Myles Shedden, Founder of Chroma | How to NOT Find Product Market Fit

Pablo 0:00 This one's a bit of a different episode. Normally on the show, we interview founders who have made it well beyond product market fit. I got some feedback telling me to also have some stories of things that didn't work. I interviewed a good friend of mine, Myles, really sharp, hardworking founder who started a company called Chroma, which ultimately never really got to true, profitable market – product market fit. He made some pivots along the way. He got some real traction; but it just never got to the other side. He ended up having to shut down. This episode's full of lessons, part of it, frankly, of what not to do. That's just the name of the game. I think as first time founders, and I've been there, we make tons of avoidable mistakes. That's just the reality of it. After you go through it once, two or three times, you start to figure out the things that work, the things that don't, what leads to fake growth, what leads to real product market fit. It helps, of course, to hear from the mistakes that other people made. Myles here, he shares. He goes really deep. He's very forthright. On top of that, he gives his perspective on what he wished he had done in different times of starting Chroma. Here it is, check it out. Welcome to the Product Market Fit Show, brought to you by Mistral, a seed-stage firm based in Canada. I'm Pablo. I'm a founder turned VC. My goal is to help early stage founders like you find product market fit. Today we have Myles, the founder of Chroma, and now a VC at Longboat Capital. I'm really excited for this episode. It's really different than what we've done so far. If you look at all of our episodes to date, they're all success stories. This is how you do something right. I was talking to a listener a few weeks ago. His feedback was, I know a lot of these people fail. I want to hear some stories about people who try to do it and it didn't work out. I couldn't think of a better person than Myles. I've been involved one way or another in his journey and really seen it up close. Really thoughtful founder, really sharp founder. Ultimately, Chroma didn't work out. I think what we're going to do here is just pull out the mistakes you made along the way, the big ones. Everybody makes mistakes. What were the really big ones that if you could go back, you would change and maybe how you would do things different. Anyways, Myles, tell us. At the beginning, how did you – what was the first idea for Chroma and how did you come up with it? The First Idea of Chroma Myles 2:30 I was at SkipTheDishes at the time. I was living with our head of product, one of the co-founders, our head of growth. We all lived in a house called the Skip House at the time. Our landlord came by with paper lease and he said, “Guys, can you fill out this lease, scan it, and email it back to me?” All of us just sat there and were like, “This guy manages 4,000 homes. There's no way he drops off all these leases.” It stuck in our heads. When we were working, we would come back and say, “Man, I still can't believe this guy's dropping off these leases.” When we talked to him, he said, “Yeah, sometimes I drop off a couple a day.” Have you ever heard of DocuSign sort of thing? They've already solved this. As we all started to move towards leaving SkipTheDishes, we started to look – think about other things that we could do. I've always believed that you should do what you're good at. The things that we were good at were operationally intense businesses that benefit from network effects and are geographically spread out where you can launch cities, zones and things like that. And it was all from our time at SkipTheDishes. We started to put the two together almost as simple as restaurants, couriers and customers are like landlords, property managers and renters. It's not a perfect fit. That's how, in our heads, I think, it shook out. We started to say, “Well, if we could streamline and build a platform behind the scenes that all of these users could use, we could streamline a lot of these things and just make it a lot better of an experience.” Turns out, many people had that idea much before us. That's basically the genesis of the business. We tried to leverage what we were good at in a somewhat analogous space to try to build a big company. Pablo 4:18 It's not uncommon, right? I think the hope is you find a customer pain point, then you figure out how to solve it – that resonates with you and you figure out how to solve it. You wouldn't be the first person to say, what am I good at? What are my strengths? Or even the other way, which is I've built this product, where should I sell? I've built this technology. I guess that's very common. I have this IP. Where can it work? It doesn't always fail, but there is something a little artificial about it. Maybe just, it's a bit more of a risk that you'll never land on something real. I guess the next question is, you have this thinking. What validation do you do, if any? Myles 5:01 When I joined Skip, we were – the company was already successful. I wasn't there in the early days. I wasn't part of that discovery element. I think I maybe didn't know how to do it properly. What we did is we had conversations with as many small and