What I learned from interviewing 11 brilliant marketplace minds

Sjoerd: [00:00:18] Hi, I'm Sjoerd CMO at Sharetribe and I'm your host. Welcome to the final episode of Two-Sided. This is also an experimental episode because we are going to try to do something a little bit different today, a summary, uh, sort of, "too long, didn't listen" version of what I thought were the five most interesting take aways from the entire season. Feel free to also use this as a guide to maybe check out episodes that you haven't listened to before. I will introduce the different topics, so you'll have to listen to a little bit more to me talking today than usually, and then for each point we'll listen to some anecdotes and outtakes form the different episodes that illustrate the point, hopefully very well. But before we go in to this content, since this is the final episode and I'm recording this here by myself from my home studio, also known as the garage, I'd love to hear from you if you have feedback, ideas, or especially if you happen to know some awesome marketplace entrepreneurs I should talk to, reach out to me. You can email me to sjoerd@sharetribe.com. I'll spell that out so that is S-J-O-E-R-D@sharetribe.com or you can tweet to @sharetribe and I'll read it and react. I'd really love to hear from you. Let's get started with the five biggest takeaways of the first season of Two-Sided. So, the first one is also straight but it's the most important one. If you are in the early stages of building an online marketplace, if you are to take away one thing from this entire season, let it be this: constrain your marketplace, make it small. This is something that already came up in the very first episode with [Lenny Risicsky 00:02:03]. So just for context, Lenny, he worked at Airbnb for seven years and then after that while trying to get better at advising marketplace startups, he interviewed a whole bunch of folks at 17 of today's biggest marketplaces, including Airbnb, DoorDash, SunTec, [SE Huber 00:02:17] and a whole bunch more. Out of his research came the follow...

Lenny: [00:02:24] Maybe the most interesting work under this work is just that there's essentially just four steps that every marketplace goes through to get started and to crack the chicken and egg problem. So therefore I'll go through all four and then I'll kind of dive in to each one. So the four steps turned out to be, constrain your marketplace. So the smallest, kind of the minimal viable marketplace. Then concentrate on one side of the marketplace, supply and demand. Then build supply then build demand. So those are the four rough steps, and so to dive in to each piece, constraining the marketplace... Basically the first you want to do... And every marketplace did this except for one, which is kind of interesting, uh, find a way to reduce the size of your marketplace to either a market, say like, New York or a category, say like antiques, whether that's new or IKEA furniture or [inaudible 00:03:13].

Sjoerd: [00:03:13] So Lenny mentions a couple of things in here. First of all is, how can you constrain your marketplace? There are basically two axis along which you can constrain your marketplace. Number one, geographical location. So this can be town, city state, country even, but also neighborhood. And then number two, by category. So that is either a product or service. This kind of depends on your marketplace. And I think it's important to understand that you probably can not make this small enough, like, later in the talk Lenny mentions a very cool metaphor that he's learned from Sarah Tavel, who is a partner at Benchmark, a venture capital firm. Lenny: [00:03:47] And then three is something that, uh, that I think about as Sarah Tavel, who is also a partner at Benchmark, has this metaphor of trying to build a white hot center in your marketplace, which is essentially, it just like something that's working really well, even has a tiny, tiny scale and only once you feel like you've got that working, does that make sense to expand and start growing.

Sjoerd: [00:04:07] Mm-hmm [affirmative].

Lenny: [00:04:08] And so my advice is try to find something that's working well, whether it just feels like this is working. Supply is meeting demand, people are happy, they are coming back.

