by Y Combinator2/28/2018
Convoy matches shippers with freight carriers, which helps the trucking industry run more efficiently because currently, over 40% of the trucks on the road are empty.
Anu Hariharan is a Partner at YC Continuity.
Craig Cannon [00:00] – Hey how’s it going? This is Craig Canon, and you’re listening to Y Combinator’s Podcast, today’s episode is with Dan Lewis and Anu Hariharan, Dan is the CEO of Convoy, and Anu is a partner here at YC. Convoy matches shippers with freight carriers, that helps the truck industry run more efficiently, because currently over 40 percent of the trucks on the road are empty. If you’d like to read the transcript of this episode you can find it on the YC blog, Alright, here we go.
Anu Hariharan [00:27] – Thank you so much Dan for joining us on the podcast today. Dan Lewis is the CEO and founder of Convoy, we are really excited to have him here. Dan, I wanted to start with your background, primarily because now you’re building a trucking marketplace. I was curious, what did you do and how did you arrive at this idea?
Dan Lewis [00:50] – It really comes back to a combination of background supply chain and technology. My family run a mom-and-pop off-supply company, distributing office supplies all around the Puget Sound in Washington. Every summer my job was to drive big spooner vans delivering stuff. I learned a lot about the idea of warehouse operations and delivering, but it wasn’t really something that I thought, I’m going to go and build a company and do this, I was pretty young, and then after school I ended up doing a lot of work in transportation and technology, I worked for the Panama Canal for a while. I ended up living in Spain doing management consulting, thinking a lot about outsourcing of maintenance and parts distribution, did similar work with some airlines and companies in the United States, and then shifted over and started working in tech. Actually, my job in college was building websites for the Yale Athletics department, which was a great job. It was super flexible I got 10 bucks an hour and I could work form my dorm room, but that kind of led to an interest in technology, and then after college and after doing some consulting. I was at Microsoft for a while, and I handled other technology start-ups and bigger technology companies, and so then I ended up at Amazon. One of the things I saw at Amazon, that kind of inspired me was I realized that when I was growing up we used to go to the store to buy stuff, and we would buy it the same day we wanted it, and that was option value, people called it procrastinating, I call it option value,
Dan Lewis [02:20] – you can decide the day you want to get it or not. Then online shopping came around, and you have to choose do I want to go to the store and get it today, or do I want to buy it seven days in advance, then it was five days, and three days, and then two and then one. Amazon really took hold and started raising the bar for delivery, and making it faster and faster, and all of a sudden with Prime, the decision between do I want it really soon, and do I want to buy it online wasn’t our conflict anymore. Amazon really started to take over in retail, and it was really that that caused me to think okay so the supply chain is winning, it’s not the location of the store, it’s not the ambience, how nice the sales staff is how easy the parking is, all that stuff, it’s the fact that I can get it really fast, and that was driving massive shifting behavior. We looked at a lot of different companies out there, I was looking to do a start-up in general, and I decided to leave Amazon and go pursue this, and when I was looking at transportation a lot of the companies were last mile, so delivering Instacart or Postmates or Doordash, Uber Rush or Google Shopping Express, all these different companies that were doing that, there were not a lot of companies focusing on the middle mile, getting everything in the right place so that you could do that same day delivery. That was the initial background, but the inspiration that led to thinking, let’s go and build a logistics company.
Anu Hariharan [03:49] – And one of the things I know, having talked before, is you actually worked on even fleshing the idea out a lot more before you started actually building it. Can you talk about that process, and what work did you do to really understand, “Okay the seems is interesting, but is this real?” When did you actually get convinced that you are going to go build this?
Dan Lewis [04:12] – I was liking what I was doing at my job, but I really wanted to start a company, and I have two kids, and my wife’s at home with the kids and everything like that, “Okay if I’m going to take this risk and walk away from my job, I better know what I’m doing.” Because I had done consulting, I didn’t really know another way to do it, I was like, “I have to go and do a bunch of research,” and so I quit Amazon, and before we decided what to build, we spent weeks full-time out in the field researching, so what I initially did was a bunch of online research to figure out what the general space looks like, and then I did three things in parallel. I wrote to probably a hundred different people, maybe a third of them from my background, people I knew from previous jobs related to supply I could talk to about trucking, two thirds of them were people I just find on LinkedIn, and just wrote to them or just called them.
Anu Hariharan [05:06] – And all of them in the trucking industry?
Dan Lewis [05:07] – All of them in transportation trucking, could be brokerages, warehouses, truck drivers. Actually when my family… my mom is an ESL teacher, and growing up we always hosted foreign students and foreign families in our house, and one of them was from the Ukraine, and this guy is named Sasha, he ended up running a trucking business with his family, and his brothers out of Eastern Washington, so he was actually the first call that I made. He was talking to me how he drove a truck delivering produce from Washington to Florida and some other runs as well. I started to do a bunch of phone calls, but then I went out in the field, and I was like, “I have to go talk to people, because that’s the fastest way to learn and make sure I don’t make a big mistake,” and I remember, it’s kind of funny, I remember two things, I wanted to look professional, but I also wanted to fit in, so I bought a bunch of gift cards, and I went to truck stops, and I wasn’t quite sure how I could engage people first. I went inside and looked around, and went inside my car, I was like I got to think I’m going to do this, and the easiest person to talk to was the person who was filling the tires with air. I went over and talked to that person and then started having more and more conversations at these truck stops, and learning more and more. I was trying to gather as much information as I could. Did the same thing in warehouses, just walked into a bunch of warehouses, got kicked out of a Whole Foods warehouse. The door was not supposed to be open, I just walked in and started talking to people,
Dan Lewis [06:38] – and they were like what are you doing? But over six or seven weeks, all that research and those phone calls, and people just want to talk to you about what they’re doing. We learned a lot of information, and we wrote three or four business plans, I put them together, and then had a lot of detail, and tons of quotes and anecdotes and real examples to back up the theory that you get from the data that’s online, and when we took that to investors, that was different, and I hadn’t realized how unique that was until after I did it, I thought that was normal, but then after I did it, I realized not every potential founder goes and does that level of research before they start proposing their idea.