large landlords as we could. It's not about just having those conversations. It’s about what happens in those conversations. We had somewhat made up our mind. Before we had any of those conversations, we started sketching out, here are all the buckets of work that you would have to do as a landlord and a renter. Here are all the features that we could build in order to streamline that work. By doing that, we came up with a solution before we knew the problem. Just like, oh, we are going to intuit our way to a platform that solves all these problems. When we started to have those conversations with landlords, small and big, and renters for that matter, we asked Asking the Wrong Questions Myles 6:00 the wrong questions. They always say, judge somebody for their questions, not their answers. We asked the wrong questions. In retrospect, we led people to what we wanted to hear. I'll give you an example. We would say, "What are your biggest pain points?” Trying to pretend as if we were asking the right questions. They'd be like, “Well, we have to use all these softwares.” Blah, blah, blah, and renters don't pay on time. We don't have the good renters and we don't know until after. They damage places, maintenance and things like that. This is embarrassing in retrospect, but we'd be like, “Well, what if there was one platform that could do all of that for you?” They were like, “Yeah, that sounds amazing, honestly.” They'd be like, “Well, how much would it cost?” We're like, “It'd be even cheaper.” They'd be like, “Yeah, no brainer, we would for sure do that.” That was like all we needed. We were like, “Cool, we'll build it.” In retrospect, it sounds so dumb. We basically just created a fake scenario where people would tell us what we wanted to hear so that we could go do it. In retrospect, now that I know a little bit about that industry after two years in it, we started it the – totally the wrong way and with the wrong wedge. We didn't actually listen to customer's pain points. That's why, a year and some into the business, we realized that we needed to pivot because the cart was so far in front of the horse that it would've been – we would've been this – in zombie mode raising 500k, 500k, 500k to try to build out this platform that no one would use until it could solve all their problems. I could see that that was just years away. It was the opportunity cost was not right. That's why we had to pivot. The discovery we did was totally backwards. Pablo 7:47 What you're talking about there is honestly – it's the most common mistake that first time founders make hands down, early stage zero to one. That's the most common mistake. I did the exact same thing. Lee and I, when we started Gym Track, we did the exact same thing. We went out. We had the system for gyms to automatically track their wearables. Our customers were gyms and gym operators. We went out to 25 gyms and we got 25 letters of interest. We did it by telling them what they needed to hear. It was wild. It was what you're talking about exactly. We're going in there and we're like, “Well, imagine it could track everybody's workouts.” They’re like, “Well, people here do this workout with these bands. Would it track that?” Yeah, of course it would track that. Yeah, why not, right? Everything is yes, yes, yes. You're in some weird sales mode where you don't actually have – it's weird, right? You want them to say yes because there's some maybe ego thing plus something, some momentum piece that you just want to be able to be like, yes, I'm onto something. I'm going to get to build it. You get nothing out of it because you don't even have anything to sell. One thing, if you had a thing that you'd be like, oh, great, then sign here, give me some money. You actually haven't built a thing yet, so you're just getting false signal. It's wild. Myles 9:06 Totally, yeah, I think – you’re bang on. I think one of the interesting things is, it's very low rung thinking. It's very ego driven. It's very tell me what I want to hear driven, as opposed to scientific thinking where you're like, let's try to understand. They call it customer discovery for a reason. Let's try to discover a bunch of stuff we don't know. We were starting from a position of, we already know what the answer is as opposed to being like, we don't even know what the problem is. In retrospect, when we did pivot and didn't move into doing proper – what I think is more proper discovery, we were a lot more successful. To your point, I think it's common. That doesn't make me feel any better, but I do think it's common. Founders Should be Scientists not Builders Pablo 9:51 To your point, I think this founder as a scientist is the right analogy. Founders are builders. That's, in a sense, the problem in the really early stages. Before you build something, you've got to know why and what you're going to build. Founders don't start off as scientists. Obviously, they start off as builders. That's why they're founders. That's why I think of the four startup mode. There is research mode. It gets cleansed over. It gets skipped over. When you do that, unless you're either solving your own consumer product and you're the person on the other side or you're in the industry already for decades, in which case you've effectively already done the