just push yourself to the brink of burnout and exhaustion.
The other thing that it brings up for me is the lack of goals, lack of planning thing, it’s a problem. There are reasons that larger startups and larger companies have all these planning meetings, forecasts, and have goals. I know that so many of us—myself included—we leave the corporate world because there’s so much […] that you have to deal with. Then we don’t want to go start our own startup and put a bunch of […] into it, but there’s a balance.
If you throw it all out, you can find yourself in that situation where you’re not taking care of yourself. Frankly, you’re not being a good steward of your own mental health, nor of your company. It can actually hurt the growth of your company. There’s a lot of negatives to come back. I’m glad we touched on that because folks should be aware that it’s not all the way on the left, all the way on the right. There’s a middle here where I do think even in smaller startups that we need to be mindful and do some thinking and planning about this stuff.
Piggybacking on that, I’m curious, during these two years of fast growth, you’ve already referenced a lot of ups and downs. The growth is amazing, it’s fun, it’s exhilarating, the pain and anxiety you’ve brought up can be debilitating, and it can take a toll on you. Can you take me to a moment or a series of moments that you remember that perhaps were a low point during the journey?
Arvid: I have a couple of good ones and most of them (funny enough) are technical. That’s what I was doing mostly. I was responsible for building the product, for maintaining the infrastructure. FeedbackPanda was built on essentially a cloud-native thing. We put our software into Docker containers, and they were running on Google cloud in the end, but we didn’t start there.
We started with a small, affordable, call it “local” cloud provider—some company in Germany. They were great until they weren’t. We ran into connectivity trouble, and they had issues providing the service. Our service would crash. That was already at a point where we had a couple of a hundred customers. Here I was having to deal with the maintenance that I had to do to actually get the service back up at random times because this other startup that was just the same size as we were had issues configuring the cluster, the infrastructure that they were providing for us to run our software on.
That stuff started to accelerate. First, it was a couple of minutes of downtime a day. Then it was half an hour, and then it turned into full downtime for a day. I was just sitting there. I think we were in Canada even at that point visiting Danielle’s family. I was just on my phone, trying to get their weird little JavaScript UI (that didn’t work) to restart the containers because they didn’t have a CLI yet or an API even where I could communicate with their system and then restart stuff automatically.
I was sitting in somebody’s home—that wasn’t my own—trying to fix the service that was running on infrastructure—that was also not my own—and that I had no control over because they also were like four dudes from Germany trying to build the next AWS. Big mistake, right? It was cheap, and I didn’t want to look into the Google Cloud documentation, so we chose that. Big, big mistake.
It took the knowledge of a very capable friend, and a couple of hours late at night to actually migrate this whole thing over—just a couple hours by the way. I just pushed it as far away from me as possible. I tried to believe that they could deal with the problem, get back actually solving it, and allow us to have our service running, but at some point just couldn’t, so we’ve migrated over to Google Cloud. It had zero seconds downtime for the next 400 days.
That was a big problem. I was just trying to stay focused on it because I couldn’t sleep. Our monitoring system was always essentially waking me up with a call. I had to go restart some containers and try to manage it. It was brutal. I did not sleep well, and I still have some sort of monitoring PTSD whenever a call comes in. Just one split-second, “Is it the server again?” even though I don’t even own the business anymore. I still have that to this day.
Rob: People ask founders why they sell their companies. It’s like these times. I’ve been there, so I know what that feels like. Danielle, do you have thoughts on this?
Danielle: Yeah, I was right there beside him, obviously, even more powerless when a service that you’re using just goes down.
Rob: Was that your low point as well?
Danielle: 100%. It happened more than once with this particular server before we moved over to a better solution. There’s one other point, though, just to add a different perspective. We’re co-founders and partners, so when something personal in life happens, it was really happening to both of us. For example, last winter, my grandmother was very sick, and she ended up passing away. Arvid really picked up the slack for me, but being a co-founder with your partner does mean that you’re experiencing those personal events together whether they’re good, celebrations, or are tragic. Just something to consider. It’s not a low-low.
Arvid: It is a low, just not a business low.
Rob: It definitely makes it a lot more complicated to have, that dual relationship of co-founder and partner. I get it. We’re running short on time but I have two more questions for you that I’d love your thoughts on. The first is when you started building the FeedbackPanda, did you intend to sell it from the start? Was that thought in your mind? If not, when did that thought first occur to you?
Arvid: Before we actually started building the business and before Danielle was even teaching English online, I was working as a software engineer in Hamburg in Germany. We live in Berlin, so that’s a two-hour commute there, and then two hours back. Three days of the week, I would commute for four hours, and I pretty much had nothing else to do (because it was on a train so I wouldn’t have to drive) reading or listening to podcasts.
I was reading a lot of books. A couple of the books that I read at that point were the E-myth by Michael E. Gerber and Built to Sell by John Warrillow. Built to Sell has really stuck with me from the beginning because I learned in that book (in particular) that building a sellable company is building a company you can run forever, and building a company that runs without you where you as the owner actually benefit from the company running, and other people, systems, or automation are doing the work.