Sjoerd: [00:04:17] Sure as an entrepreneur, you should dream big and you should set goals but don't the mistake with the marketplace of casting too big a net early on. So this last outtake also takes us to why you should constrain. So we discussed how you should constrain but category, but why? So, you should try to create the smallest circumstances in which the following can occur: so number one, supply is meeting demand, two, there are enough transactions, three, people are happy and they are coming back. Lenny, basically spells out here a sort of hierarchy of needs for your marketplace to be successful. So, let unpack that a little bit. So, supply needs to meet demand. So usually this is phrased as "liquidity". So let's go into that for a second. So liquidity is basically the situation where there is enough relevant math on both sides of the platform, so supply on the one side and demand on the other side. That transactions are likely to take place. Now, that sounds very simple but it actually requires a load of work to achieve. For example, if you are a local marketplace, and with that I mean supply and demand need to be in the same geographical location, so best example is like a car sharing marketplace where if you own a car you can rent it out there and if you don't own a car you can find a car to rent there. So if I'm renting a car, I need to be in the same physical geographical location as the car and since this one is about private cars, probably also, the owner of the car needs to be in a reasonable physical vicinity. So let's say we launch this in Helsinki, Finland where Sharetribe is from, If we want any transactions to happen on our fictional car sharing marketplace, we need to have a significant amount of cars in Helsinki on the one side as well as a significant amount of interest through people who would like to rent in Helsinki, Finland. So, just getting those two sides onboard through one GO and one category of product is already a lot of work. Now, imagine if you would add another category of product in this marketplace, motorcycles, bicycles, helicopters, whatever, if you were in only one location, you need to then add supply and demand once for that new category. But if you are and 10 locations, suddenly adding one new category becomes a lot of work because you need to fill that hole on the other side 10x or times whatever you perceive as the amount of locations that you have. Sophie Adelman, shared some terrific insight in episode six on that or how she saw that and how she works for Hired, which is a marketplace for recruiting technical talent.

Sophie: [00:06:48] So how I think about it is, so you have products in GO and each, if you are thinking about the products and GOs, each one of those squares within that matrix is actually it's own little marketplace. So you can think about supply and demand between candidates and opportunities and in different GOs... That would be three sided anyways in that regard, but then each one will have a liquidity ratio between supply and demand. So you need to think about that when you are building it out. And sometimes that will mean you making decisions about you not moving into new GOs because that will create more complexity, sometimes that will mean you cutting the number of products that you offer. In our case we offer software engineers, data scientists and designers, but, and product managers. We found that software engineers have high liquidity whereas designers didn't, but actually if you break it down via different GOs, you might find that central London might have high liquidity but East London did not. And so you really have to be really sensitive to these things and think about the constraints that will affect your product being delivered to the market.

Sjoerd: [00:07:57] So Sophie is basically saying here that you're not just building one marketplace. The more GOs and the more categories or products you have, you are basically building a mini multitude of different marketplaces, like, each of which has their own liquidity ratio, and so if you have too many of these marketplaces early on, the more work it will be and the longer it will take you to actually get liquidity going. And liquidity, by the way is not stable.

Sophie: [00:08:23] I think, one lesson that I've learned from working two marketplace businesses is you need someone who is always thinking about that supply-demand dynamic and you're always constantly adjusting between the supply and demand. It's a fine balancing act which is what makes it fun because you're not just bashing one thing, you need lots of different tools the whole time.

Sjoerd: [00:08:40] Now, once we get the first transactions, the next step is of course to get enough transactions. You need to achieve high liquidity in order to figure out if this is a scalable business idea. If you only get, you know, you can only imagine, coming back to the car sharing marketplace example, if you only get a couple of transactions in Helsinki per month and scaling to a new city takes you another two months then likely something is missing in your marketplace idea. And then of course the last step, if you get enough transactions, you want to make sure that people that are having them are having them are happy so that they come back for this repeat transactions, this repeat purchase ratio which is very important. Now, only once you figure these out should you start to think about expanding. And so the point is with this first takeaway that these objectives are much simpler to achieve on a smaller scale. And so this topic of constrain is something that came back throughout the season. Of course, because I asked her about it, but we saw evidence in every conversation. For example, Ruthie Amaru from Freightos, a marketplace for freight, told us in episode three.