research one way or another, if you glance through this, you're not going to hit. Anyways, going back to your story, you get all these yeses. You take it to being something positive. How do you leverage out – what starts happening as you start building this product out and try to get it into market? Myles 10:45 Really predictable outcomes, actually. What happened was, when you build a business that conducts operations, and frankly any business, you build it around the software that you use. What I mean by that is, if you have to do 10 things and your software solves four, you need people or processes to solve the other six. In the landlord world, most of whom already use – particularly larger-scaled landlords already use some type of software. If we go in and say we can do payments, which is what we did, way better than what you do today. They're like, “Wow, that's really cool. This is way better. What about maintenance? What about new renters and new applicants? What about background checks? What about this, what about this, what about this?” What we came to learn was that even if we did one of those things better, in order for them to use our product, they would lose all the other ones. All the four other things that the software – their current software does, even if it doesn't do the one as well. They were like, “Well, who would do this work? We don't have people to do those things because our software does it.” Their answer predictably was, “Let us know once your software can do all of the stuff that you said it would be able to so that it can swap out for the current stuff we use. Because until then, we're not just going to go add all these people.” Once we heard that, we were like, yeah, obviously. What our maybe more ego-driven answer was, “Well, we better build this stuff quickly, as fast as possible.” We hired more engineers. We hired more product people. That was when we started to take a step back and say, “Wow, this is like a three-year build.” Pablo 12:25 How many people did you get to without any meaningful revenue? Myles 12:30 I think 17. Now, not all of those were full-time. We had some contractors. People spending part of their days on Chroma was like 17. To put that into perspective, a friend of mine has a startup in Toronto that has her and her sister. They do $50,000 a month in revenue. It's probably more complicated than what we were doing. They just do off the shelf stuff and they just grind. Pablo 13:04 Well, walk me through this. What does it feel like to have – there's nothing wrong with having many employees. Obviously, you want to grow a business, you’ve got to hire people. There's something weird I've noticed because I did this same mistake, same thing at Gym Track. There's something offsetting about – or some weird feeling about having all those people doing work and not any real customers. What was that like? "One more feature" never works Myles 13:30 I think what we told ourselves was, "It's coming, it's coming, it's coming. We have to keep building or else, if we stop building, then customers won't sign up.” I got this really great piece of advice from a good friend of mine who's a VC in Vancouver. He said, "Honestly, I've never seen a company become successful after they release that one last feature or the next feature.” It's always the first thing they do. Think about Facebook. Back in the day, you could only write on someone's wall and set your relationship status. It was the simplest company in the world, and yet everybody signed up for it. It wasn't until they had a billion users or whatever that they launched groups, marketplace and blah, blah, blah. When I think about a wedge, that's what I think about now. What is that first thing that you're launching that people sign up for? We just didn't have one. Payments, and frankly, payments – landlords have been getting paid for 5,000 years. They find a way. That's not the pain point. Out of all the things we could have chosen, we probably chose the worst one. That comes back to what we thought at the very beginning, which was if we had all that money, 50 billion a year in Canada going at 500 billion in the US going through our system, that would be inherently valuable. Even if you just think about that, that is obviously not the problem. That's just a totally adjacent thing of money coming through your system. Well, it only happens if you're solving someone's problem. That was not the problem, just a whole bunch of very naive, silly mistakes. Pablo 15:05 Well, it's funny. Again, it's the same thing. It's starting at the end and working backwards versus starting from the beginning. You're going and you're saying, “If we had all this payment volume, we'd be a really big company. How do we get that?” Versus, what problems should we solve today? Where does that lead to tomorrow? Here's another question. What do you think now? You're now on the other side. You’re looking at companies. What would you say to a founder that's a good sign that you have too many employees? Having too Many Employees Myles 15:32 I'll give you an example. I consult a company. When I joined, they did, I think 8,800 a month in revenue and had a burn rate north of 50 I think, something like that. My first reaction was, we're upside down. For every dollar revenue we generate, we