We never really intended to sell the business, but we built it as a sellable business anyway because that was (to us) the optimal way to build a business. That when we actually got people that were interested in purchasing the company, then we thought about selling it. From the beginning, we just really wanted to build a cool, efficient, and mostly a highly-automated business because at least I am very lazy and I didn’t want to work too much. Turns out I had to work quite a bit, but we still made it highly automated, and hence sellable, which really helped when it came to actually sell the business.
Rob: That’s a good point that you make. I think founders get involved in this binary thinking about, “I am never going to sell my business,” or “I’m building it to sell.” Building it to sell from the start helps you no matter what. The idea of you said you’re lazy, that’s an adaptive trait for a founder because you’re lazy, but you’re willing to put in hard work is a thing. Lazy, I think about just means you’re highly efficient, and having a goal of having an automated business bringing in tens of thousands a month is something all of us would aspire to.
Wrapping up, I’m curious, are you going to do it again? Do the two of you have the desire?
Danielle: Coming 2020, no. Just kidding. I would love to hop back in. In fact, I co-founded a company while we still own FeedbackPanda with the hopes of building out this project. Absolutely, I would love to get back in. It’s quite interesting. I feel that our roles—since we’ve sold—have reversed. I was very focused on our brand image and nurturing the audience, and now, I’ve really been looking into the no-code movement and trying to build up this product, another SaaS essentially.
Arvid: Whereas I have started pretty much building an audience and writing. We really switched. When we sold the business, we both fell into this place where we didn’t know what to do. We’ve been making our way out of it, and I did this through writing. Started a blog, started a podcast, started a newsletter, and just committed to actually putting the information that I had in my mind, that I had learned over the last couple years both with FeedbackPanda and all the prior things that I’ve been part of, put that in writing, and then I started being on Twitter, talking to people, and building an audience.
I wrote a guide and wrote a book, eventually. All these things, that is where I’m going, but of course, I’m still a software engineer. I still want to build things. We’ll see where we really end up, but there’s a lot of projects that are already on the horizon. There’s always people coming knocking on our door with new ideas, suggestions, of course. I don’t think we’re done. We are now at a point where we can actually make much riskier and more interesting decisions in the projects that were involved in. Why not, right?
Rob: Yup. It’s a great place to be in. It’s something a lot of us—many, many of us—aspire to. Thank you guys so much for coming on the show. If folks want to keep up with you on Twitter, you’re @arvidkahl and @SimpsonDanniK. We will link both of those up in the show notes. Arvid, you’re doing a lot of writing at the bootstrappedfounder.com. If folks wanted to check out Danielle’s website, it’s simpsondanielle.com. Thanks again for joining me today.
Danielle: Thank you so much.
Arvid: Thanks a lot.
Rob: If you have a question for me or one of our guests, please, leave us a voicemail at 1-888-801-9690 or email questions@startupsfortherestofus.com. If you’re not subscribed, you should search for startups in any podcatcher, and we’re going to be in the top few. We do have full transcripts of each episode—even though they are a few weeks behind—at any given time. Thanks so much for hanging out again with me this week and I’ll see you next time.
Welcome to this week’s episode of Startups for the Rest of Us. I’m your host, Rob Walling. Each week on the show you hear from startup founders who are ambitious, but some of them bootstrap, some self-fund, and others raise small amounts of money. The uniting factor is that they want to build great companies, they want to build great lives for themselves, and they don’t want to go down the traditional venture track of go big, go home, I have to build a unicorn or bust.
This week, we hear from Arvid Kahl and Danielle Simpson, the co-founders of FeedbackPanda. Arvid and Danielle did an attendee talk at MicroConf in Croatia just a few months back, and I wanted to invite them to the show because of how interesting their story is. Over 2 years, they bootstrap FeedbackPanda to $55,000 MRR with no employees. FeedbackPanda is student feedback for teachers. They’re selling into the teacher market. They had rapid growth via word of mouth and a clever referral program. They really caught a nice wave in language learning.
We dive into all kinds of things today. They eventually sold the company for a life-changing sum of money after two years. In this conversation, you’ll hear about Arvid and Danielle’s struggles, their victories, their failures, the anxiety, the highs, and the lows. We dig into a lot of pretty fascinating things. I hope you enjoy this conversation, and with that, let’s dive into the conversation with Arvid and Danielle of FeedbackPanda.
Arvid and Danielle, thank you so much for joining me on the show today.
Danielle: Thanks so much for having us.
Arvid: Yeah, hi.
Rob: Welcome to Startups for the Rest of Us. I am just so fascinated with your story—growing in two years as bootstrappers from $0 to $55,000 MRR is just a fascinating story. I want to start off almost with the part of a typical hero’s journey at looking at the end of that journey. I’d love to hear, what did it feel like when you saw the money from selling FeedbackPanda into your bank account, when you refreshed that and you all those zeroes?
