GrubHub Case Study: Selling By Day And Working At Night

In 2004, Michael Evans jotted a business idea on a mustard-stained napkin. He wanted to build a web site to connect hungry customers with local restaurants that delivered. This year, GrubHub, the company he launched, will do $70 million in food orders.

In this program, you’ll hear how he got traction for his business by going door to door and selling his idea to local restaurants during the day, and then heading home to build his site at night. You’ll also hear how he scaled his business by learning to hire and by focusing on a few key features.

Michael Evans

Michael Evans


Michael Evans is the co-founder of GrubHub, a site that helps you find every restaurant that delivers to you.



Full Interview Transcript

Three messages before I fire you up with another great program. First, check out all these stunning custom logo designs that one client got from 99designs is the largest marketplace for crowd-source graphic design. Look at all these beautiful custom web designs that another client got. Go to and describe your needs. Artists from all over the world will flood you with custom designs. You only pay for what you want. Look at these cool T-shirt designs. Get your own designs at

Next, what do I mean when I say that will give you the best CD rates and the best savings rates? I mean RateBrain updates its data daily. RateBrain has more fresh and comprehensive rates than the competition, and with RateBrain you can even get an alert when a rate changes.

Finally, how many entrepreneurs have I interviewed here on Mixergy who told you that they wished they hired better lawyers when they started? If you’re a tech entrepreneur, shouldn’t you talk to Scott Edward Walker of Walker Corporate Law? Scott Edward Walker is the entrepreneur’s lawyer.

Here’s the program.

Andrew Warner: Hey, everyone. My name is Andrew Warner. I’m the founder of, home of the ambitious upstart. And you know what this is. This is a website where a mix of entrepreneurs and business people teach how they built their companies so that you can go out there build your own company and come back and do what they’re doing, which is teach and share your experiences of what you learned along the way.

So how do you take an idea that’s sketched out on a napkin and turn it into an industry changing leader? Joining me is Michael Evans. In 2004, he said, “If I’m hungry, shouldn’t I be able to go online and figure out what food I can have delivered?” It’s a damn good question. I should have been asking that back in the late ’90s. It’s important. So he launched GrubHub as a hobby. Today it’s generated millions in revenue, and I’ll ask him exactly how much and how he did it. Michael, welcome to Mixergy.

Michael Evans: Hey, thanks.

Andrew: By the way, man, such a good idea. I don’t know why this thing wasn’t up and running before and I’m so glad that you put up GrubHub. So here’s what I did. Before this interview, in preparation, I typed in my office’s address on, and I found that there are 136 restaurants in this neighborhood that will deliver here. I didn’t even know there were 136 restaurants in my neighborhood. But there are 136 that will deliver. That’s a lot. How many do you guys have in your system?

Michael: There are about 11,000 restaurants in the system, and the experience you just described is really typical. It started out basically as a hobby, and one of the things I was trying to find out was which restaurants deliver to me. And just answering that question for myself, back when it was 100 restaurants, I was like, “Wow. This is bigger than I thought it was.” And so that’s very typical. If you live in one of the cities that we service, the number of restaurants that deliver to a particular address is usually anywhere from 150 to 250 restaurants is sort of a typical number.

Andrew: Oh, wow. So you started this as a hobby. What were you doing as a full-time job when you launched it?

Michael: I originally worked at Classified Ventures at and at and Those three companies are all here based in Chicago. I was a web developer. And it was a combination of I saw a real problem that needed to be solved along with my co-founder, Matt Maloney, and didn’t want to work for The Man anymore.

Andrew: What was Classified Ventures?

Michael: Classified Ventures is . . . and are their two most well-known websites. They do lead generation for those different industries for the newspaper industry, for Tribune and Gannett.

Andrew: So when this was a hobby and you were just playing around with it, what was that first version like?

Michael: It’s funny because I think that the real key is getting the simplest product possible out the door. And so that first version and maybe I’ll answer this a little bit longer. So my partner and I had been talking about the fact that there is a need for this because what we were doing is we’d go on the Yellow Pages and we’d be like, “Okay, this restaurant’s in my ZIP code, but it doesn’t deliver to me. This restaurant might deliver to me. They’re not sure.” There’s not a real place to get any of this information. So that first version was really just like a neighborhood guide that showed in my area of Chicago what restaurants delivered to my address. It was very personal and that was why it was a just such a hobby. So that first version was very much just a listing of restaurants, sort of a neighborhood listing, that would deliver to a particular spot.

Andrew: What platform did you publish on?

Michael: Everything is open source. We used the LAMP stack. Is that what you’re asking?

Andrew: Yeah. And I’m wondering how much functionality was built into this. Was it static pages at first? Was it WordPress and you built something custom?

Evan: It was all custom, because I was a web developer trying to figure out how to do user design, which fortunately today we have people who are much better at that than I am. But the initial version, this is before Google Maps came out or anything, so I had actually edited a map program to put static maps that you could click through to find restaurants that deliver to you. It was all sort of GIF image maps in the original version. So it wasn’t quite static, but it was also a far cry from where it is today.