spend five. I think there's maybe a point where, to some extent, that's acceptable. Especially in SaaS-based businesses or software businesses, things like that, I get it. For me, a big indication is maybe less about how many employees you have, but more about what the business profile is and where are we in that journey? If you're, call it, I don't know, pre 10 grand or 20 grand a month, if you have more than five people, I don't know what you're doing. Especially if you're start – trying to start small. I you go raise 10 million because you're a third time founder, yeah, for sure. You might be able to buy your way through some product market fit to some extent. My sense is, if you're trying to organically build a startup and iterate your way to a really nice business – and frankly, Facebook started the same way, which is – I don't necessarily like the company, but I think it's a very great story about how to build a company. They had three or four people at the beginning until they were quite large. My friend in Toronto who is building her startup is a great example. You should have meaningful traction and have some clear indication of product market fit before you start bringing people in. I think it's just capital at the time was so easy to raise, particularly coming from the Skip background, a lot of people would be like, “Oh yeah, cool, these – it sounds like these guys know what they're doing. They'll build another SkipTheDishes.” Frankly, we weren't even the ones that built it. Nonetheless, we were able to raise not crazy capital, but a million bucks pretty easily. That allowed us to get really sloppy in terms of how to go to market, how to come up with a plan and how to do customer discovery. We were just like, “Well, we'll just start hiring engineers.” I oftentimes said first time founders over index to product and under index to distribution. When I'd say that, people would be like, “Oh yeah, that sounds right, roughly. Who's on your team?” It was just me, a head of product and six software engineers. Just a bunch of people that don't know anything about getting a product to the right customer. Not only did we probably know the right answer, we just completely ignored. Pablo 17:58 On that question about too many employees, did you ever feel like you were making work up just to keep people busy? Myles 18:05 No, I don't think so. No, because there was so much to build. When we were trying to build that platform, there was so much to build. You could just look at the platforms people use today and be like, “Okay, I get that problem. We understand the applicant problem. Go solve that a better way.” Not really, there was a lot to build. When you're trying to build a big platform like that, I don't have any problem with doing that in theory or in principle, but you should find the right place to start. We started at the totally wrong place because we started, to your point, at the end and with what we thought the solution was instead of what the problem was. There are real problems in that space. There is, I think, an opportunity to build a platform in that space. You've got to go about it the right way. Pablo 18:48 Moving forward on the narrative, there's a point at which it shifts a little bit in that at least you put out a product that does have pull. Tell us what that product was and how you got there. In lending ,Having Crazy Pull Does Not Mean you Have Something Myles 19:00 Yeah, about a year and two months or a year and three months in, we had to take a hard look in the mirror and say, "This is not a venture scale business in its current form. We would have to go raise a bunch of money and spend a bunch of time building this.” It would've been a slog and probably not a great use of everybody's time. There's some really, really capable people on that team that were – that had left high-paying jobs to come and work at a startup on little or no salaries. In having a couple conversations with some advisors like yourself included, people just said, “Look, I don't know if this works. I think you’ve got to maybe rethink things.” At that point it was clear that we had not done the right discovery. We started to say, okay, what is the right discovery? I think we read books like Crossing the Chasm, The Mom Test, the classic books that everyone should read before they do a startup because they're really important. I think if we – if I'd have read Crossing the Chasm before we started Chroma, we would've started it totally different. Maybe not even in the same space to be honest, but I didn't. About a year and two months in, we realized that we needed to pivot. We knew there was probably something on the renter side of things, maybe less so the landlords. It’s a pretty competitive space. There's a graveyard of companies who have tried to build a platform in that space and failed, but probably something on the renter side. We ran a bunch of surveys on Mechanical Turk from Amazon, SurveySparrow, a handful of these things. We did primary research with friends and renters from Reddit. We sent out a bunch of pretend ads that people would click on to see what keywords and things were driving interest. We started to do what we'd call actual proper discovery. We'd say, rank, if you could do one of these four things, which one would you