Danielle: Oh my gosh. It was such a mix of relief of never having seen that amount of money before, just shocked to have that amount on one bank account, total just happiness. How about you?
Arvid: It was pretty strange, it’s quite a weird feeling. The weird part about it to me was how nothing really changed. You have all these stories you hear about lives being changed and it does change in a numerical way, but we were still in our apartment that we’ve been in before for the two years prior to building the business every single day. We were still sitting on the same chairs and the same couch. The only thing we did was really get out two glasses and some affordable champagne.
Danielle: In the morning.
Arvid: Right, because that’s when we saw it due to the time difference between the States and Germany where we live. Just cheers and then that was it, right? Then we went right to the transition because that is also one of the bigger things we had to do at that point. It was a wonderful feeling. Also, I guess, just a couple hours later and a couple of days later, the feeling of dread and the feeling of, “What now?” came to the forefront of my mind at least because you just wonder, “Okay, now this is done,” and you have to shift your thinking. There was a lot of joy and a lot of slight- to medium-confusion.
Danielle: It kind of oscillated between all of those emotions for the next few months, I would say.
Rob: You had such a massive transition, and I hear this often from entrepreneurs who sell their companies, and I went through the same thing a couple of times. It makes sense that your response was elation followed by confusion followed by some day drinking, which is something I highly recommend. Once you sell your company, please, day drink a day or two. Just take a day off.
You know what’s fascinating is, Arvid, you just said, “Our lives didn’t change. We still were in this apartment sitting in the same chairs drinking affordable champagne.” What I find when I talk to makers is we don’t do it for the money, we do it for the freedom. We do it so that we can make and work on interesting projects, and build things that we want to build rather than buy a Maserati or buy a big house. I am curious. Did you guys buy anything interesting or cool with the money that maybe you had never indulged in before?
Danielle: We definitely took a very nice vacation. We spent ten days in South Africa, and it was amazing to be in the middle of the African bush and be in the middle of nature, and have something that we didn’t have for the past two years. The privilege of being fully present in those moments where we weren’t looking at Intercom while we’re also trying to look at an elephant or something like that. That was quite indulgent for us.
Arvid: That was the true luxury was having time to ourselves both as people who have been working 24/7 for two years and as a couple. Just to be able to be present with each other. No Maserati could buy us that kind of luxury as much as a vacation—just being away from it all. Material things, I don’t think so, and I think you’re absolutely right. Makers do it for the making, not for the raking. You don’t try to rake it and you just try to continue making stuff. That is what you want to do.
We were at an event, at a Berlin startup community just a couple of weeks prior to the whole due diligence phase, the negotiations, and all that stuff. There were people talking there about this post-economic state of mind, and that’s what you want to reach. You want to reach financial stability that allows you to make decisions that do not hinge on you being financially stable from the decisions you make in the business. It’s like you can make risky business choices because you know that personally, your finances are in order.
That’s one of the things that actually led us to consider selling the business is that all of our assets were bound in it. We had a great business, but it was also a lot to lose at the later point, later stage, in the second year of the business. That is one of those feelings that we had in addition to the normal dread of running a gigantic business with just two people.
Rob: Absolutely. Most entrepreneurs are just way under-diversified. Most entrepreneurs who are having success like you had literally had millions of dollars tied up into a small business, and they might have $100,000 in the public stock market. It’s completely under-diversified. I don’t think a lot of people think about that.
In the last episode, Ruben Gamez was on, and he brought up a really good point that I feel like we should hammer home is you can’t run a business like we run and expect it to be here in 10 or 20 years. A lot of these businesses just don’t make it because of the massive shifts in Google, APIs, and protocols. How many SaaS companies do we know that really last that long? It’s a lot more difficult than it seems. I’m curious, can you talk about how much you sold the company for?
Arvid: We cannot talk about how much we sold the company for, but what we can say is the fact that we sold it for a life-changing amount of money. That’s why we got out that champagne because it was actually something to celebrate, but we can’t be specific about the numbers.
Rob: Of course. Most acquisitions, especially at these types of sizes, are under NDA. I’m going to do some loose math, don’t confirm or deny this. When I think of a company selling to a financial buyer, $55,000 MRR mostly profit because you had no employees at $660,000 ARR, 3x–4x multiple puts it somewhere in the $1.5–$2.5 million dollar range in my head. If it was a strategic buyer, it would be a little more. Just so listeners have a context, I have zero inside information, and again, I don’t want you to confirm or deny. That’s probably the kind of range we’re talking about if this was a market rate sell.
Arvid: Interesting.
Rob: You like that?
Arvid: Yeah. Great. We won’t confirm or deny anything, but definitely an interesting insight. Interesting also into the math because that’s what most people are really interested in, we were also interested in trying to figure out how much our business might be worth. There were numbers—multipliers—from something below one to something above ten. People were talking about SaaS, content, e-commerce and all these different kinds of things.