Anu Hariharan [07:19] – And what were some of the two or three big learnings for you, from that research, in terms of how you decided to launch Convoy and what pain points to solve?
Dan Lewis [07:29] – The first thing I learned was one, people in the industry were super open and interested in talking, and interested in proving things. It was very welcoming in that respect. A quick aside is that coming from a tech background and from a consulting background, I was always used to setting up appointments, or like putting something on the calendar, and in trucking you just pick up the phone and call. I remembered so many times I tried I said on the phone to somebody, like, “Mo just call me, I’ll pickup,” so that was different, but probably the biggest insight we had early on was trust, there was a lack of trust between parties in the industry, and people generally get along, but there wasn’t a deep-seated trust that shippers were confident brokers weren’t taking advantage of them. In fact shippers often felt like brokers and some of their transportation parties weren’t reliable, or they had issues, or they didn’t trust the pricing. Brokers also felt like when they were working with trucking companies, that they didn’t get the respect they deserved for all the work, the truck companies felt like they were not necessary. Truck drivers were constantly telling me that they felt like brokers weren’t looking out for them, and that larger trucker companies weren’t looking out for the smaller guys, and they wanted someone they could count on, they wanted someone who would treat them fairly and more transparently. There was a lack of data, lack of transparency, and that led to some mistrust between the parties,
Dan Lewis [08:57] – and honestly the shipper is trying to get the best rate possible, the trucking company is trying to get the best rate possible, and the broker or the 3PL in the middle is trying to do the same thing on both sides. Everybody is trying to negotiate all the time, so it does breed a little bit of mistrust. We realized that the first thing was let’s focus on building trust, let’s focus on bringing transparency consistency and visibility into pricing, and visibility into where the trucks are, which answers a lot of questions. One of the areas that raised some lack of trust is shippers often didn’t know where the truck was, and trucking companies and truck drivers wouldn’t always honestly report where they were.
Anu Hariharan [09:37] – There was a trust issue on both sides, not just on the driver but also on the shippers side?
Dan Lewis [09:41] – Well shippers yeah that’s right, they didn’t always feel like they were getting a full straight story. That’s pretty normal when you have a big industry and a lot of different parties involved, and a lot of negotiating involved every day. ll the parties were kind of a little bit, a love hate relationship was how one of them described it to me.
Anu Hariharan [10:02] – Got it, and how its Convoy today do you think solving that trust issue on both sides?
Dan Lewis [10:07] – I’ll give a couple of examples. When we have a job that we’re going to offer to the truck drivers in our in our community, our partners, we figure out which carriers this job is best for and we offer it to them, and we show a price upfront. That’s a guaranteed price, and it’s a guarantee that there is a real job behind that. One of the things that was common in the industry, sometimes brokers or 3PL’s or even big trucking companies will post jobs online that won’t actually jobs they had yet, they post them to try and get a price to figure out what the truck do it for, and then they can take that price and go back and try to get the actual job. We don’t put jobs up that aren’t real jobs, and that was important and a guaranteed upfront price, we’re not going to go back and ask to pay you less later. There was pricing confidence, and real job confidence, and we pay the drivers really quickly. They’re also used to being stiffed sometimes and worried about not getting paid, and so we paid them really quickly, so they will have confidence that Convoy was actually respecting them. On the other side with shippers, all the trucks that run loads for Convoy use the Convoy app, so that provides real-time visibility for the shipper to understand where the load is, and when the truck’s arriving it provides, and actually another good example there is if a truck shows up at a facility and its delayed, typically more than a couple of hours in the industry, that truck is going to expect to get what’s called a detention payment for being delayed.
Dan Lewis [11:36] – Now there is often debates around that, when did you actually arrive, and the trucks going to say they arrived earlier, and the shippers going to say they arrived later because they don’t want to pay, and so what we can do without real-time visibility is solve problems like that. We actually build something that is pretty new in the industry, its automated detention, we just track it and we pay the driver and they can just do it right from the app, so they don’t have to go through an arduous process of trying to get that.
Anu Hariharan [12:00] – That’s great, so on the trucker side you do it with guaranteed pricing upfront and a real job, and quick pay, whereas on the shippers side you giving them tracking info and even helping solve for delayed shipments.
Dan Lewis [12:13] – Yeah that’s right. If you know when it’s actually there, you can see what it is you can get timestamps that are real, it answers a lot of the debates.
Anu Hariharan [12:20] – That’s great. Let’s go back to your one of Convoy, because a lot of our audience are actually early-stage start-ups. One of the big questions they often have about the marketplace, because yours is a marketplace, you have truck drivers and shippers on the other side, so the first question most of our viewers often have is how do you bootstrap a marketplace, and so how did you guys go about doing it, what was year one in Convoy, how did you find your first few truck drivers and your shippers?