Andrew: All right. So if you’re building your own stuff, in addition to doing it because you’re a developer and that’s what you do, wasn’t there also a bigger vision here? An idea that this would be a big company?

Michael: Yeah. I mean we knew we were on to something when we would go and tell people, “Hey, you can discover all the restaurants that deliver to you to this ZIP code here in our city. And it’s like 250 restaurants.” And the response is just overwhelmingly, “Oh wow. That’s awesome.” And so at first, we didn’t have really a revenue model to think of or anything like that. We knew [inaudible 06:09] But it quickly became clear to us that restaurants wanted to get more orders and people wanted to order from restaurants easily and the two of those things came together at . . . it was almost like a stereotypical situation where my partner and I were sitting in a restaurant both having a burger and he’s like, “Well, what if we sell a freemium model where we list all the restaurants and then we just start providing those restaurants that are willing to pay for orders?” So we sketched it all out on a napkin . There is a big glob of mustard on that napkin. We actually have it still. We scanned it. There’s a link to it on our blog.

Andrew: I mean that very first version. Were you thinking that this would be a business? Or did you just think, hey, let me see if I can put something together for my own needs?

Michael: No, I think even in the first version I didn’t want to have a hobby. I knew that it was a hobby at that point, because the distinction of having not made any dollars versus making some money certainly what I consider the hobby versus non-hobby distinction. And even in that first version I knew that there was something to be had here. So definitely it was part of a grand plan pretty early on.

Andrew: Can you give me again the business model that you sketched on the back of the napkin? And yes, I did see it on your blog and I saw it mentioned on GrubHub too.

Michael: Sure. The initial idea was let’s list all the restaurants that deliver to a particular address, and then those restaurants that are willing to pay just a small subscription fee, we will show them in a premium place, really promote them through our email list, make sure that they generate more orders. And that original idea, it’s gone through a lot of refinements, but it’s still at the very core of what we do. We send orders to restaurants that want orders and are willing to pay for them. And so that initial idea and that initial business model has survived all six years. There’s been refinements, but it’s still 75% what we sketched out on that napkin.

Andrew: I see. So I typed in my address. I see House of Philly. I see Kabob Grill. I see New Dynasty. There’s a chance that those three companies are at the top because they paid for that placement.

Michael: Yeah. There’s a lot that goes into where restaurants show up that’s based on how popular they are and how close they are to you, what their service levels are. But one of the factors is what they’ve chosen to pay us.

Andrew: Okay. So how did you get the initial restaurants into your system?

Michael: Hammered the phones. One of the things I thought early on was, well, I don’t have to keep this idea secret because nobody else is going to be crazy enough to actually call 20,000 restaurants, because that’s a lot of work. And Google’s not going to do it because they don’t like calling people. They’ll ask us for the data, but they’re not actually going to do it themselves. And that turned out to be true. So while I was writing the websites in the evenings my business partner Matt was making phone calls. And eventually that’s a part of the process that we very quickly hired people to do. But early on, it was just hammering the phones, doing the extra work.

Andrew: Maybe Google wouldn’t jump in. Wasn’t there a concern that somebody else with some money to invest might toss people at this problem and have them bang away at the phones until they could outdo you?

Michael: Yeah. There were a lot of people were afraid of that, but I wasn’t one of them. One of the things that I think is clear is that investors don’t like to spend money on, “Hey, I can throw money at this. I can throw people at this problem in a really expensive way and outspend this really motivated entrepreneur.” You’re not going to find a lot of investors that are excited about that prospect, so I wasn’t terribly worried about it because I figured they’d much rather just throw the money at me, to be smarter about it, and then they would help me out. And that turned out to be true.

Andrew: What were you calling up the restaurants and asking them for at the time? Were you trying to sell them on listings, or just trying to sell them on taking the orders that you guys generated online?

Michael: Just collecting the data information, one of the really key pieces is where they deliver to. So finding the exact, down to ten feet of what spaces they would deliver to and which wouldn’t, that was a really big piece of the call. There’s some really intricate local data involved in that. If you live in San Francisco, none of the restaurants are going to deliver to the Tenderloin, but they’ll deliver everywhere else. And in Chicago, there’s their own little neighborhoods where people deliver. Some restaurants are willing to deliver to a boat on a lake. You just don’t have any idea unless you actually call and talk to them.

Andrew: I see. And so you had to call up just to find out where they would deliver. But how do you get the orders to them? I think it was Alexis Ohanian, who I talked about, who wanted to build something like GrubHub for mobile back when he launched Reddit and it was just too big a problem to tackle at the time. I think one of the issues was how do you get the order to the restaurant? How did you solve that?

Michael: We used carrier pigeons is what we did.

Andrew: See, no investor’s going to jump in and take your carrier pigeons business. It’s smart.