There’s a lot of information in the market, there’s a lot of misinformation in the market, and every transaction seems to be extremely unique when it comes to the actual numbers that the multipliers—whichever ones are chosen—actually work on. As you said, we had essentially just two of us as employees and that’s founder-employees. It’s a different math than if we had four employees but the founders didn’t work in the company. All these kinds of things just go into a very complicated piece of mathematics. Ranges are usually the best you can do.
Rob: Yeah, and that’s the thing is you’re in a unique situation because most companies doing $55,000 MRR have multiple employees, have a team of three, four, five people, which would make it much less profitable than what I’m expecting FeedbackPanda was. Sales multiples also depend a lot so much on growth rate. A company that’s declining versus flat versus growing—5% month-over-month, 20% month-over-month. That’s where this gets very complicated and there is no one formula. It’s just negotiation at that point.
I am curious, if you had held out longer, if you waited another 6, 12, 18 months, do you feel like you could have sold for more? Were you continuing to grow and do you feel like you could have made more money having held out?
Danielle: Absolutely. There were channels that we hadn’t yet activated. FeedbackPanda is in a quite price-sensitive niche, but we regularly had customers proclaiming on Facebook that they would be willing to pay more. We just didn’t flip that switch. If you look at our previous members as well, we had a steady growth rate for our entire lifetime. We weren’t really slowing down.
Arvid: We were piggybacking on the growth of the Chinese education market, too. That one is still growing. FeedbackPanda’s customers being recruited by Chinese online English teaching schools, we were selling to mostly North American teachers that were hired by Chinese companies. There’s a lot going on in China when it comes to online teaching, which is the presence of the internet in even more rural areas and all these kinds of things.
That growth didn’t stall. There was no plateau in sight. There was a lot of competition on-site, which again if you look at Chinese businesses, it is a pretty good sign of a healthy industry because as long as there’s competition, there’s no clear winner and there’s growth. Our growth trajectory was very much aligned with the Chinese education system growth trajectory and that one was still and is still growing.
Rob: That’s great when you can catch a wave like that. Was that a bit of luck or did you guys see that that wave was coming and hop on it?
Danielle: It was both. We got lucky that we were in a position to see the wave coming and that we were in that position ready to build something. I saw the problem and we just hopped right in.
Arvid: Yeah, I think so. The lucky part was that Danielle injured her leg at some point. As lucky as this can be because that made you have to work from home, so we needed to find something to actually do. As an opera singer, you can’t really work from home that much. It’s not easy to sing in an apartment, but it is easy to teach. That worked. These Chinese companies just came up at that point. They’ve been around for (I would guess) three, four, or five months at that point, not much longer than that.
We were just at the beginning of that wave and we recognized it because people were there to see it. That’s an important point is we just fell into the industry and then we saw it for what it was. As much as most teachers are entrepreneurial, they’re not software tech kind of entrepreneurial. They’re entrepreneurial in their own ways at building content businesses and brands like influencer brands, but not necessarily a SaaS that could scale almost infinitely in the branch or in the niche itself.
Rob: I have this mental model. I like the way you said, “Well, there was some luck, but also some skill,” I will say. I have this mental model; I should probably write a blog post about it. It’s that success comes down to these three factors: luck, skill, and hard work. In the case of the two of you, you’ve already talked about the hard work. You said you work 24/7 in essence for two years. The skill that the two of you have built up over the years applies to that to a certain degree, as well as being able to forecast and notice the Chinese market taking off, and then that little bit of luck.
I’ve seen startups have varying degrees of this. You’ll see someone get extremely lucky that they happen to come across an idea, and then they really don’t need that much skill or hard work to get it done, but it’s very, very rare, and you can’t control luck. I like looking at these really meticulous, repeatable startups where most of us are willing to do the hard work, and it gets building up skills over time, and then hoping for that bit of luck that pushes you over the edge but not counting on 80%–90% of it to be luck.
Arvid: Yeah, you have a lot of overlap with opportunity surface theory at this point, because skills and hard work build the opportunity surface, then you just need something to actually strike it, and that would be the luck component.
Rob: The niche of selling to teachers, I’ve heard some horror stories about it, to be honest about price sensitivity, and needing a lot of support, is that correct? Was it a tough space to be in, in terms of just needing the volume of customers? Your price points—at least today—are $15 a month, so by MicroConf’s B2B SaaS standards, that it’s a pretty low price. I would expect high churn and price sensitivity. Is that accurate?
Danielle: Price sensitivity, 100%, but high churn, not at all. The teachers in this niche were already pre-selected, so they were somewhat technical. They knew how to integrate a SaaS into their workflow much more than had we gone to a brick-and-mortar school, tried to then convince the school board or the principal, had to go and train teachers how to use the software. The teachers already teaching in this niche were very self-directed. They were having a lot of trouble solving this problem that we ended up solving for them.
The high churn was not even an issue there, but the price point, we actually started lower. We started at $5 a month understanding that this audience might be price-sensitive. Then when they were converting so quickly, and seeing the value so quickly, then we doubled the price to $10 a month, got rid of the $5 a month plan altogether, and then we ran it at that price point.