Dan Lewis [12:49] – We thought, we went out and did so much research, that I had a pretty good list of truck drivers and a pretty good list of shippers right at the beginning. We thought okay, we’ll go talk to a bunch of truck drivers and get them signed up, and then we’ll go to shippers and start bringing loads into this marketplace and sort of developing it as a broker, and truck drivers weren’t interested to be honest, we went out and talked to a bunch of them, and they were like, “Well do you have any loads yet,” we said no, as much as they love the idea its vaporware. To them it was this new company, never heard of us before, and they also as I mentioned didn’t have a lot of trust, so they were kind of skeptical of who is Convoy. They aren’t going to downloading an app and start using it, they are busy and are looking for their next load right now, so we said okay got it, we are going to come back with some loads, so then we switched over and went to shippers, and we spent a lot of time again building the trust, and that’s where we could sit down with the companies who shipped freight, some early customers were print shops. Fsor example and you go talk to them and figure out, what is it you guys need, where are your pain points and they would tell us, it’s okay we’re able to do this for you. We were able to get some shipments from them, and because those loads don’t have to pick up for 24 or 48 hours from the time they’re offered to us, we had time to take that job and turn around and go find a truck.
Dan Lewis [14:12] – We actually used every single job as a marketing opportunity, and we’d use that job to go find a truck, offer them a job, they’d say, “We want to do this work for you,” we’d say “Great, download the app, and it will show up inside the app,” and it was the right kind of way to get them to start engaging with us, through a real job, and then to get the job and complete it, they would have to use the app, and that was fine for them as long as it was a real job. That then start building some buzz, especially in the Western Washington community where we got started. Then truck drivers would tell their friends, and we started getting more on, and we built referral programs early on, once we had a few truck rivers that trusted us, and it was pretty small initially, but honestly it really came down to getting the job and using that as a marketing opportunity to get a truck to sign up for the market, versus trying to get supply before we actually had any action going.
Anu Hariharan [15:02] – Yeah and most marketplaces say that, usually it’s true for them the supply always goes to where demand is, at least initially,
Dan Lewis [15:09] – Yeah that’s right then it shifts later.
Anu Hariharan [15:11] – Yes, did you start off in the Pacific Northwest, is that right in Seattle?
Dan Lewis [15:17] – Yeah we started off in this Pacific Northwest doing straight truck work, so lots of truck deliveries around Western Washington, as well as some full truckload deliveries, and honestly I think we did up to about 150 miles when he first started out, so it was pretty tight. The idea was in trucking you have local truck drivers, they typically don’t run more than 250 miles one way because they want to get home that night. They don’t have a bed to sleep in their cab, and you have regional and long haul. We focused on the local folks initially, and the advantage there was that when you’re trying to build a marketplace you’re trying to build density. You want to have a dense number of suppliers and buyers in that particular market. Because we focus just on jobs within a couple hundred miles of Seattle, every truck driver that we signed up was available the next day for another job, because they lived in that region and they slept in that region every night, and so therefore we were able to get a higher density of truck drivers quickly in that region, and start seeing the marketplace to take shape, and start learning how the business worked.
Anu Hariharan [16:22] – Got it. How many markets are you in today?
Dan Lewis [16:24] – We are just about nationwide now to be honest. Parts of the south and south-east are where we’re trying to expand to now, but we operate out of all the major metros now. We think of the world, kind of metro to metro. We think of it in terms of the shipping lanes that exist.
Anu Hariharan [16:46] – Got it. Let me ask you this, because it’s interesting, it’s always interesting to see how marketplaces scale. It’s very widely established that, Uber and Lyft went city by city, because they are the riders that are requesting rides within the city. Airbrn was all about travel, so one of the things they did very early, was they went after all the high tourist cities, New York, London. Rather than just doing US, they just did global cities that had a lot of tourist traffic, how do you guys think about scaling, so how did you decide market number two after pacific north-west, or even from local to regional, what lens do you use to expand?
Dan Lewis [17:23] – That’s a great question. We always thought about what the minimum viable footprint to serve our customers. We served both trucking companies and shippers, but initially the folks were giving us jobs and loads that we could take to go get truck drivers on the platform, our shippers, and we really had to understand what they needed. Initially as I mentioned we focus just on the North West, and we thought about two different ways to expand, one was to take the initial services that we offer, so the types of trucking that we support, which at that time was a full truckload drive-in as well as flatbed, and expand that into more geographies and more lands, or instead stay in the same geography and add more types of trucks. Most of our customers that we had to date were much more interested in us expanding to new geographies with them than offering additional types of trucks because most of them we were already servicing the types of trucks that they needed. And so we decided okay if that’s where our customers are looking for us to expand, that will help us expand, because it will give us more freight and more volume in the business as we expand. We essentially looked at, where did our existing customers want to run freight next, and that’s how we started thinking about expanding, so sort of what is a minimal viable footprint for a customer where I can feel like I’m really servicing them as a partner, and that’s challenging in freight, because I think you’re right,
Dan Lewis [18:50] – in a lot of marketplaces it’s a specific local geography, and everybody stays in that geography, whereas in trucking, one company based in Seattle or based in Portland, might ship to a hundred other cities. Another company might ship to 90 cities, with 60 overlapping with that other one. You have to look and see what are the most popular common freight lanes of your customers, and where they want to start shipping to next, and that’s a really kind of healthy way to start expanding into new geographies. That’s how we initially started going more broadly was through our customers desire for us to service them on additional lanes.