Michael: No, no one is going to compete with those. In all seriousness, the restaurants choose how they want us to communicate with them. So any restaurateur will know what their best channel for getting an order is, whether it’s a phone call or a fax or an email or using one of our own web based point of sale system, all those things. So they choose and we just make all of the options available to the restaurateur. And then the second part of that is you have to really make sure that you confirm that the order has gotten to them. So there’s a whole system in place that makes sure that the restaurants have received the order within three minutes of us transmitting it or there’s someone on our customer service team making sure that food’s on the way. Because like you said, it’s a tough problem and I know that there are customer service issues in a lot of businesses. But one of our mantras is “nobody is ever hungry.” If somebody places an order on GrubHub, they must be fed because there’s nothing worse than an angry, hungry, disappointed consumer. There’s nothing better than a satisfied, full, fed consumer. So I’d say a big chunk of our business today is making sure that those orders get through and that we know that they got through.

Andrew: But in the early days, were you calling them up? Were you faxing them? You couldn’t have had somebody standing at the ready to do all that.

Michael: Actually from very early on, we actually did have somebody standing at the ready to confirm that the orders were there. A lot of those systems have become automated in time, and we have people watching the automated systems to make sure nothing goes wrong. But even early on, we knew that was a place we were going to have to put some extra effort.

Andrew: So was it you and Matt who, when an order would come through, would sometimes call up the restaurant and say, “Hey, I need a pizza with extra tomato sauce on the left and onions on the right.” Was it you guys who were doing that?

Michael: Yeah. If that’s what it took, that’s what it took. But very quickly you realize that’s not a very scalable process, so you start to put automated things in place to take the place of people making calls.

Andrew: Like what? What was the first bit of automation that you added?

Michael: So you just mentioned a phone call, right? Why does a person have to make the phone call? Why not have a system make the phone call and say, “Type in your confirmation code”? So that was the first thing that we automated for sure.

Andrew: Interesting. All right. And you coded that up back when it was pretty much still a hobby and it was you and Matt?

Michael: Yeah. I mean you’ve got to keep in mind that both Matt and I were software developers, so the technical challenges weren’t the kind of things that we were going to shy away from. It was sales and marketing that we needed to really buckle down. I got a “Sales for Dummies” book because I knew how to write software and I knew how to write automated phone IVR systems and I knew how to code database schemas and all that stuff, but it turns out that the key insight to selling a restaurant which is just walk in and say, “I’ve got a product that you should buy” and just believe it. That’s actually . . . “Sales for Dummies,” it’s a great book.

Andrew: Tell me more about what you learned from “Sales for Dummies” book that was actually practically useful.

Michael: The most practical useful tip was walk in through the kitchen. Go into the alley and walk in through the back door, because you want to talk to the owner and they’re in the back, they’re not in the front. So that was it. That was the most valuable tip.

Andrew: And that was in the actual book, go to the back?

Michael: Yeah, it actually was. That wasn’t one of their overall arching themes, but it was in one of the little funny example stories. I was like, “Hey, I’m going to try that.” And it worked.

Andrew: Mike, I’m sorry to obsess over the early days so much, but I’m fascinated by details. You would actually walk into stores, door-to-door, the guy who’s now running a multi-million dollar company, and we’ll get to how big the company is today, but you were walking door-to-door coming up with clever ways to just meet the owner and selling him on using your system?

Michael: Yeah. I was doing that. I had a background as a software developer, and I very quickly realized why salespeople promise things that don’t exist in the system because you will do anything to make that sale when you’re going to door-to-door. And I’d be like, “Oh my God, I’m going to be up 20 hours tonight writing the software that I just sold to this guy.” But I was so close to the customer in those early days that in some ways I miss it. I actually still go on sales calls just to make sure that I understand what the things are that are valuable to the restaurateur.

Andrew: Do you happen remember one of the things that you promised when you were talking to a restaurateur?

Michael: Yeah. Let’s see. One of the ones was, “Yeah, it’d be really great if I could email the people who have used the website before. Or you could email them for me whenever I have a coupon.” I’d be like, “Oh yeah. We have that system. No problem. I could do that.” So I had basically just promised a complete CRM solution that didn’t exist. And then two days later it did.

Andrew: And is that something that’s still useful today?

Michael: Yeah. Sure. Certainly within being respectful of not spamming consumers and things like that. But it’s definitely something that is valuable. And people want to get coupons from their local restaurants. Yeah. For sure.

Andrew: I’d want that. If I was getting something delivered . . . I tend to get the same four things delivered at home. If they had some kind of deal, I might pick one over the other three because of that deal.

Michael: So what four things do you like getting at home? Like what cuisines do you like?

Andrew: Do you know what my thing is right now? Chopped salad. It’s this local place, they will cut your salad, they will put it in a wrap for you or in a bowl. I didn’t have much salad when I was in Argentina. Now that I’m in the US, I’m trying to get as many vegetables as I can in me.

Michael: Okay. So go on GrubHub and look for restaurants that deliver salads to your address, and you’ll see a couple of other options as well. Because even though you might go to the same cuisine or the same place, it’s worth taking some variety too. So it’s definitely worth checking out some of those other choices.