Arvid: It was $10 for a year, and then we decided to increase it again by 50% to what is now $15 monthly. We released a referral system at the same time so that people could get the cheaper price, which was the actual price—$10 a month at the point—if they were referred or were referring, so people could go back to that price, so it didn’t cost them that much.
There were a couple of mistakes that we made there when it came to grandfathering the plans. We didn’t really ever set a horizon on that so it was an infinite or indefinite grandfathering of $10 a month plan if you had it before—I don’t know—December 31, 2018, then we would keep that forever, which also kept retention high (I guess) because people didn’t want to lose their plan, that kind of level.
Danielle: That goes back to what Arvid was talking about at the beginning with this post-economic state of mind. We felt we had to grandfather all of those teachers in because we were more worried about losing them as customers than the potential gain of having to pay 50% more.
Arvid: Yeah. We didn’t really want to experiment much with pricing because we really felt that if that is too much and we lose our customers, the whole business is going to implode. That’s the kind of fear that I had at that point, which is why referral system with this really, really high 30% coupon or something for life. Lots of the decisions we made came not from a place of optimism but of severe pessimism of potential future.
Rob: Do you think those decisions were correct? Would you make those again?
Arvid: No. They were correct for us and the business at the time, but if we were doing it again from a different financial security perspective or knowing that it wasn’t as crazy, sure.
Danielle: Can I just interject here, too? Had we been isolated in our different roles, if we weren’t the CTO, CEO, and the people talking directly to our customers on the support desk—hearing the sometimes quite sad stories about people who couldn’t make a $10 a month payment—we really empathized with our customers. I know Arvid has a huge heart and would often gift subscriptions to people. That was more of just feeling that was the right thing to do. For us, it wasn’t a business decision, it’s more of philanthropy.
Arvid: Yeah. There was a lot of emotion.
Rob: I can see that. I’m curious, I want to find out, why did this grow so quickly? Was it just the referral program? You’ve talked about word of mouth being the main driver.
Arvid: The referral system only came in a year after we actually already had all this growth.
Danielle: The Chinese schools that were hiring all of these teachers were doubling the teachers that they had contracted. It was like a gig. It was like Uber for online teaching. They hire these American teachers, and they had 10,000 teachers when I started in 2017. They doubled by December of 2018, so they had 20,000 teachers. When we sold in July of 2019, they were at 75,000 teachers. This was one company alone. We were really benefiting from this hiring rush of not only this one company that’s just one company that was the largest, but several other Chinese English companies coming up.
Arvid: We sold to the teachers directly. We were benefiting just from the sheer number of them being recruited because if we had tried to sell to the schools, it would have been a whole different thing. It would have been much more complicated, and there’s bureaucracy, and then there’s the whole thing with getting into the Chinese market, which is it is technically impossible without having somebody in there to do the work with you, so we just decided against that.
Partnering is also hard, so by selling directly in like a B2BC selling to the individual agents with a budget, which the teachers were, we could leverage the growth of that market without anything in the middle. The one great benefit of teachers is that teachers love sharing, and teachers organize in groups. Essentially, they organize as tribes even.
The teacher community around these English as a second language online schools, organize at Facebook groups. They are in Instagram communities. There is a lot of sharing, a lot of communicating going on. If you can get into these communities as a part of them—as Danielle did because she was a teacher at that point—then you see what people are talking about, what they were sharing with each other, and you can become part of the conversation, become part of the actual people trying to solve their problems.
That’s where we did our problem validation, that’s where we found all these problems that people had. We saw how many people of these teachers had the same problems, were they critical, were they just nuances or annoyances. That’s also where we started our marketing, and where we did all of our marketing, which wasn’t really too much. It boiled down to a number of comments on some Facebook posts, that’s what we did because the tribe was so strong that once we actually put our solution in there, and people recognized it as a product that they could really use, they started sharing it for us.
The referral system we put in place because we were just raising our prices and we want to make it still affordable to the people that were currently purchasing the subscription to our software. The actual growth comes from the tribal structure of the teacher community and from us finding the water cooler—finding the place where they were hanging out.
Danielle: I would say it was lucky that they even knew that they had a problem. Teachers are so used to doing the extra work for zero pay. We really did get lucky that they recognized that this shouldn’t be the case and that there could be options for them.
Arvid: When you think about it, there’s this whole thing about prospect awareness funnel by Eugene Schwartz when it comes to like I’m completely unaware, and problem-aware, solution-aware, product-aware until they are fully aware of your product. We were lucky that people were already at the problem-awareness and often solution-awareness stage because, while we didn’t have competition in the actual industry, there were competitive alternatives.
People were using Excel sheets and Word documents, or they were already starting to share their templates that they used for their feedback on Google Docs or in certain groups on Facebook. We already saw people solving the problem in a rickety way, everyone for themselves, and we just really consolidated this into a centralized system that also had the strong network effect, which mirrored the internal structure of the tribe that we had found on Facebook.