Anu Hariharan [19:24] – Got it, when did you first expand out of the Pacific North West (PNW), how long were you on that 150 mile local zone before you said hey I think we have product market fit let’s expand?
Dan Lewis [19:35] – We were probably operating in that zone for probably 4 to 5 months.
Anu Hariharan [19:43] – Which is quite normally actually.
Dan Lewis [19:45] – And so the exact progression which I’m happy to share was, Western Washington, Washington State, I remember drawing the zones on the map, we’re going to allow shipments to work in this zone, using a map of trucker drivers, Washington has some really remote areas. Then we did Washington, Oregon, Idaho. Then we did the I5 corridor, going down to LA, then we started doing the Western seven states, so we really start to expand within the same contiguous area. Then we went to Texas. In each of these different regions that we went into, we thought a lot about different major highways and corridors, and the major lanes were freight runs. That’s really the key is to understand where most of the volume is going to becoming and making sure you’re building a lot of supply there, and probably one of the things, for people that are building marketplaces now, that was the most challenging part of this, was our customers were constantly asking us to go to new areas, and for good reason, they were excited to work with us, and they wanted that we were offering in all of the distribution centers around the country for example. We had to be really disciplined to push back and say no. Our sales team wanted to sell more broadly, because they saw opportunities, and their customers would be saying look I can give you 50 loads a week coming out of the city that you don’t support yet, or this region, and we had to initially say no, and that was pretty unique for us. Most brokerages would take freight in
Dan Lewis [21:12] – the lower 48 states wherever it’s available, and tried to mark it up, and make a profit, whereas in our case we were focusing on building density and specific lanes and markets, and so we didn’t actually go national, we went kind of lane by lane, metro by metro initially. Since we’ve had a bunch of national customers, and so our goal then shifted to, as I mentioned minimal viable footprint, for the marketplace to really servicing those customers, they really needed us to be in more geographies around the country, otherwise we couldn’t really have that partnership with them, and then once we want to service them around the country, well now we need truck drivers around the country, so we start working with truck drivers that run nationally, and then those truck drivers are looking for their next load in whatever city in the country they’re in, so all of a sudden the footprint to really serve that ecosystem expands pretty quickly. That’s been one of the growth areas and challenges for us, is figuring out okay we have to expand it quickly, but still moderating in the sense that we’re not going to do everything for everyone, humans are infinitely flexible. If it’s all manual and it’s all driven by people, that person can make a decision about which geography at what price, and they can call somebody in figure it out, whereas if you’re trying to automate it, and run through a technology system, you’re trying to automate the transactions and matching in the marketplace, you have to know the scope.
Dan Lewis [22:34] – It’s really hard to train for everything right, so that’s why when you’re doing this digital freight matching and taking the approach we are using, you have to limit geography, whereas a lot of other folks don’t, they just go everywhere, and that really is indicative that they’re probably not using a lot of technology.
Anu Hariharan [22:48] – And that’s probably not the most optimal, even though human beings are really flexible, there’s only so many permutations and combinations they can do in their head, but a system can do it way better.
Dan Lewis [22:59] – Especially if you learn over time ultimately you can do a lot more.
Anu Hariharan [23:03] – Now let’s shift a little bit more to company building. How big was Convoy in year one, like team size?
Dan Lewis [23:11] – Year one, by the end of year one I think we were about 30 or 40 people. We grew relatively quickly, I don’t think I mentioned to our initial investors, I don’t think we had our bank account open when we were trying to take our first seed checks. We had done so much research upfront, we didn’t worry initially about incorporating, or getting the name set up, we didn’t think about all those things, we simply just thought “What is the right business? How do we get customer data? How do we understand what to go build?” Then we had a great story that we put together, and we went to raise our seed around, and it happened relatively quickly.
Anu Hariharan [23:51] – How many people were you when, was it just you and Grant the co-founder, or did you have more people raising the feed round?
Dan Lewis [23:57] – Yeah it was me and Grant, and then it was five engineers.
Anu Hariharan [24:01] – Okay, so you already have five engineers.
Dan Lewis [24:02] – Yeah and they weren’t full-time yet, but they were all committed to the project, and they were working on it part-time, some of them were employed elsewhere, some of them weren’t, but they were working on getting it going, and the thing I learned through that process was that it felt to me that angel investors and seed investors look for three things: the first thing they want to see is that they believe in the founders, they have a conversation with you and they walk away thinking I believe this person is going to be capable of going away and building something, checkbox one right, box two is, the idea seems reasonable, for some investors for some angel investors that means they understand the space, and they don’t invest in things that they don’t understand, for others it’s they can think through it, and they ask around, and they feel compelled that the idea could be reasonable. Most entrepreneurs that want to start a new business, can check those two boxes, they’re going to come across as a compelling individual because they’re passionate about this, and they put a lot of time into it, and the idea is maybe reasonable, the space is reasonable. The hardest part is when that angel investor is asking themselves can this person go deliver on a vision quickly, and that usually comes down to do they have the engineering talent available, or the specific skills available to them to go build the company right away, and I had learned that over time, so one thing we thought was critical when we went out to angel investors,
Dan Lewis [25:23] – was we wanted to have already working with us a group of engineers that were willing to go and that already said they wanted to start, and I found when I talked to folks that was one of the most compelling parts of it, yeah they were excited about the idea and us, but then they said, “Okay, but can you build it?” I wasn’t going to build it by myself, and my co-founder is a great engineer, but again you need a team to do some of this stuff, and so that was really the question was could we pull the people together, and already having a few people working on it, made it much easier to raise the interest.