Ruthie: [00:09:45] We changed the pitch to, "Hey, can we bring you more business?" And obviously a bunch of them said, "Yes." We got a nice group, we got them ready to go, and we published and we were ready to go with worldwide coverage for anybody who wanted to import or export. And that's in work, right, you can not start a marketplace that's that big because we'd have, you know, small importers, we discovered that there is a very big difference between an importer who is bringing in 30 prototypes from China, you know, just to play around, uh, and a Walmart, right?

Sjoerd: [00:10:18] Right.

Ruthie: [00:10:18] So suddenly you discover, "Oh my goodness. Th-" And then you discover, wow, people who are importing in to the United States are totally different to people who are importing into India a- an- and the world is so big. You look at things from the outside and they all seem so simple, the minute you step in... And thinking about that point, I heard a great podcast by the founder of Fiverr who said... No, it wasn't Fiverr I think it was actually Get, who said, "You know what, when we started Get, which is a tax paying app, we started it in one block." There was like, one little square block, right?

Sjoerd: [00:10:49] Yeah.

Ruthie: [00:10:49] And I was like, "Ah, that." Right?

Sjoerd: [00:10:52] Yeah.

Ruthie: [00:10:52] So we basically narrowed down... We went from the entire world and everybody in the world to very small importers and exporters from China to the US.

Sjoerd: [00:11:03] So they are trying tp be the freight platform for anyone in the world to anywhere in the world and it just didn't work. So they narrowed it down to very small importers and exporters and only between China and the US. So in a way, the did a double constrain, both in category: small importers and exporters as well as NGO for the entire world for just between those two countries. Now, it doesn't always need to artificially be constrained. For some marketplace it just comes naturally, for example, Kickstarter, in episode 10, Charles told us the following...

Speaker 6: [00:11:38] So with projects like that, right-

Sjoerd: [00:11:40] Yeah.

Speaker 6: [00:11:40] ... It was projects like, we were all embedded in the arts community of different sorts, right? And A, that was the thing that bonded the three of us as co-founders, but also got us to focus on, almost without distraction, like, this particular market.

Sjoerd: [00:11:56] Yeah.

Speaker 6: [00:11:56] And so those are some of the first projects that got launched-

Sjoerd: [00:12:00] But many other guests have had the same experience as Ruthie. They needed to artificially constrain their marketplace. Here is Charles Armitage from Florence which is a marketplace for elderly care homes, can hire temporary nurses.

Charles A.: [00:12:13] Yeah, so we started aggressively in West London.

Sjoerd: [00:12:15] Okay.

Charles A.: [00:12:16] Hyper local marketplace, so you know people don't want to travel more than, you know, half an hour to a shift so that limits your geographical constrain, and you know, day one, and let's say you are trying to pick up an Uber in London and there's one driver and one rider in the whole of London, well, the marketplace doesn't work, does it?

Sjoerd: [00:12:31] No.

Charles A.: [00:12:31] So it's heeding that point of liquidity as quickly as possible as, you know, essential and it's ultimately the main task. We kind of know anecdotally it's coming more and more evidence-based as we grow but the platform works organically when you've got about a hundred shifts per week and about 60 works to do those within a one hour, sort of travel time radius.

Sjoerd: [00:12:49] Okay.

Charles A.: [00:12:49] And this is what we kind of arrived at as our current success metric. And at that point, you know, shifts get replaced by care home and it gets filled in a suitable amount of time with the right worker with the right skills. Now, day one, again, not knowing what we know now, you know, we ran in to... If I'm a nursing home I might need a load of different startups, I might need a nurse, I might need a care worker, I might a, uh, you know, cleaner, a chef, whatever it is. And we kind of went in expecting to provide all of those staff types day one, and certainly both nurses and care workers, and we realized quickly that by spreading ourselves thinly across both nurses and care workers, we were really, really impacting the user experience of the platform because all our efforts, you know, at those early stages, your time and efforts are really, really valuable, so spreading too thinly was really deleterious. So we actually fairly soothed into the process probably two months in. So we actually decided to stop doing care workers and we said, "Well, actually we are going to concentrate on doing one thing well, which is matching nurses to nursing shifts in care homes and dominate that in one small chosen market. And then we gradually went to London and then it took us months and months to go to the south east and then a year to go [inaudible 00:13:55] city, and then by that time we kind of started to work out how to make the machine work right across the country. And we've only recently reinstalled care workers on the platform, because you know, we wanted to reach that liquidity for nurses and shifts across the country before we went on to the next thing.