do the most? Blah, blah, blah. A couple things that were like what is the biggest problem in terms of this, this and this on a monthly basis? We spent a bunch of time coming up with surveys, questions and trying to understand real people's problems. We launched the rent now, pay later product after strong indication that there was a lot of people that wanted to split their rent payments that wanted to pay when – pay their rent when they got paid, not necessarily on the first of the month. There's a stat that I'm not sure how exactly accurate it is, but 50% of Canadians roughly live paycheck to paycheck. If you live paycheck to paycheck, paying your rent with one full paycheck means you have to budget the whole other one and so – over the month. It can be really complicated and challenging for a lot of people. I'm not sure that housing prices and interest rates have helped that. A lot of people go to payday loans. A lot of people go to – or just late and get charged late fees, pretty exorbitant late fees by landlords, or they get insufficient funds, which is like $45 from your landlord and 45 bucks from the bank. It’s pretty punitive for folks who can't make the payments. We basically said, “Look, pay us half your paycheck on or before the first. We'll pay your landlord the full amount on the first. Then you pay us the other half on your next pay.” To fairly little marketing, we had pretty crazy demand for that. Now, in retrospect, we're giving away money. A lot of people sign up when you give away money. I'll sign up if people give away money. I think where we got to was that we had finally built something that people wanted a lot. That was the first time that we were like okay, cool, we've finally done – we've created some value here. I think what — where we started to get a little – in a little bit of maybe trouble was, as rates rose, we started to get compressed on unit economics because our cost of borrow went up or would've gone up. The credit worthiness of our customers declined, broadly speaking. That business model works, I think, only when capital is cheap, available and rates are low. Pablo 22:56 I'll stop you there for a second because I – touching on something that's broad which is fintech lending especially is a weird business. I mean that because in most cases, when you have crazy pull, it means something. As you said, if you're asking, “Hey, who wants money?” You're going to have crazy pull. Now it doesn't mean anything. The only way that it means something is if the underlying unit economics of that are very much in your favor. I think what we saw over the last few years is, a lot of what we thought was fintech innovation was just cheap money. That's all it was. You look at a bunch of these buy now, pay later platforms, a bunch of other fintech lenders in alternative spaces. It was like, oh, money's free and these people in this world aren't really getting it as easily as they should be. I'll come in the middle and I'll do that. It worked until, all of a sudden, money was worth something and became expensive again. The side you're lending to wasn't able to absorb that, or could only absorb it up to a certain point. Now you just get compressed. This middleman thing doesn't work. A lot of businesses died because of that. In a sense, it was good that you were still small. It wasn't as bad, the fall. I think you saw it earlier than many. It's a good thing to note because, again, usually, pull means something. When you're lending, when you're giving away money, when you're giving away something, it doesn't mean nearly as much. Myles 24:27 Yeah, to that end, I think we – it's almost like a sheet code a bit in terms of product market fit because you're like, to your point, “Who wants money?” Everyone's like, “I do.” Is that really creating value? I'm not saying that leverage and capital can't create value for people. What I'm saying is, it's a lot easier to create value that way, at least from a pull perspective, than building a product that people will actually go and pay for that isn't money. I think, yeah, I think you're right in that regard. I think you're also right in that we were lucky to some extent to see it earlier than, I think, lots of people, not as early as that maybe we would've liked. There was a real scenario where we were going to go – we had been approved for a $15 million line of credit at – for credit committee from a major bank contingent on a $3 million equity raise. I don't know if we could've got that equity raise done. We might've been able to. We had to have a serious conversation amongst the founders of – the three questions we asked were, do we think this business model's rock solid? Do we think that we could actually – we go through a downturn for 18, 24 months, 36, whatever it ends up being, and is Chroma a place that A players still want to work? Then finally, do we believe that the equity value of the business will increase substantially over the next couple years, such that we can raise another round at the end of this and become a successful company? The answer was ultimately no, no, and probably not. That's why we decided to return capital to investors. I don't want to say that the timing is wrong. I don't want to blame external factors because it was frankly our discovery, our