Anu Hariharan [25:56] – And where did your first five hires come from? Was it from your network, or how did you recruit them?
Dan Lewis [26:03] – I was part of another start-up called Wavii that Google had purchased a few years back, and a couple of the folks had been part of that company that start-up in Seattle, and they’d been at Google for a little while, I think one of them had already moved back to Seattle, but they were interested in moving back to Seattle, so those were two of the engineers, a couple of them were from Amazon, had left actually Amazon a little bit before around the time that I had, were working on their own projects, actually we were sitting at a table, my friend and I kind of had a table we were working at, and we invited other friends and engineers in Seattle that were working on their own projects to come and join us. As they were sitting next to us we were kind of like hey do you want to help us with this little thing, and we got them kind of excited about it, and we actually recruited two of the people to help us just by inviting them to come sit next to us, to work on their projects. So those both were folks who had been on Amazon before, and then a fifth person I met through a friend, he was also previously at Amazon, so it was pretty tight, and I knew four of the engineers well before going into this.
Anu Hariharan [27:07] – By the end of the first year you had 30 to 40 people, and how much was the engineering versus the rest?
Dan Lewis [27:16] – At that point product and engineering was probably 40 percent.
Anu Hariharan [27:22] – 40 percent got it.
Dan Lewis [27:23] – About 40 percent, and the company quickly shifted as we started getting more into an operational mode, we started shifting more towards sales, and operations and customer success and customer support, and those roles ended up growing faster than product engineering, just proportionally. All of it is growing, and we have a great product engineering team, we brought a lot of great data science talent in recently, but proportionally it’s more sales and marketing operations.
Anu Hariharan [27:54] – That’s actually common. Most people find it counterintuitive but it’s actually common for most marketplaces in the early days, where their ops team, and maybe the sales team start scaling much faster than the engineering team, but there’s obviously a point at which you start turning that dial, but for growth, it was true for Airbnb, and it was true for Uber and Lyft as well. It’s actually quite common, so how big is the company now, how many people?
Dan Lewis [28:22] – We are just over 200 people now.
Anu Hariharan [28:24] – Just over 200 and in less than three years?
Dan Lewis [28:26] – Less than three years.
Anu Hariharan [28:27] – Okay so one of the faster scaling companies, how has your job shifted as the CEO from year one when you are managing 20 to 30 verses today, what have been the big changes you’ve had to do?
Dan Lewis [28:40] – Yeah I mean my role is constantly changing, and I’m always thinking about that too, because I do worry that I won’t see a change that needs to happen and do it well. When I first started I was, I was visionary recruiter. I was out trying to convince customers, convince other employees to join us, I had a diagram I drew, andI said okay I need people with marketplace experience and with transportation experience, and to add a few different dimensions to this, I’m going to go find those people, so I was pretty aggressive about going out and trying to recruit and find the right people for the company. Then we kind of got into it, and I switched my hat, we had maybe 10 to 15 people, and I really switched to product and kind of business development. Then I was thinking, okay what is year one, and Convoy was founded April 1st, 2015, and we launched to customers at the end of August, so it was very fast in terms of that run-up, and so all summer there was a core group of folks cranking away and building the first version of the product, so I really shifted into like designer product manager, and thinking about our go to market for that initial launch.
Anu Hariharan [30:00] – We usually call that the Doer-in-Chief.
Dan Lewis [30:02] – Yeah I was the doer in chief for sure, and then we announced and we launched, and then I shifted into fundraiser, and we were like, “Okay now we have something that we feel like it’s working well, it’s out in the market, we’ve built a great v1, we’ve built a great solid initial team, let’s go raise our Series A.” We went and raised around, from Greylock that fall. So the fall of 2015, and closed right at the end of that year, and that was my first time having gone and done a round like that, and that year, again it’s a different type of job, and I had to learn how to do that, and how to think about it from that perspective, and really think about what do we do next and adding a little bit more structure around the business and around the plans. Then I shifted back into like Salesman-in-Chief, or Chief Salesman. I remember going and spending a lot of time with customers, throughout 2016, building relationships. Unilever had called us actually, a really forward thinking shipper, and they had asked us if we were interested in participating in a pilot that they were running with different sorts of digital freight companies, and we ended up becoming their partner out of that. I spent a lot of time learning about their business, spend time with them and then spent time thinking about who are other similar types of companies we can go and build relationships with, and that was really where we start to expand more down the West Coast, we started working with them in Southern California and Texas, and we were looking for more businesses there as well.
Dan Lewis [31:31] – That year was really focusing on initial sales, and it kind of kept evolving from there, a couple of other hats, we did a lot of really strategic business development in early 2017 thinking about okay now we have some real customers, how do we get deep partnerships right. If you want to go accelerate and disrupt an industry, it’s great to have other companies in that industry and that business who want to support you doing that, and are going to make an investment in you doing that, so we kind of shifted our, so I became more of a strategy focused, strategy business development focused in 2017. My role will continue to evolve as I step back, and help the leaders at Convoy, becomes stronger, bring in the right people, develop people at the company, and really start building the culture, thinking about culture vision, and executive leadership development at the company, is really where I think it goes next.