Sjoerd: [00:14:10] And also in our last episode, episode 11, Brian Clayton from GreenPal which a marketplace for lawn mowing, also hyper local shared some similar views and the great, catchy book title which basically summarized this entire point.

Brian: [00:14:24] So our is a local marketplace, so we are geographically constrained, so you know, fast forward we are in every major city in the United States and every one of those towns, cities had to be built from the ground up. And one thing that we did do right in the early days was that we red every resource we could and one book that stood out to me was a book called Mail It Then Scale It, and I knew that we needed to nail the playbook in our home town first before we went to other cities and it was excruciatingly tough because it was slow. And so we knew that we had to make it happen in Nashville consistently and reliably before we try to go to any other city. So we fit three years in Nashville and then our first market outside of Nashville was Tampa, Florida, another year there and then it wasn't until year three that we launched another handful of markets. And then once we started understanding, okay, this is how we recruit supply, this is how we recruit demand size, this is how we make sure this goes right, I don't know 75% of the time and these are the things we are doing to make sure things go right 99% of the time-

Sjoerd: [00:15:34] In here, Brian actually gave another great reason for scaling down first, is that you can optimize your playbook. You should try to achieve this, let's say operational excellence at first and then you can scale. Well, I hope we drive the point home by now, so let's move to another interesting insight. You don't need a complicated, extensive product in the beginning. So I spoke to Josh Breinlinger in episode two, who is an investor at Jackson Square Ventures who also was early employee at oDesk and Rev.com and he gave a great insight, because what is actually the product of your marketplace? The product of your marketplace is not the software that you're building...

Josh: [00:16:16] For most marketplaces that I see, I think the important thing that I understand is, the- the product is the liquidity, right? It's not about a really pretty user experience, it's about the interactions you have in the marketplace. And so you know, in the real estate world, the product is leads or in a used car marketplace, you know, if you are trying to sell a used car, it doesn't really matter that much what the app looks like, it matters did you find a high quality buyer real quickly? And so if someone can come along and figure out how to aggregate that demand quickly, then you have something phenomenal that people will pay for, so you know, a lot of times embracing that means just getting them more leads or getting them more business, which is ultimately what most people want.

Sjoerd: [00:17:02] So the product of your marketplace is your liquidity and however you make that liquidity happen, that doesn't matter that much. Listen to what Brian had to say...

Brian: [00:17:12] We had something very similar in the first two years of the business. We had a crappy code marketplace and so it was a very hard to use product, it was very difficult, it was clunky, it was unreliable and basically for the first year our product was some crap that we cobbled together and a little chat bubble in the lower right hand corner. And basically- and since we would hand crank everything, the introduction of the quoting process, the home owners actually calling the service providers and making sure that they would show up on time. If they didn't show up, calling around like crazy trying to find, uh, a replacement. We spent three years in Nashville, Tennessee trying to perfect the process of a home owner trying to log on to the platform, gets five quotes in less than a minute, reads reviews and looks at data about these service providers and then hires one and then they actually show up on the day they are supposed to do the job and do a good job.

Sjoerd: [00:18:12] And that wasn't just the case for Brian. Chelsea, also showed their rehearsed version...