failure to do discovery that led us down this path. It took two years to figure out that what we built wasn't really that valuable and then pivot into something that wasn't a great business. It was a really wonderful product. Our customers loved it, but it was not a great business. Pablo 26:24 Let me ask you this, because I think relentlessness is such a big topic and not giving up is a huge part of being a founder. You decided not to give up for a long time before you ultimately decided to pull the plug. Now that you've had some time to think about it, what's your advice on that? When should you just say, you know what? This just isn't working. Let's all go do something else. When do You Pull the Plug? Myles 26:49 I think it's probably when it's just – probably a couple factors, one, when you dread getting up to go to work, it’s probably – and you don't see a way to get out of that. Going through hard times, of course, at Skip, we went through like a ton of them. At Chroma, we went through a ton of them. In every job I've ever had, you go through a ton of them. When you're like, I – at least I'm not the right person to be part of this team anymore because I'm not giving it my 100%, that's probably the time, broadly speaking. I think there was more call it finance, business-based reasons for us. We asked those three questions. The answers were no. When all those answers are no, what are you doing? I knew that everybody on the team could go and get three, four, five times what they were making at Chroma. They've got families, livelihoods, rent to pay and things like that. To go out to investors and ask for 3 million more dollars and to go to a bank that I've known for 15 years or 10 years and ask for 15 million bucks with the prospect that it's probably not going to work and that it was, probably, us just not admitting and having too much ego to say, “Hey, this is not a good plan.” That's why we decided to wind it down. Chroma was not the last company that I want to build. When I think about the future, it was about how, okay, at any point, how do we steward our shareholders' capital, at best? How do we try to do the right things for the people that have put their trust, money and confidence in us? That's why we decided to wind down. Recap Pablo 28:15 That's perfect. Well, perfect time to wrap it up. I think as a recap, you came out of a really successful Canadian company, out of Skip with a few other really strong people. You had a really honestly compelling team, which was an asset; but I think you started thinking about it from, okay, we have this asset. How do we leverage it? Versus here's this pain point, here's this market. How do we attack it? That ultimately led to this motion where you ended up spending, let's say a year, two years trying to fit something that the market was just not ready to take in. You ultimately shifted and really did true research. When you went through it, you did real surveys, you talked to a bunch of different potential customers and found a real problem. Unfortunately, the timing wasn't perfect. The innovation was more around lending at a time where that wasn't really going to – couldn't apply. I'll say this, the amount of learning that you got – because it's the same thing as me back with Gym Track. The amount of learning you got in that time period cannot be replicated anywhere else. That's why in a way, if you actually are – as long as you're actually going all in and you're really trying, if you're a founder and you go through failures, which the vast majority of founders will go through, it's not wasted time. That's part of the lesson I want to draw out here. Yeah, okay, here's how you don't do it. You should definitely do customer discovery and research and all this good stuff. The flip side is, if you really go through it, you actually are making something happen and you're going all in on it, you will learn so many things that it's like a super cheap MBA or whatever. You think of it like that because, honestly, you come out of it a completely different person. Myles 29:52 Yeah, I think I was really torn when we decided to wind down between doing something else right away, founding a company or going the VC route because I felt like I had just – all of this stuff I wish I knew when we founded Chroma, I had just learned. I was like, now I think we could do a 10 times better job building a successful company. At the same time, building companies is really hard. I decided to try to leverage what I'd learned throughout that process to support other founders who are building great companies, maybe take a little bit of a backseat for myself for a couple years and try to support the ecosystem in Canada, particularly in the energy space, which is what we focus on, which is what my background was prior to SkipTheDishes. Yeah, you’re bang on, though. I don't know if the lessons are better if you fail than if you're successful. I think they're probably somewhat comparable. I definitely remember every single mistake, every single big mistake that we made, how we got there, why we made it and what not to do. When I give advice to people, it's not here's what you should do. It's here's what you should definitely not do because I can tell you where that ends. You're right. It's a good learning experience.