Anu Hariharan [32:27] – We actually say role changes from Doer-in-Chief to Company Builder-in-Chief. That means you wear multiple hats at different points in time. One of the things I know you spent a lot of time on, is mission statement and building core company values, and really reinforcing that within the company, for culture, so talk a little bit about when did you define a mission statement, did you have it before you even started the company, or at what point, and also at what stage of the company did you actually list out the core values?
Dan Lewis [32:59] – Our mission statement I’ll just say it upfront, is “To transport the world with endless capacity and zero waste.” sAnd we actually didn’t codify that, that was sort of how we thought about it, we knew that 40 percent of the time trucks were running empty, there was an incredible amount of waste, by reducing waste we could improve the economics, improve service, everything gets better, it’s a win, win, win. That was kind of the vision, and we codified that just about a month ago. A little bit over two years in we said let’s write this down let’s really formalize around that.
Anu Hariharan [33:30] – What prompted you to codify this one month ago?
Dan Lewis [33:34] – Yeah it’s interesting, we were thinking about how is our company growing, there are different sizes of company where the culture starts to change, and you need new mechanisms, and you need new ways for people to stay connected and in touch, and share information. When we passed 150 people which I think is referred to the Dunbar number, because it kind of refers to the number of people this researcher believed you could have direct social interaction with and relationships with, or direct relationships with, we felt we wanted to take some of the things that we talked about and believed, and make them more like constructs of the business that everybody knew coming in would learn about and think about and it would kind of represent who we were as we started to get bigger, and then to the values, we have a set of company values, we use them all the time, we use them for every recruiting, every time we interview somebody–
Anu Hariharan [34:33] – How did you guys come up the values?
Dan Lewis [34:37] – I came from Amazon, and I’ve gone back there a couple of times, but Amazon has a set of leadership principles, and that construct we liked. We said that’s an interesting concept let’s look at that, and actually I had a chance to chat with, so Jeff Bezos was one of our investors, he talked about how they form their values, which was, I think it was about a year in, or maybe a hundred people, they decided at that point they would codify their beliefs, and what they thought worked. We had had similar ideas, kind of before we raised money, right around that time, me and the engineers I mentioned and my co-founder, sat around and said what are the values of the company going to be, and we decided way too early to decide. We wrote down some things, we don’t really know who we are yet, we can’t just say who we’re going to be, let’s be that and then write it down. It actually came about a year into the company, in early 2016, we decided okay we’re getting a little bit bigger, we were 30 or 40 people at that point, and we wanted to have consistency as we started to scale. New people were coming in to do interviews, we were trying to scale a little bit on management and review process and things like that, and kind of remember who we were, and so we said okay what do we like about ourselves? What are the qualities that we think, one of them is know why, that was really important, because if you don’t know why, you can’t act independently, you always have to go back and ask for the next step if you don’t have that context upfront,
Dan Lewis [36:06] – or love problems not solutions. We saw too often teams being named after the solution, like it could be the review’s team, or it could be the search team, where the solution is that, but the problem is I need customer confidence, or I need to find something, so we always want to think about how to be re-orient around the problem so we don’t forget the original problem we’re trying to solve and fixate too much on the solution that we came up with. We created these values that help steer some of the way we think, and we thought were really valuable, that happened about a year in. That was input from at least half or maybe more of the company at that point, thinking about what this could be, and then over three months we kind of wrote them down, debated them, we sat in a Thai restaurant for like four hours at the end, and we said were not leaving until we all agree this is v1, and we wrote them down. It’s a living document, and it’s something that we’ve actually evolved, we actually just added a new value about a month ago, which is bring out the best in others, and that’s one that represents a lot about where the culture is developing. It has been a pretty core part of what we’ve done in terms of trying to build a consistent values throughout the company.
Anu Hariharan [37:16] – And how do you test for that in your interviews, how do you make sure that the next person walking into the door is the person that you want, not just from skill set and competence, but also from a cultural fit?
Dan Lewis [37:30] – Yeah we have a pretty thorough interview process, both in person and we do some off-line stuff. We actually upfront assign people that are doing interviews different values to learn about, and then we have different questions, either they can make up their own questions, we have banks of questions we have thought of that kind of represents those values. We actually make sure that we ask them, we make sure we are a little bit comprehensive in terms of the types of questions we ask to really understand if we believe somebody represents those values. Often with any interview process is always going to come down to someone’s personal opinion as well, that always weighs in, but this is one way to think about being consistent there, and we almost always have someone from outside of the group, that’s part of the interview process as well who’s really looking at the values and saying, I’m not personally going to benefit from this person joining, in terms of my team, but I want to make sure they are right for the company.
Anu Hariharan [38:29] – And does that person’s input get any higher importance or lower or is it the same as the rest?
Dan Lewis [38:34] – The person who comes in from outside the group? Yeah, that person actually has to be brought in.
Anu Hariharan [38:37] – Has to be brought in so that’s your test for the culture?
Dan Lewis [38:40] – That’s one of the ways we test yes.
Anu Hariharan [38:41] – Got it. Now you’re almost at 200 people, do you still interview every candidate, what point did you stop, and I’m sure you’re still interviewing quite a few of them, but how do you decide on how involved do you get in the process?