Chelsea: [00:18:16] So one of the things I was going to say about marketplaces early on is that all marketplaces when they start to facilitate it, the- a lot of people think that they just work but actually behind the scenes you got people who are doing this matching, because in the marketplace it's all about getting the supply matched to the demand and the demand matched to the supply and helping them find each other and it makes that connection then for the transaction if that transaction is to occur. So when we first started we didn't have a platform, we had a very basic website with a form that people could fill in, both for candidates and for employers, but people actually just called us up so we had a phone number on the website and people would just call up and say, "Hey, I want to do an apprenticeship." If they filled out the form online... In the beginning it was incredibly manually and the first iteration was done on paper and spreadsheets on Google and we had the roles up on a whiteboard, um, in the office, and then we had the candidates in spreadsheet and we need to manually do the match, send the email and it would all be done like that. Now, what's great about that is, if you can build a business where it is... You've got the supply and demand dynamics, it's a marketplace but you can actually generate revenue without any technology, then you know you have the ability to develop something that will be more facilitated and you can invest in building that marketplace.

Sjoerd: [00:19:12] At Freightos they just had a button that just send an email.

Ruthie: [00:19:43] I mean, I'm sure we wouldn't be the first one to build the search and then a select and then a big old book button, then our book button was pink at the time, I remember that. And all the book button did, I was certain nobody would ever hit that book button, but what that book button did was actually send an email to me-

Sjoerd: [00:20:02] [laughing]

Ruthie: [00:20:03] Or someone else, an operations manger. We'd optimistically, you know, hired an operations manager and I was like, don't worry you're not going to get... And then we get this email, right and we're like, "Okay, now what? Somebody actually hit the book."

Sjoerd: [00:20:18] Yeah.

Ruthie: [00:20:18] So we're like okay, well, you know, it's sent to the operations manager, you call the buyer, I'll call the seller and let's see if this is actually a real transaction, if we can make this happen. And he called the buyer and he says, "So, how do I pay?" And he's like, "Uh, pay? Uh, PayPal." [laughs] And then he said, " Oh, okay. What's the payment fee?" And he's like, "Uh, three percent." [laughs] You know, the processing fee. So the first few transactions were, you know, obviously took a lot of hand holding-

Sjoerd: [00:20:53] And so some of you are probably thinking, "Well, of course I knew that already." But you'd be surprised that even though the lean startup and similar methodologies have been forming the world, still a lot of people don't realize this, even some of our guests learned this the hard way. Charles shared this fantastic story of the very first version of Florence.

Charles A.: [00:21:10] I remember we had a little bit of cash to start off, not very much at all and neither of us really knew anything about how to build a product, how to talk to users, anything like that so we kind of thought, "Well, there's this problem and we need this platform to do a hundred things." You know, we need to manage the whole life cycle to do all these different things and we need it done in the next two weeks and we want to spend this amount of money from it, and so, you know, you go to the internet and ask the internet if someone can do that for you and someone somewhere puts up their hand and says they can and you know, we ended up spending few months quite a lot of money to get not even anything approaching an MVP that worked, and that was a bit stressful. But it was a really good point because what we then did is in that process you obviously start speaking to users and you know, I went and sat in a care home for a few months and drove the nurses to share some things like that and you know, you speak to people and you kind of come to the realization yourself that actually, what I need to do here really to solve 80% the problem is a very, very simple product and actually we ended up conforming back to just Google sheets where the nursing homes would go into Google sheets and put in their shifts and then 10 nurses would be found, [inaudible 00:22:16] which they, "Yeah, interested in that, that one, that one and that one." And then that was it, there you go you got a platform there.

Sjoerd: [00:22:21] So another great lesson, you don't need to have a fancy platform in order to get liquidity going. Now, that brings us to the next subject, how do you keep people on the platform? There are of course like multiple ways of doing this, but two that really stood out to me this season... The first one, and this seems to apply mostly to marketplaces where one or both sides of the platforms are businesses or professionals, so B2B and B2C is just this: build amazing tools. So these tools should most be around time saving or real pain points, that's what professional are experiencing, and almost all B2B marketplaces that I've talked to this season, they had this as a component. So for example, episode five featured CREXi which is a commercial real estate platform. Michael DeGiorgio, the founder, he told me about how they build an easy way to solve a number of pain points that every real estate broker had to go through on a weekly basis.