In FeedbackPanda, sharing was built into the process. People would write templates for their feedback, and allow them to generate it easily, and then they could share it to what we call the cloud. In the cloud, people could find templates by other teachers, use them, modify them, and share them back.
There was this whole community building that happened around the actual data in our product, which had this gigantic network effect pulling in even more people because once they signed up, there were literally hundreds of thousands of templates for them to pick to make their job much faster.
Rob: It’s fascinating to hear the whole tribal structure that you’re talking about because I have not heard of a bootstrap SaaS app that has grown as fast as you did by making some Facebook posts. Because you said, it’s a tribe, water cooler. That was the key, so this feels to me—at least looking across the apps that I know about that I have had fast growth—like a one-off. The uniqueness of the teachers and how it was like a confluence of events, (a) it’s a tribal structure, (b) you found the water cooler, and (c) it was growing so fast that people were just diving in. Do you agree with that or do you feel like your approach here could be reproduced in other spaces?
Danielle: I’m not sure if this makes it fall into the category of one-off or repeatable, but I think what we did once people actually got to FeedbackPanda, we talked directly to them, and we were selling them on our product. Arvid said there wasn’t much competition. There was another SaaS that tried to offer something similar to what we were offering and to my knowledge, they don’t have even a fraction of the teachers that signed up for FeedbackPanda.
I think that doing those things that don’t quite scale or you don’t see them scaling, talking to a teacher daily trying to get them to that moment of really understanding how the product works, and then turning those first conversations into knowledge base articles where teachers can find the information for themselves, or communicating in a way that is like a soundbite that they can take from that conversation, go back to the community, share the information, that communication style was actually how we activated the tribe for us. Those direct conversations using user-engage at the very beginning and then migrated to Intercom.
Arvid: We always try to be in direct contact with people. The tribe is one thing, but we did amplify it. We needed to amplify it, we needed to project a brand into the tribe that was both part of the tribe and solving something for them on a professional level. We were the teachers that helped other teachers. I think that’s the transferable component as well. It works in software development. If you sell a product to engineers as an engineer, and you’re part of their community, you have a reputation as being a good engineer—if you have that—that really helps because you sell it through your reputation, you sell it through some sense of authority, and having a following already.
You were talking about this in the episode with Ruben as well. If you don’t have an audience, if you don’t have a market, there is an audience somewhere, you just haven’t found it yet. I believe that there are tribal or at least community structures for every single thing, for every kind of thing we’re part of. I’ve been talking with a lot of consulting clients recently, and they are in specialty foods, they are in document processing for specific kinds of jobs, they are in many kinds of groups of people.
Whenever I talk to them, after a while, we figure out there is a niche community. There is a group congregating somewhere. There is a meet-up for this in this city. There are three people that go there at all times. There is this online bulletin board where people that look like it’s from the 80s, but that’s usually a good sign that shows that there’s a community that has been around for a long, long time hidden away in this one niche location. You just really have to find it, and become part of it, and not do the marketing way where you start trying to push your product from the beginning. You just become part of it because you want to be part of it.
You want to talk to these people, you want to help them out, and then eventually, you can put your product in there because it’s a natural progression from being a part of the community to wanting to help the community.
Danielle: There was a lot of integrity that we built through our brand. Something I didn’t expect to do was release a product that wasn’t quite perfect or finished. Anybody who knows me knows that I like to get things as close to perfect as possible before revealing it to anybody, but this would have been detrimental to FeedbackPanda. Arvid quite nicely balances me out in this regard.
We released the product, then we got a lot of great feedback from the teachers in the community. They thought that he was a wizard because they would give him their feedback, and then in a day or two, he would release the feature that they had wanted. I’m not saying that you should build every feature that somebody asks you for. We really looked through all of those examples, what really deserved to be a part of the product, but even that was just reinforcing this dependability on FeedbackPanda.
Rob: It’s great to be makers who can ship fast in a space where all the software sucks because you don’t have to be that good to be really, really impressive. That’s cool. I love anecdotes like that. We used to ship stuff in the early days of Drip the same day. Someone would send a supportive comment, “How come it doesn’t do this?” I would literally build it two hours later, send them an email. Those were some of the funnest ones.
I’m curious. You’re growing this company, it’s obviously moving very quickly, why did you not hire someone to help? Was there just not enough work that you needed help, or was there some other reason for that decision?
Arvid: There’s a short and a long answer. The short one is I’m an idiot, and Danielle could tell you why.
Danielle: I would say that be careful when you get advice from books and podcasts without really critically examining whether they apply to your situation. We were both guilty of taking this advice, “Hire late, hire late, hire late,” to heart. What we made now in retrospect can say, we should have built a roadmap that said, “Okay, when we get to this goal, when we reach this amount of support tickets, or whether it’s tied to MRR, or to the amount of work that you have, we could have said this is when we’re going to hire.” We didn’t do that because we didn’t expect to grow that quickly. We thought we had more time.