Dan Lewis [38:58] – I’m no longer interviewing everybody sadly, I really enjoyed that. I talk to a lot of our other entrepreneurs that I knew or who had invested in Convoy about how they did that, and I thought one thing that was consistent with a lot of the ones that had been successful was that they had actually interviewed for a long period of time as many people as they could, so I’d probably stopped interviewing everyone after about a 150 employees.
Anu Hariharan [39:19] – You almost interviewed everyone, until a 150?
Dan Lewis [39:22] – Yeah I would do a phone screen or I would do some sort of different stage of the interview, but I would get involved and have a conversation. Honestly it wasn’t, so the thing that’s important to me was it wasn’t about me evaluating that person, it was that I wanted them to evaluate me. When you’re joining an early stage start-up, the CEO has a pretty outsized impact on your career, and the company you’re joining, and I wanted them to come in saying I believe in the company, I believe in the leader of this company. I don’t want them to meet me three months later and go man if only I met Dan before I would never have joined. It was actually important to me that they had a chance to meet me, and so what I do still is with everyone who joins Convoy, I try to get a one-to-one with them within a couple of months of them joining, or sooner, and really just sit down and listen to how they’re thinking about it and what they want to do, and we get to know each other a little bit, because it’s important to me that they have a chance to learn more about me, because they’re making a bet, and I am honored and feel like a lot of responsibility for the people who decide to join Convoy.
Anu Hariharan [40:28] – Now there are many founders we see, even at YC, and often one of the things they talk about is hiding mistakes, and it’s inevitable, but are there things that you went through at Convoy which were great learnings that you think everybody should have in their interview toolkit?
Dan Lewis [40:44] – Yeah, a couple of things. One is trust your instincts. It depends on the type of hire, but almost every hire has a significant impact on a lot of people, and if a couple of people in that interview have instincts where they feel like there is something wrong, whether something off culturally, or there’s some sort of, there was some issue related to those things, we really need to trust that. When we have ignored that that hasn’t always resulted in the right outcome, or maybe we didn’t listen as much at certain times, and we were like this person seems just right for all these other things, and so early on we learned a couple of things there. The second is we have a written component to almost all of our interviews, and a lot of them with even done quantitative tests and things like that, because we are speed dating, we’re trying to learn a lot about each other quickly, an interview in person with someone is going to show you one side of that person, but their ability to write and structure their thoughts on paper, or to do an exercise is also going to show you a lot about them as well. For different types of roles, thinking about different exercises helps you learn much more about that person in the same amount of time. We find that some people will be great and in person discussion, and not as great on the other components, or vice versa, and we needed to consider that.
Anu Hariharan [42:12] – Can you give an example of a written exercise or something?
Dan Lewis [42:15] – Yeah so one example would be, we might ask a couple of strategic questions, or one strategic question and one very tactical question, and then ask the person that’s interviewing to put together a written response to that. Then we’ll read that written response and try to ask them about it during the interview. Not only do we learn how do they write, which is an incredibly important skill for most jobs, but also how they structure information, how thorough are they, how seriously do they take it. I ask a couple of questions that are a little almost out of the ordinary sometimes, and I found those are really important for a start-up, I’ll even ask them to someone interviewing for a C-level position, and I don’t want to say them exactly, because I actually want to keep using it.
Anu Hariharan [43:03] – Okay so that’s why were not giving the question away.
Dan Lewis [43:04] – I’m kidding, but it could be something that is totally out of the ordinary. It’s like, “Hey, work on this kind of exercise,” not expected, I guarantee they didn’t think about that coming in, and I’m half interested in the response, because it teaches me. I can learn a lot from that, but I’m also interested in how did they handle that question. Are they little entitled, like, “Yeah I don’t think I should answer that, why would I have to answer that question?” Putting somebody in a position where you’re asking them to do something that maybe they think is extra work or is a little bit off topic from what they expected the discussion to be about, is a great way to learn their attitude. Are they entitled? Are they not? Those sort of qualities you really want to understand during an interview process.
Anu Hariharan [43:48] – And now that you have 200 people, and you’re probably hiring a large proportion of your leadership team as well, how do you assess if someone can scale to the next level or the next role, how do you do that?
Dan Lewis [44:01] – That’s an extremely important question and really challenging, and I know it’s something that a lot of different CEOs I’ve spoken with have had challenges with. Here’s how I think about it, the most important indicator is hiring, so if someone wants to get to the next level, and are saying how do I get to the next level, how do I scale with the company as the company gets bigger, the thing I think about, what I’ll say is, “Well can you hire the person that’s doing your job right now that’s as good or better than you at that job? Can you convince that person to come and join Convoy and work for you?” And if you can, that’s how you scale. You bring in the people that can help you scale by building a team that helps you step up to the next level. If you can’t hire those people, if you can’t recruit and attract them and they’re not ready to work for you, then maybe that’s an indication that you’re not at that next level yet, because you can’t recruit the person to work for you that will help you get to the next level, or that is at that level today.
Anu Hariharan [45:01] – That’s a very interesting test.
Dan Lewis [45:04] – Because often times what you’re really deciding is, if someone is at a certain level, are you going to hire that person’s boss? Let’s say the company gets 10 times bigger and you have a much bigger organization and need more senior leaders. Will you hire the person’s manager, or does that person step up into that role and hire their replacement? You certainly can’t say, you don’t want to say because you are a little more junior, your replacement can be even less experienced than you, you have to say no, you have to pick somebody that raises your bar, that’s better than you, can you hire that person in. If you can and they’re excited to come work for you, then that’s a path to the next level.