Andrew: Here’s what else I discovered through you that I didn’t know existed. Well, first of all, right now I’m on the phone with you and I see Shawarma Spot. I love Shawarma Spot and here’s what it says in the description. “Healthy, salads.” Where was that? Afghan Kebab House. Man, when I lived in Manhattan I used to love having Afghan Kebab House deliver to me. I didn’t realize they had it outside of Manhattan. I’ve got to put this away page away. I’ve got to go back to one of your facts pages.

Michael: I’ve been there. I have placed . . . this afternoon was order number 401for me. So I’ve placed a lot of delivery orders through my own website. And I think that’s really important that it was useful for me on day one and it’s useful for me on day 2,000. I think that’s key.

Andrew: I’ve got to make sure that I don’t come across in this interview as a fan boy. Clearly I am a fan of what you’re doing. I love the food that you’ve got lined up there, but I want to make sure this interview’s useful. So let’s see what else you did. You walked door-to-door and you learned about what your customer wanted. What else am I curious about? I’m curious about now that I see how you got all those restaurants online, how do you get the customers to come in and buy from those restaurants?

Michael: So one of the big things for us is word of mouth. People love the website and then they tell their friends, right? So from a business perspective, you guys know Net Promoter Score, right? I’m sure you’ve heard of this.

Andrew: It happened to come up in a previous interview. I interviewed a company that actually helps companies generate their Net Promoter Score and manage the results that they get, the feedback. Net Promoter Score is “how likely are you to recommend us to your friend.” The higher, the number the better you’re doing as a business.

Michael: Our number comes in around 62%, which is very high. It’s not world-class yet. Apple and Google are in the probably in the low 70s, so we have a few more points to go before we’re in that realm. But that speaks to why people love the website and then they tell their friends about it. Almost our entire marketing plan is based around building a great product that people love that they’ll use the first time, they’ll use every time after that, and then they’ll tell their friends to use. That sums up our entire marketing plan really.

Andrew: There wasn’t anything in place to systemize that. In other words, I’m on and I see always, “If you get your friends to buy from us, we’ll give you ten bucks.” And that seems like one of their marketing techniques, one of their big ones. Did you have anything like that?

Michael: I’m not against the idea. But just internally for me, if you’re a buddy of mine and I can get you fed well, that’s enough of a reward for me that I really want to make that recommendation. So instead of putting a lot of time towards developing a referral fee program, I guess why pay somebody . . . would you really want to be paid for something that you would rather help a friend out for free? We’re not against the idea, but it just seems like every time we thought about it we’re like, “Well, what about this extra feature that makes the ordering process better?” Instead of worrying about paying people off? And I think that that program works for a lot of companies, but we’ve just been much more focused on making sure the experience is a good experience.

Andrew: But word of mouth takes a while to build up. You’re in the early days trying to get people in just to validate your idea. Did you do anything to juice it? Did you do anything like hand out coupons, walk out in the street and talk to people? Anything?

Michael: I’ve got to tell you, we have tried everything. One of the things that’s been really valuable for us is we do a lot of transit advertising. The one type of advertising that we do do is transit, because when you’re going home from work and you’re hungry, if you see an ad on the bus or the train, you’re more likely to place an order. And now that our iPhone and Android apps are out, that happens on the way home as opposed to when you get home. So we’re seeing a massive shift from people using the website to people placing an order so that they’re racing it home against the delivery driver. So that’s the one thing that we do do to sort of, in your words, juice the word of mouth equation. It’s how do we make sure? We get people when they’re hungry. So that’s the other thing that we do.

Andrew: Okay. At some point this from being a hobby to a business, and you say on your blog that an advisor helped you make that decision. Can you tell me what he said that persuaded you to change things?

Michael: Yeah. He told me to quit my job. The very first piece of advice that I got that was really valuable was metaphorical hunger is great, but if you actually have to sell or eat ramen noodles you will sell. And especially since I was building a website based around food, I was very, very aware that . . . I quit my job, I made $140 that first month, and man that is good motivation for making a little more the next month. So that was the big piece of advice that I got from an advisor.

Andrew: And the other thing that you decided right from the beginning was that you had to make money day one. Why?

Michael: Again, the issue is if you make a product that someone is willing to pay for because you’ve solved such an important problem for them, you know you’ve made a good product, right? And the better the product, the more someone is willing to pay for it. So in some ways getting customers to pay for my product was a validation that the product was moving in the right direction. And products that made it easier to sell, that provided more value got prioritized over products that didn’t make it easier to sell. So again, it’s not an academic exercise. It’s one’s livelihood and it’s important to make it work. So I do think quitting your job and actually paying yourself and actually making money as soon as day one as you can possibly do it is important.

Andrew: Do you have an example of a direction that you took with your business because the data, meaning the revenue that was coming in, was telling you to go in that direction?