Michael: [00:23:17] And then, you know, another kind of based upon is all really simple, which I think was part of the beauty of it. On like a Friday, you'd then go take all those notes and summarize it and send it to you client and say, "Hey, this is all the marketing I've done. These are the people I've spoken with, this is who I know is interested in it, this is the feedback I'm getting." And it's then these all like, overview reports for all the deals you are trying to sell. Every Friday. And I figured, you know, let's make that a click of the mouse. So, not only do you know real time who's looking at your property, but you can also supply them with information and a lot of the document signing and the process of like, signing a confidentiality agreement and exchanging documents to get the buyer to know more about the property, who that buyer was, all the notes you had, literally, all the marketing you've done, you literally had a button at the end and it spits out that report for you to send over to your seller or client so they know exactly how defective, you know, the sale is going.

Sjoerd: [00:24:10] So CREXi built something that made half a day's paperwork suddenly doable with the click of a button. Of course it's a fantastic value proposition. Scientist.com which is the modern place for ordering scientific work did something similar. Here's what Kevin Lustig had to say when I asked him how they keep the transaction around the platform.

Kevin: [00:24:29] The market place in many ways can be considered a three way market place. I didn't want to bring that up too early in our discussion. And because most of our clients where the business comes from are large pharmaceutical companies and these large pharmaceutical companies today they have about 100,000 people and they often have about 1000 people in procurement. And what they do is, they're working in 200 different research areas and they put a few people in charge of each research area-

Sjoerd: [00:25:00] Yeah.

Kevin: [00:25:02] ... and these people then often get to determine how items within their categories will be purchased, okay? So procurement is the third side. You have the pharmaceutical researchers, that's one side, you have procurement who is particularly focus on making sure that all the regulations are followed and that they did in good value and then you have the suppliers as the third side. And so for many of our biggest clients, the chairman who's actually gone out in particular areas and told the researchers that the marketplace is the only way that you can acquire human samples, because it's the only way that pharmaceutical company can be 100% sure that when those human samples are delivered, that they were obtained from ethical suppliers in an ethical way, that the patient that donated the human tissue has actually given their permission to use it. So basically we have really used the compliance as a way to get recurring revenue, as a way to prevent the scientist from going off platform.

Sjoerd: [00:26:12] And also GreenPal, the marketplace for lawn mowers, they build out the tool set for the business side of the B2C marketplace.

Brian: [00:26:19] So, as a service provide, let's say typically somebody calls you off of your Yelp page, "Hey, I need a quote. I live at you know, 1532 Main Street, can you come out and look at it?" You have to drive out there, look at it, walk around, maybe meet with the person, write them a written quote, leave it in the mailbox, hand it to them, hope they call you back. "Can you come out Thursday?" "No, I can not come out Thursday, maybe Friday." "Oh, I really need it done it by Thur-" You got to go through all this for every single person that tries to hire you. When you use GreenPal, you get 30 of these opportunities a day, it's right here and you know, you're presented with all of this information, okay? It's like here is the aerial imagery of this person's property. Here is the street view of this person's property. Here is how many square foot their lot size is, here is the average winning prize on the platform for that zip code. They want their lawn mowed very two weeks, not every week, they are actually expecting a perfect job, they've indicated that and they've also indicated that they want to hire somebody for the rest of the season. And so, we give the service provider all of this rich data around the service request, that by the way, the home owner was able to pop in less than a minute and we're able to make this thing so much smoother than just doing it the traditional sense on both sides of the transaction.