Arvid: We didn’t really set goals much. The one and only goal that I set for myself was, “I want to grow this business to the unimaginable number of $50,000 MRR,” and that’s going to be impossible so might just as well set it as a goal. Once we hit that, “Yeah, what now?” Even before, once we hit, $10,000, “Oh, this works,” or $20,000, “Okay, now we can pay ourselves.” Then $30,000, $40,000, “Oh, this is still growing.”
There was never a moment of scheduled reflection. I’ve been writing about this recently, I just remembered. The thing I called is continuous validation, but not just of a product but actually of your business. Are you still in it for the same reasons? Are you still fine with what you’re doing? These kinds of things, had we done this—and we did it sometimes, but by far not enough—we would have understood—at least a year in—that it would be fine to hire a customer service person.
You just alluded to it (I guess) that there wasn’t enough to do for a full 40-hour a week position. To me, in my complete absence of knowledge and experience, if you hire somebody, it has to be a full-time position. That was what I thought. Of course, you don’t. Of course, you can have somebody do this for the two hours a day that most people are asking questions, and then do the rest later, or early, or whatever. To me, in the middle of it and always being interrupted by conversations on Intercom at random hours, it felt like, “Might just as well do it myself until we have somebody to take it all away from me completely,” which meant that I blocked all of this.
I thought, “I could do it. I can take care of it.” Yeah, it caused me a lot of sanity, caused a lot of anxiety that I had to learn how to deal with. In retrospect, once we sold, we had to hire our replacement so there was no way around that. When we did that I figured out, “Oh, this is actually easy and enjoyable and I should have done this immediately.” That’s one of the big learnings for me.
Rob: It’s a good point, and you’ve touched on two problems really that a lot of founders face, especially in the bootstrap space. One, we have the savior complex where we feel like we need to do everything, and it is hard to delegate, or you have a mental roadblock that won’t let you hire someone part-time, you just don’t think about it, or you just push yourself to the brink of burnout and exhaustion.
The other thing that it brings up for me is the lack of goals, lack of planning thing, it’s a problem. There are reasons that larger startups and larger companies have all these planning meetings, forecasts, and have goals. I know that so many of us—myself included—we leave the corporate world because there’s so much […] that you have to deal with. Then we don’t want to go start our own startup and put a bunch of […] into it, but there’s a balance.
If you throw it all out, you can find yourself in that situation where you’re not taking care of yourself. Frankly, you’re not being a good steward of your own mental health, nor of your company. It can actually hurt the growth of your company. There’s a lot of negatives to come back. I’m glad we touched on that because folks should be aware that it’s not all the way on the left, all the way on the right. There’s a middle here where I do think even in smaller startups that we need to be mindful and do some thinking and planning about this stuff.
Piggybacking on that, I’m curious, during these two years of fast growth, you’ve already referenced a lot of ups and downs. The growth is amazing, it’s fun, it’s exhilarating, the pain and anxiety you’ve brought up can be debilitating, and it can take a toll on you. Can you take me to a moment or a series of moments that you remember that perhaps were a low point during the journey?
Arvid: I have a couple of good ones and most of them (funny enough) are technical. That’s what I was doing mostly. I was responsible for building the product, for maintaining the infrastructure. FeedbackPanda was built on essentially a cloud-native thing. We put our software into Docker containers, and they were running on Google cloud in the end, but we didn’t start there.
We started with a small, affordable, call it “local” cloud provider—some company in Germany. They were great until they weren’t. We ran into connectivity trouble, and they had issues providing the service. Our service would crash. That was already at a point where we had a couple of a hundred customers. Here I was having to deal with the maintenance that I had to do to actually get the service back up at random times because this other startup that was just the same size as we were had issues configuring the cluster, the infrastructure that they were providing for us to run our software on.
That stuff started to accelerate. First, it was a couple of minutes of downtime a day. Then it was half an hour, and then it turned into full downtime for a day. I was just sitting there. I think we were in Canada even at that point visiting Danielle’s family. I was just on my phone, trying to get their weird little JavaScript UI (that didn’t work) to restart the containers because they didn’t have a CLI yet or an API even where I could communicate with their system and then restart stuff automatically.
I was sitting in somebody’s home—that wasn’t my own—trying to fix the service that was running on infrastructure—that was also not my own—and that I had no control over because they also were like four dudes from Germany trying to build the next AWS. Big mistake, right? It was cheap, and I didn’t want to look into the Google Cloud documentation, so we chose that. Big, big mistake.
It took the knowledge of a very capable friend, and a couple of hours late at night to actually migrate this whole thing over—just a couple hours by the way. I just pushed it as far away from me as possible. I tried to believe that they could deal with the problem, get back actually solving it, and allow us to have our service running, but at some point just couldn’t, so we’ve migrated over to Google Cloud. It had zero seconds downtime for the next 400 days.
That was a big problem. I was just trying to stay focused on it because I couldn’t sleep. Our monitoring system was always essentially waking me up with a call. I had to go restart some containers and try to manage it. It was brutal. I did not sleep well, and I still have some sort of monitoring PTSD whenever a call comes in. Just one split-second, “Is it the server again?” even though I don’t even own the business anymore. I still have that to this day.