Michael: Yeah. Really early on, I mentioned that we had a subscription fee. And that turned out to be a challenge, because then let’s say I had 100 restaurants on the website. To double the amount of revenue coming to us, I needed to get 200 restaurants on the website, right? So we changed it around to a per order basis and then to double the amount of revenue on the website it was consumers. And so that’s wasn’t a direct sales pitch. That was consumers finding the product valuable, the diners that are hungry. So that was the biggest change from the napkin, and that was driven entirely by how we wanted to scale the business.

Andrew: I see. What about this? I was interviewing Noah Kagan, and he said that he worked at Facebook in the early days and Mark Zuckerberg kept telling him over and over, “Our goal is to grow the audience. That if we grow the audience the revenue will come in afterwards.” And it seems to have worked out well for Facebook. There must have been a part of you that said, “Let’s find revenue, grow the audience and then later on we’ll find a way to make money if we are the leader in this space.” Why did you decide against that? Why’d you push out on that idea?

Michael: We bootstrapped for three years before we took funding. And I didn’t want to not want to take a salary for three years. So I think there’s a set of people who are really good at building platforms and really good at grabbing market share and then building a revenue channel afterwards. It’s just sort of not the way I’m built. I wanted to build a product that people were willing to pay for. And that would drive our decisions about how we were going to really grow the business because we were providing value to people. So I think it really just comes down to that’s not in my makeup about what drives me to run a business.

Andrew: Do you think running out of Chicago maybe changed the way that you think about it? That maybe if you were in Silicon Valley, maybe even if you were in New York, you might have thought more along the venture capital route?

Michael: Maybe I would have. I don’t think so. One of the things that I decided pretty early on, several years before I even started GrubHub, was I have a unique skill and that’s the ability to create a product and something of value that someone will pay for. And there’s no cost of capital for me to do that because of my skill set. And going all the way back to the Industrial Revolution, that probably hasn’t happened in history before. So just thinking about where entrepreneurship started 100 years ago and what it is today, I was like, “Hey, I can do this without investment. I want to see if I can.” I’m challenged by that. And actually the reason that I ended up getting venture capital, we were already growing at 100% per year. We were already profitable. And I said, “I think taking VC money is going to really raise the bar in terms of what I have to do. My partner and I, we’d have to have a board of directors, we’d have to hit revenue goals, they would push us to grow faster, it would focus us more towards the end goal of the business. And so the decision to take venture capital wasn’t one necessarily about increasing resources to speed up growth, but more it was about personal growth. And so I think where I was living probably wouldn’t have made much of a difference. And there’s a lot of VC funded businesses coming out of Chicago these days. Groupon, Sitter City, a couple others. So there’s definitely cash here to be here. So I don’t think that that necessarily would have made the difference.

Andrew: Okay. You mentioned end goal. When you took the money, was it because you had an end goal of selling out at some point?

Michael: The day before the cash hit the bank, I probably didn’t have that goal. But it’s one of the explicit expectations that a venture capitalist makes sure that you agree to when you take cash. And you’re no longer the head of the company. Or you’re the head of the company but it’s no longer just yours. So that’s an agreement that we made, so that became one of the goals of the business once we took the money. Now the reality is the best way to do that is just to continue growing the business. You don’t focus on it too much.

Andrew: How profitable was your business before you took on investment?

Michael: I think profitable is ebullient. It was profitable. Are you talking about what were the margins and stuff like that . . .

Andrew: No. How much profit? By the way, I like the way you described that. It’s not binary. There’s a lot of profit, there’s a little bit of profit. Do you know about how much money in profit you were bringing into the company?

Michael: It wasn’t huge. The profit, if I were to divide that between myself and . . . we do profit sharing among the employees, I think it still would have represented an opportunity cost versus the other jobs that we could have in our industries. So at the time we took financing, it was probably not long-term sustainable profitability. But we weren’t not going to be able to make payroll because we knew we were making more money than we were spending.

Andrew: I see. Enough to make payroll, enough to give yourself a nice salary, but not a competitive salary with what you would have gotten if you took on a job somewhere else.

Michael: Yeah. That’s correct. But we might have gotten there after a year of growth with just reinvesting that profit anyway. But we weren’t there at the time we took investment.

Andrew: So you would have gotten there, but you weren’t there at the time. This was 2007. You raised is $1.1 million I read. Do I have that right?

Michael: That’s right.

Andrew: How did things change afterwards?

Michael: So we had a board of directors. It was no longer a board of advisors, it was a board of directors and we were beholden to them. Now my partner and I were two of the directors, so it was sort of weird to have that dual responsibility as well. I think it sharpened our focus and made us more aware that we needed to start hiring. So the biggest tactical difference . . . sorry, I’m taking a little while to get around this answer. The biggest tactical difference was it made us realize we needed to hire people better than ourselves to do everything we were doing currently. So we had to hire a better marketing person. We had to hire better developers than ourselves. We had to hire better sales. So that was the biggest tactical difference from both that first round of funding and then that accelerated with the second round of funding we took in 2008.