Anu Hariharan [45:43] – And that’s interesting, because I’ve heard two schools of debate on that, meaning some people think that anyone who doesn’t have the title of CEO makes it harder for them to attract any candidates, because people outside perceive, associate title with authority or decision-making power, and the other schools of thought is more like, “Hey let’s just put them in that role and test them as to whether they are able to scale,” because if they are drowning in water you’ll probably know that pretty quickly, so what’s your reaction especially to the first pushback I’ve heard which is, hey title to some extent give some level of authority, and that automatically allows more experienced candidates to talk to them. If you look at it, most CEOs haven’t, especially CEOs of start-ups, haven’t done any of the jobs of the senior execs that are hiring for, they’ve never been a CMO or a CFO, or a CRO, yet they have to be really good at attracting candidates who have 20 years of experience being a CMO, and one school of thought is it really comes down to okay who is the CEO and how compelling the vision is.
Dan Lewis [46:48] – Whenever I’m hiring someone for a role like that, a senior executive role, and we are talking to some people right now, half the job is doing the job, have the job is building the company and the culture. That is extremely important to me, and when I think about someone coming and joining, at whatever title, but certainly a senior role, that’s what a lot of people are looking for, they want to join a start-up, because they want to start build something, and help develop something that doesn’t exist today. Other leaders in the company can do the same thing, which is you’re bringing someone in to help you build that organization, and it’s really about selling and telling the story that you’re empowering that hire to do something bigger than they’ve been doing that they’re going to be excited about, and that you’re not hiring them to fill a niche and a slot on your team. You’re hiring them to do your job in some ways, so that you can step up and do more. When I think about a leader on my team hiring somebody, if that person approaches it as I’m going to hire the person and I want them to take over a big part of my job, I’m going to give them a big part of my job, it’s really hard to make that mental leap, but doing that that’s really what the next level is about. If you want to hire someone to fill a slot that secures your job, then you’re probably not level for the next level.
Anu Hariharan [48:06] – Yeah that’s fair, that’s a very good assessment. One last question, is another question we get from a lot of our founders is as you scale. From year one to year three and now you’re at 200 people, there is some tension about what happens to the team. The first 20 or 30 employees, and there’s usually tension as you scale in terms of, it’s more probably psychological than real, but how much influence do they still have in decision-making across many areas which they would have had in year one, but now obviously you have various leaders and various functions, how did you manage that in terms of, especially when the company went from 30 people in year one to 200 now, and did you a, see that tension from the first 10 or 20 hires, and if so how did you manage that?
Dan Lewis [48:56] – To make sure I understand correctly, you’re saying really early hires, how do they evolve with the company as it grows and scales.
Anu Hariharan [49:01] – Yeah and I think more is like how do you empower them to feel like they’re still as involved in critical decision-making, versus them feeling that there are now just one silo in the company.
Dan Lewis [49:11] – Yeah that makes sense. First off people are going to go into different roles over time. It’s really within whatever role that they are doing, how do they recognize that their job is to do more than their job, it’s to represent the company culture. It’s to represent the history of the company, it’s to represent the mission of the company, and that they have the responsibility, and also the ability to do it based on being there very early, to really have the credibility to go do that. One is just really positioning it that way, and talking to them and about them, and getting them involved in a lot of things outside their own day job. Another one is that one of the things that happens, and I actually think I learned this, and I want to give credit where credit is due, I first met you at a YC and Greylock event in San Francisco, and I believe I heard somebody at that event, I don’t remember exactly who, talking about one of the biggest challenges as a company scales is the early employees feeling that the company is less transparent, there’s less information flow to them. I think it’s completely true, and it’s not that there’s necessarily less transparency overall, it’s just that they’re not going to see as much because they’re not there. Again, a big part of it is actually creating a way to share more information, and that’s part of the know why value that we have at Convoy, which is people really need to know why they’re doing something and have that context, and so we actually want to make sure people
Dan Lewis [50:40] – are involved in things that go beyond their job, and have more visibility in the strategy of the company. We share a lot of that with the whole company every week, and that I think keeps it so that folks that are there early, feel like they have visibility where the company is going, and they can make a lot of sense of it. They’re the ones that their peers or other people who come are going to grab it say “Hey, what did Dan mean by this,” or, “What’s this really mean,” and they can help explain and ensure that more broadly across the company, because they have all that context and history to really know where everything is coming from. A big part of it is just continuing to share a lot, and then empowering them to be the champions of the company beyond their job, that’s what I found, and obviously, actually when I talk to a company today and they still believe it’s 100 percent true, I remember saying this at 50, 100, 150, 200, everyone in that room is going to be that person for the next 200, next 500 people that are at Convoy. It’s important that they are doing their job not just about completing the tasks of the day, but understanding the mission and the culture of the company, so that they can represent that and push that forward and pass that forward to the next group. We actually explicitly talk about that in company meetings a lot, but it just has to be something that is ground-up, it can’t be top-down.
Anu Hariharan [51:53] – Awesome, well thank you so much Dan, thanks for the time.
Craig Cannon [51:57] – Alright, thanks for listening, so as always you can find the transcript and the video at blog.ycombinator.com, and if you have a second it would be awesome to give us a rating and review wherever you find your podcasts, see you next time.
Y Combinator created a new model for funding early stage startups. Twice a year we invest a small amount of money ($150k) in a large number of startups (recently 200). The startups move to Silicon