Sjoerd: [00:27:37] An additional point here to make is that from the marketplace that we interviewed, there's been a difference in the timing of the tools. So, some of them like Freightos for example, they build the tools first and then added the network or marketplace on top. This can of course be a great way to get supply on board early and it's sometimes referred to the single player mode. I think also Lenny described this in the first episode. Another way is to start with the network and then build the tools later, like with what GreenPal did for example. So both come for the network, stay for the tool and come for the tools, stay for the network seem to be viable approaches, hopefully this point is clear by now, but I hadn't realized before how important this component could be. Now, another way to get people to stick to your platform is to build community. This is easier said than done, and of course, likely this will mostly apply to marketplace where there is a consumer or component or at least one side of the marketplace are individual, single professional freelancers for example. And not surprisingly, this community component is used for Communo, which is a marketplace for creative freelancers. In episode eight, Ryan Gill, the founder and CEO told us...

Ryan: [00:28:48] My belief was although contrarian, if we can create enough value on the platform for the supply side and the demand side, outside of the transactional side of things, we can build a real community, build real tools for them, build events, those types pf things.

Sjoerd: [00:29:02] Yeah.

Ryan: [00:29:02] We could command subscription payments, and I believe that that would also help weed out a lot of shakers, people that want to be... And when everyone has skin in the game, ideologically you care more about the outcome of the service or you care more about the outcome of the product and how you were as a client because you're paying to be on there. And so, I- I- I also have to thank [inaudible 00:29:27] they were pioneers but I believe it being "free" and I use the air quotes and I know this is a podcast, but air quotes for free. It's never free, when you know it's free, you know you are paying through the [crosstalk 00:29:39] somewhere else, and so the transaction take is always a bad solution in my opinion because it encourages that this is mediation an we don't care about that.

Sjoerd: [00:29:47] And this brings me to another interesting insight this season, subscription marketplace. We haven't seen a lot of that this season necessarily, but Ryan makes a really good case, especially if you a boot strapping your marketplace or you don't have any venture capital behind you, if you know you are offering enough value and it won't be hurting your marketplace, in fact you could go for membership or subscription model early on. There's a couple of benefits, first of all it creates predictable revenue, like Ryan says, but also the concept of having skin in the game might actually really, really improve the quality of your supply but also some other marketplaces added a subscription layer later on, like for example, CREXi, initially they offered their tools for free but once they felt that it was providing enough value they started charging separately for the use of the tool. And then let's move on to the final insight, which is maybe unpleasant to hear for anyone about starting a marketplace but getting the supply onboard early on is probably manual unscalable work. So, like Sophie already mentioned, a lot of the people think, you know, you launch the marketplace and things just automatically start running. I think I can safely say that that was the case for absolutely zero of the people that I interviewed. Almost all of the marketplaces that we talked to did some kind of formal direct sales. Jacob Wedderburn-Day, the founder of Stasher which is the platform for finding temporary storage for your luggage told us about how they got supply onboard early on.

Jacob: [00:31:21] We've got better methods now, but the, uh, early days, and to be fair it does work, was this, you know, we'd pick an area and we'd just walk around we'd see like, see a promising looking business or hotel and we'd go in and we'd either have the meeting there and then or we'd try and set up a meeting, you know, if management wasn't around. It was actually- it was amazing sales practice, I still feel like it's the kind of training... Every time we hired someone new at Stasher we'd make them do at least like one day of just kind of like foot sales like that [crosstalk 00:31:49] that's good business training and for us, I think especially l- l- looking early days when we only had a handful of posts on the platform, it was quite important to have that kind of in-person sales experience of understanding, you know, what it is they liked, what it is they didn't like, what we pitched that made them excited, what we pitched that made them nervous [laughs] and try and strike that balance in the way... Because that ultimately feeds back and helps you form the product as well.