Andrew: That makes me wonder about the first people who you hired. You said in one of your blog posts you couldn’t give them options. They wanted to get paid. They needed to take something to the grocery store and be able to buy food with it. In the early days when you were first hiring your people, how could you afford to pay them?

Michael: We started making money on day one, so we made enough money to pay them. That’s what it was.

Andrew: So it just from the profit that was coming in in the business?

Michael: Yeah. We were reinvesting a bunch of it, or all of it essentially. And that mostly went to salaries early on.

Andrew: By the way, if anyone’s listening to me, they’ve got to check out your blog. What is that personal blog?

Michael: It’s just

Andrew: I thought it was on a different URL. Anyway, you did a series . . .

Michael: I changed it two days ago.

Andrew: He did a series on there where you broke down progress in the business based on how many employees you had and other milestones. And it was just so interesting to see you take down the business into those simple terms and it wasn’t a long, detailed blog post. It was just, here’s what happened when I hired the first person. And here’s what I learned from that. And here’s what happened when we moved into a new office and here’s what I learned along the way there. Really great series. What else do I have? So you paid your people back then from profits. You now have new money coming in and you have an options pool. How do you go about hiring new people?

Michael: And to be clear. So first, we did give options early on, it just wasn’t a primary motivating factor. So we did that from day one for sure.

Andrew: That’s not easy to do either. You had to pay a lawyer to put together an options plan for you to give to your people. And that also came out of profits? That came out of revenue?

Michael: In retrospect, I should have paid a lawyer to create an option plan to give options to employees. It turned to be much more of a handshake and a promise. So I don’t know if that’s actually options. But I made those promises in good faith, and those did turn into real options as we got the funds to do that. And that all happened with the first financing actually.

Andrew: So you get money. Hiring people’s different. How?

Michael: I don’t think hiring people was different. I think the actual process of getting people in the door still came down to get the very, very best people you can and don’t settle. And that was true whether it was the first employee or the person I interviewed 20 minutes ago. And so the actual process involves just not ever compromising on getting great people. What did change was the urgency and speed with which that had to happen and how much of a mix of my time that became. Especially during those first two financings, that was 75% of my time was hiring

Andrew: I see. How much is it now?

Michael: Now, hiring is probably 25% of my time, about a quarter of my time. Hiring and also just making sure that the new employees that we have are really understanding what’s going on and they’re plugged in and everyone knows who they are. And so it’s both hiring and that first critical three months to make sure people become part of a bigger whole.

Andrew: What did you need to add more people for? What was the one area that you needed more people in the most?

Michael: Yes. We needed everything. Well, sales, because “Sales for Dummies” will get you so far. But it turns out it’s better to have a professional sales person. And marketing, I can do SEO work but it’s better to have somebody who’s actually good at it. Finance, I couldn’t even do remedial finance work and so hiring a good finance person was really key. And then software development. At some point I wanted to stop doing 100 hours a week and answering all the on-call pages and have somebody else doing that as well. And then customer service, customer service was actually one of our first hires as well.

Andrew: What kind of customer service were you doing?

Michael: Just that same mantra I mentioned before, which is the diner must always be fed no matter what. You must get your food. And so all of our customer service is always focused around that.

Andrew: I did an interview just a couple of hours before I interviewed you, and the entrepreneur who I talked to said that he had trouble finding the right salespeople. That was the hardest position to place because salespeople, even bad salespeople, are good at convincing you that they’re great. So you need to get beyond their salesmanship to figure out is there substance there. How did you find the right salespeople? How’d you get beyond that?

Michael: It’s funny. It brings back some memories. So at that time, I was walking into restaurants so I would invite someone out to coffee for a conversation. And about halfway through, I’d be like, “Hey, why don’t we go sell a restaurant?” And I would just walk into a restaurant with them and I’d say, “Go for it. Give them your pitch. You’ve heard 30 minutes of the business.” That terrified about 80% of the candidates, and I knew they weren’t the right ones. The right ones just went for it. And they didn’t do a great job in the sense that they didn’t understand all the material, but they went for it. And so I figured, “All right. They have the gumption to do this job. Now I just need to make sure they understand the product well enough to sell it.” That’s what I did early on. It was a very time intensive process. And I think that a lot of the people who went through that process probably would not agree that it was great.

Andrew: What did they need to know about the product in order to sell it well?

Michael: Here’s the secret of sales — almost nothing. And that’s true with almost every product. It’s a lot more about being confident and asking for the close than it is any other skill. So I would go in and say, “Okay, we’re going to go do a sale. Here’s a one sheet that you can use. You need to know that it provides this value and this value. See if you can convince them to the point where they’re interested, and then I’ll take it over from there. If there’s questions you don’t understand, I’ll make sure that I field those questions.”

Andrew: I see. So there’s just confidence. And how much money were you giving the restaurateur off of each order?

Michael: The restaurateurs, they tell us what they want to pay. So it varies greatly across all of our restaurants. So they just tell us what an order’s worth to them and then we send them orders.