Rob: People ask founders why they sell their companies. It’s like these times. I’ve been there, so I know what that feels like. Danielle, do you have thoughts on this?
Danielle: Yeah, I was right there beside him, obviously, even more powerless when a service that you’re using just goes down.
Rob: Was that your low point as well?
Danielle: 100%. It happened more than once with this particular server before we moved over to a better solution. There’s one other point, though, just to add a different perspective. We’re co-founders and partners, so when something personal in life happens, it was really happening to both of us. For example, last winter, my grandmother was very sick, and she ended up passing away. Arvid really picked up the slack for me, but being a co-founder with your partner does mean that you’re experiencing those personal events together whether they’re good, celebrations, or are tragic. Just something to consider. It’s not a low-low.
Arvid: It is a low, just not a business low.
Rob: It definitely makes it a lot more complicated to have, that dual relationship of co-founder and partner. I get it. We’re running short on time but I have two more questions for you that I’d love your thoughts on. The first is when you started building the FeedbackPanda, did you intend to sell it from the start? Was that thought in your mind? If not, when did that thought first occur to you?
Arvid: Before we actually started building the business and before Danielle was even teaching English online, I was working as a software engineer in Hamburg in Germany. We live in Berlin, so that’s a two-hour commute there, and then two hours back. Three days of the week, I would commute for four hours, and I pretty much had nothing else to do (because it was on a train so I wouldn’t have to drive) reading or listening to podcasts.
I was reading a lot of books. A couple of the books that I read at that point were the E-myth by Michael E. Gerber and Built to Sell by John Warrillow. Built to Sell has really stuck with me from the beginning because I learned in that book (in particular) that building a sellable company is building a company you can run forever, and building a company that runs without you where you as the owner actually benefit from the company running, and other people, systems, or automation are doing the work.
We never really intended to sell the business, but we built it as a sellable business anyway because that was (to us) the optimal way to build a business. That when we actually got people that were interested in purchasing the company, then we thought about selling it. From the beginning, we just really wanted to build a cool, efficient, and mostly a highly-automated business because at least I am very lazy and I didn’t want to work too much. Turns out I had to work quite a bit, but we still made it highly automated, and hence sellable, which really helped when it came to actually sell the business.
Rob: That’s a good point that you make. I think founders get involved in this binary thinking about, “I am never going to sell my business,” or “I’m building it to sell.” Building it to sell from the start helps you no matter what. The idea of you said you’re lazy, that’s an adaptive trait for a founder because you’re lazy, but you’re willing to put in hard work is a thing. Lazy, I think about just means you’re highly efficient, and having a goal of having an automated business bringing in tens of thousands a month is something all of us would aspire to.
Wrapping up, I’m curious, are you going to do it again? Do the two of you have the desire?
Danielle: Coming 2020, no. Just kidding. I would love to hop back in. In fact, I co-founded a company while we still own FeedbackPanda with the hopes of building out this project. Absolutely, I would love to get back in. It’s quite interesting. I feel that our roles—since we’ve sold—have reversed. I was very focused on our brand image and nurturing the audience, and now, I’ve really been looking into the no-code movement and trying to build up this product, another SaaS essentially.
Arvid: Whereas I have started pretty much building an audience and writing. We really switched. When we sold the business, we both fell into this place where we didn’t know what to do. We’ve been making our way out of it, and I did this through writing. Started a blog, started a podcast, started a newsletter, and just committed to actually putting the information that I had in my mind, that I had learned over the last couple years both with FeedbackPanda and all the prior things that I’ve been part of, put that in writing, and then I started being on Twitter, talking to people, and building an audience.
I wrote a guide and wrote a book, eventually. All these things, that is where I’m going, but of course, I’m still a software engineer. I still want to build things. We’ll see where we really end up, but there’s a lot of projects that are already on the horizon. There’s always people coming knocking on our door with new ideas, suggestions, of course. I don’t think we’re done. We are now at a point where we can actually make much riskier and more interesting decisions in the projects that were involved in. Why not, right?
Rob: Yup. It’s a great place to be in. It’s something a lot of us—many, many of us—aspire to. Thank you guys so much for coming on the show. If folks want to keep up with you on Twitter, you’re @arvidkahl and @SimpsonDanniK. We will link both of those up in the show notes. Arvid, you’re doing a lot of writing at the bootstrappedfounder.com. If folks wanted to check out Danielle’s website, it’s simpsondanielle.com. Thanks again for joining me today.
Danielle: Thank you so much.
Arvid: Thanks a lot.
Rob: If you have a question for me or one of our guests, please, leave us a voicemail at 1-888-801-9690 or email questions@startupsfortherestofus.com. If you’re not subscribed, you should search for startups in any podcatcher, and we’re going to be in the top few. We do have full transcripts of each episode—even though they are a few weeks behind—at any given time. Thanks so much for hanging out again with me this week and I’ll see you next time.