Sjoerd: [00:32:13] And Jacob touches on a great point here, that it has all kinds of benefits because you will really learn the pain points of your customers, you'll be able to sharpen the value proposition and you'll also build great relationships as Charles from Florence can confirm. Charles A.: [00:32:27] Yes, how do we get the first customers? Well, on the demand side, on the care home side, I found a list of all the care homes in London and I sat down with a phone and just called them all up one by one [crosstalk 00:32:38]. Three to one-

Sjoerd: [00:32:42] Okay.

Charles A.: [00:32:42] ... Who you know, had twenty shifts a week they needed filled so, you know, promised them that- that we would bend over backwards to try make it work and ultimately ended up having to pay huge amount of [inaudible 00:32:52] into the relationship early doors because the product didn't work. In that relationship building, that unscalable relationship building, you're learning huge amount about your customers you can make a great sort of place for developing your product-

Sjoerd: [00:33:03] And Sophie faced similar kind of challenges when she was trying to kick start Hired in the UK.

Sophie: [00:33:09] So I was running the UK market, but really I was just boots on the ground, I was, you know, just hustling, I would be in Ubers all over London, I called it my mobile office. I would get in an Uber, tether my laptop to my phone, be doing my emails and calls in the back of an Uber going from meeting to meeting and just hustling to get companies to use this platform that nobody had even heard of.

Sjoerd: [00:33:32] And let's end this episode with my favorite unscalable anecdote of the season-

Brian: [00:33:37] We need to learn how to design, build, execute, launch and distribute technology.

Sjoerd: [00:33:43] Yeah.

Brian: [00:33:43] We had to learn how to do all of these things ourselves, my two co-founders and I. And the first thing we did was we passed out over a hundred thousand door hangars around Nashville, Tennessee to try to get people to use this thing.

Sjoerd: [00:33:55] Yeah.

Brian: [00:33:56] And I got bit by a dog while doing this and I realized that ten users per dog bite was not a scalable user acquisition strategy.

Sjoerd: [00:34:06] All right, that was it. Of course there is lots more, loads of very interesting lessons that you can draw from the interviews this season, but these were my personal top five so to say. So final recap of the recap, number one, constrain your marketplace, number two, you don't need a complicated product early on, number three, build tools to solve day to day pain points from your supply sides for both acquisition and retention, number four, build community around your marketplace for retention, and five, getting the supply onboard early on probably requires unscalable manual work. I hope that this episode has given you something to think about or otherwise inspired you for your marketplace idea or if you already have a marketplace, perhaps there are some lessons here that you can immediately apply. If you don't yet have a marketplace but you have a marketplace idea, please look at Sharetribe. So we build marketplace software which allows you to get your marketplace platform up and running and into the market in a fast and affordable way. No technical chops. You can find us at https://www.sharetribe.com. Also, this is a right place to send a big thank you to my colleague Oli and Mira, who were instrumental early on in giving me very constructive feedback about how I interview and big thanks also to our designer Jan, for the awesome designs. I still think we should make a the-shirt of the logo, almost let me know if you are interested in buying one. And a big thank you to [Namanja Koyaya 00:35:39], who has been editing this podcast, for indulging my last minute changes and also of course for making me sound a lot more fluent than I actually am. So, if you're also considering doing a podcast, you can find him at podcastproducer.org. Now, finally to repeat what I said earlier, I'd love to hear from you. If you have feedback, if you have questions, if you happen to know awesome marketplace entrepreneurs I should talk to, if you have any additional insights into the discussions that you've heard here, reach out to me through email, sjoerd@sharetribe.com. S-J-O-E-R-D @sharetribe.com to tweet to us @sharetribe.com, I'd love to hear from you. And of course the final thank you goes to you, our listeners. That was it. I'm signing off. Thank you for listening.

Speaker 1: [00:36:35] Thank you for listening to Two-Sided, the marketplace podcast. If you enjoyed today's show don't forget to subscribe. If you listen on iTunes, we'd also love for you to rate and give us a review. If you got inspired to build your own marketplace, go visit https://www.sharetribe.com it's the fastest way to build a successful online marketplace business. Until next time.