Andrew: Oh, wow. It comes down to what? Is it just a couple of bucks per person?

Michael: Yeah. It’s not much. It’s a pretty nominal fee. We do entirely based on volume. We want to send restaurants so much volume that they have to hire people to continue to service this new flood of food. Like I mentioned before the interview started, we did about $70 million in food sales this year. Some of our top restaurants are making close to a million bucks a year just in food. So they’re very happy with us, and they’re totally fine paying just a little bit per order and they tell us what that’s going to be. We don’t necessarily enforce those rules to them. It’s very, very profitable for the restaurants.

Andrew: There are restaurants that are getting about a million dollars in business from

Michael: Yes. Several.

Andrew: What kind of restaurants are we talking about? Domino’s? That kind of thing?

Michael: No. They’re all independent operator. We don’t have any chains larger than ten restaurants on the website.

Andrew: So a guy like this Kebob House, or someone on their level, could end up getting a million dollars in business from

Michael: Yep. They could.

Andrew: This is the worst interview for me to do. My mouth is watering as I do this interview. I moved away from your page but I’m still on here. I found something that was inspiring in the story that you told about your early days that you would sell by day and improve at night. Can you describe what a typical day was like?

Michael: One of the things that kept me sane was you could only talk to a restaurateur from 1:00 to 5:00 p.m. You don’t go in at lunch rush and you don’t go in at dinner rush. A typical day would be I’d wake up at like 11:00 and then I’d go and sell restaurants between 1:00 and 5:00, sort of get my leads figured out, figure out what it’s going to be, maybe I’d make a few phone calls and set up some appointments. Then I’d just hit the streets from 1:00 to 5:00, come home, order some food. Actually a lot of times I wasn’t hungry because the restaurateurs would feed me a lot while I was selling them, even the ones who didn’t sign up. They must have seen some hunger in my eye or something like that because they definitely would feed me. Then right around 6:00 or 7:00, I’d start writing software and I’d go to bed around 1:00. Start the next day again. Seven days a week for probably about six months like that.

Andrew: Wow. Right now I can imagine my audience sitting there listening and saying, “This guy just has it all made. Everything worked out perfectly.” We’ve got to humanize this story. Do you have a big setback that you can talk about how you overcame?

Michael: One of the things I would go back and change is a lot of people had ideas for me, both for the business and the website. They’d say, “What about adding this feature? What about adding that feature? What about changing the business model that way?” And people had those ideas when I asked for them and they had those ideas when I didn’t ask for them. I mean, everybody had input. I was not gracious at accepting all of that input. I think I was a little argumentative, and I wasn’t as thankful for people who really spent the time and effort to help me make my business better. I wasn’t as appreciative of them as I should have been. So I’d say that one of the things that I did wrong was that. One of my advisors basically knocked me upside the head and told me to knock it off. That was a good message to hear and I was glad I was able to hear that.

And then in terms of more technical business things that we did, I think that doing user focused design earlier rather than later is something that you should do. And that probably fits into the same character flaw of not asking for advice as much as I should have, or not accepting it. Now I understand that user centric design and user interface design is one of the most important parts of our business. Early on it was whatever I thought worked I was going to throw up there because I was the entrepreneur.

Andrew: Give me an example. What did you throw up that wasn’t user centric?

Michael: It was like a link farm at first. It was all SEO optimized, and it was ugly. I didn’t ask people, “Will this feature help?” I just said, “This feature will probably help.” And so that would be links to restaurants or there would be other companies in the area that wanted to advertise on our website and I would add those on as advertisements, which didn’t really add any value to the end user and probably was a waste of my time. So I guess both of those things are in the same vein — listen to your users and listen to people who actually take the time to give you advice. Even if you disagree with them, you should accept their advice graciously, right, because they’re doing you a favor.

Andrew: It is sometimes hard to hear that. I ask for feedback all the time in my interviews. I want it. And sometimes when I get an email where someone saying do something a little bit differently, I get a little upset. “Come on, this is what I do.”

Michael: Okay. I’ll try to get angry when you ask me that question.

Andrew: Like what? Beyond being more gracious . . . and I hope that I’m being gracious with every single person’s who given me feedback. Guys, I want more feedback. But beyond that, was there something that you wish you would have listened to that you would have done differently? Some advice that someone gave you that you turned away back then?

Michael: The one thing that I tried to do too much of, since I was writing my own software, I would just try to do everything. So a couple people told me I needed to focus more. If you’re going to be a website that shows delivery and does online ordering, then focus on those things. Don’t focus on the ancillary advertising features for partners and things like that. So it was about focus. The scarcest resource that an entrepreneur has is their own time. It’s not the dollars in the bank account, and so prioritizing is absolutely critical.

Andrew: How did you go to $70 million? Is there a big milestone, a big step that you took, one thing that you did that really jumpstarted all that revenue?

Michael: I’ve mentioned this before to some people. Making that first dollar was really hard because we had to make a product that was worth people paying for. The step that I took that really made the big difference was the second dollar from an original customer. Getting repeat business and thinking about how you get both restaurants and diners to come back and repeat purchase was the thing that really made it not just a business but a super charged engine for growth. And that’s just because when a person comes back to the website and purchases a second food item or purchases another dinner from us, it’s just as much revenue as the first one was. But getting that person to do that is a fraction of the cost from either marketing or product or whatever perspective you’re thinking of. So the real super charger for growing a business is getting the repeat business model figured out.

Andrew: How’d you do that?

Michael: I mentioned before that our entire marketing strategy is how do we get people to place their first order, place another order after that, and then tell their friends to place an order. That was not the original strategy. That came out of a lot of learning that this repeat business is really critical. So that’s how we did it. We came to that marketing idea and away from how can we do partnerships with every person under the sun to see if we can drive traffic to the website.

Andrew: So how do you get somebody who’s ordered once to remember that it was GrubHub that they ordered from and to get them to come back and order from you?

Michael: Provide as much variety and convenience as possible and make sure they get fed.

Andrew: Is there any other tool? Like maybe emailing them a week later or emailing them afterwards?

Michael: So imagine somebody who didn’t get fed or didn’t have a lot of variety, email is not very helpful. No. It’s make sure they get fed, make sure they’re happy, and make sure they have variety. Then all of those other tactics to remind them that that was their experience, they’re valuable certainly. But they are not core to what we do.

Andrew: I see. So there’s a little bit that you do, but there’s not any one of those tools that you can, “Andrew. Once we mastered email we were on top.”

Michael: No. I mean think about there are 50,000 businesses out there that have mastered email, but there aren’t a lot that have mastered getting people fed. That is the one thing that we do really well is we get hungry people their food. In Jim Collins, “Good to Great,” he talks about the one thing that you do better than everything else. We get hungry people fed, and so that’s worked out really well for us.

Andrew: Finally, I’m looking at my notes to see if I covered everything. There are two things I didn’t cover. First is why’d you raise in 2009 another $2 million?

Michael: We had grown to a pretty good level, and we wanted to accelerate the number of cities that we were expanding into through 2009 and 2010. So we did the raise specifically for that. So now we’re in 13 cities. At the time we did the raise we were in four.

Wearne: Do you know what? I realize you’ve said a few times here, “SEO.” How big a part did that play in marketing?

Michael: The nice thing about SEO is you build it into the product. So I’ve been talking about how the product is the king in terms of getting people to the website. One does that with an eye towards SEO, for sure. So that’s one of the big inbound funnels for how people find out about us in the first place.

Andrew: I see. So every one of these restaurant pages that are one your site are optimized for search engines, designed to make it easy for people who are looking for them and Google to find them.

Michael: They’re designed with exactly two principles in mind. How do we get a person who’s hungry to be able to place their order as easy as possible? And then how do we also, secondarily, make that attractive to Google? It’s really important to design websites for people first and machines second. But you have to do both. You don’t forget machines in that equation. It is optimized for search engines, but that’s secondary to the people using the website.

Andrew: I’m looking at one of the pages right now again. The Kebab House. What about this page is designed for machines? I could see where it’s designed for humans like me but . . .

Michael: Every item of food on that restaurant is showing up as text, right? So who else has that? That’s one big piece right there. It’s just the nature of the business that we built. And beyond that, very simple things like not having duplicate content and having unique URLs for that restaurant and things like that.

Andrew: Yeah, you’re right. Why do you think restaurant pages are all in Flash and are so bad to use on phones and so bad for highlighting and copying.

Michael: I love restaurateurs and they are so good at making food. Their nephew might be okay at writing web pages. So maybe that’s the best answer I can give to your question.

Andrew: The last note that I have here is advice for other entrepreneurs and other people who are building online businesses

Michael: If it’s an online business, the key is figuring out the repeat usage. If your business only attracts somebody to pay once for a service, make something of value that they’re willing to pay for multiple times because it’s just not worth the uphill battle of getting new users every time you need revenue.

Andrew: What a great place to leave it. The website as everybody heard me say over and over again . . . it’s 4:52 right here and all I had was a little salad today. That’s why I keep harping on the food of this website. And also it’s a great place to order food. I actually didn’t even know you guys were in town until I went to a tech cocktail party, you guys gave out these $10 coupons. I showed it to my wife. I said, “I don’t even want the $10. It’s a new company. I don’t want to take their $10 from them. But, man, I didn’t know GrubHub was in town.”

Michael: Feel free to use that ten bucks. It’s all you.

Andrew: It’s actually sitting there in the kitchen. I always feel bad using them, but it’s easy when it’s online. I don’t have to present it to a restaurateur. Anyway guys, check out And check out Michael, thanks for doing the interview. It’s great to meet you.

Michael: Thank you very much.

Andrew: Thank you all for watching. And Anthony, thanks for hooking up this interview. Bye.

This transcript brought to you by

Who should we feature on Mixergy? Let us know who you think would make a great interviewee.