Andrew: Hey there freedom fighters, my name is Andrew Warner. I’m the founder of Mixergy.com. That fist over here means its home of the ambitious upstart, that’s what’s it about here. Hey, have you ever tried selling something to customers who just couldn’t understand what you’re creating until they finally bought and used what you’ve created? Well, how do you do that? Jeffery Zwelling is a Mixergy fan who says that happened to him. He’s founder of Convertro, business intelligence for marketing professionals. When he tried to tell marketers how his company could help increase their sales, first, they didn’t understand how it worked or how it helped them. Then they didn’t believe him and some even thought he was selling snake oil. This is the story of how he won customers over and how he got Convertro to be big to help customers and how he sold it. This interview is sponsored by Toptal, a new sponsor for me. If you’re having trouble finding a developer that’s the right fit and the ideal fit for your company check out toptal.com. Toptal is the company of developers, you tell them the kind of developers you’re looking for. They will search through their network for the right fit. I know because they’ve done that for me, we just hired someone through them today. They’ll test the candidates for you, they did that for me too and they’ll only introduce you to the perfect person or 2 people and you pick the one you want to work with. You can hire them full time, you can hire them part time or if you already have a team of developers, who just needs some extra hands for few hours a week you can do that with them too. Check out toptal.com, my sponsor and thank them for sponsoring. Jeffrey, Welcome.
Jeffrey: Hey, thanks. It’s great to be here. Big fan of the show and I’ve enjoyed many of the episodes.
Andrew: You know, I really appreciate you saying that. You said
that before we’ve started and that you were listening back before anyone even knew what Mixergy was, which is really cool. What did you get out of listening to Mixergy?
Jeffrey: In particular, I find the detailed questions that you ask people, you don’t let them necessarily just promote their business, which, of course, is what they want to do always. But you really dig into the nitty-gritty of what people do to be successful, and I found some of those things to be incredibly insightful.
Andrew: Oh, thanks. I appreciate you’re listening and saying that. And one of the things I learned about you in researching you is that you were working at a company that you called YLighting, and you have a problem that eventually led to the company that we’re talking about here today to Convertro. What was that issue, what were you trying to solve for yourself?
Jeffrey: So in the case of YLighting we sold very expensive lighting and furniture, average order was $800, life time value was $2,400. What that meant in terms of marketing optimization is that by the time someone went to make a purchase decision, to buy that $8,000 light, they’d spent a fair amount of time doing research on determining which is the right light, as well as where they were going to buy it from. So all of the traditional tools that people used to measure the marketing effect, just didn’t work for us. Specifically, at the point of that they were doing a purchase, when we would apply a conversion to that marketing it was all searches for our brand YLighting. And YLighting it’s well known among a specific niche wasn’t well enough to assume that people have discovered us simply by that brand. And it was highlighted in fact, at a time when, Blooming Dales placed a $40,000 order on our site at for retail. And I as I often did in those days; I would call up a customer after their order was delivered and say, “So tell me why did you buy from us?” And I got these stories in Blooming Dales in particular, where she told me, “Oh, a year ago our architect, recommended this light. And I went online and I ended up on YLighting never intending to buy from you. We were going to buy from our purchasing department but he was an idiot. Then I signed up for the newsletter and I had great customer service experience. In a year later they buy a $40,000 for retail. Now if I want to understand how to get more Blooming Dales, the problem that I had was that, if I looked in Google analytics, it said, they searched for YLighting because that’s what they did once they had that PO approved. But the reality was it was the search for the particular light that the architect had recommended that led her and introduces her to our brand. And so I needed to make sure that I was reinvesting in the marketing channels that were bringing me customers’ top of the funnel, not the ones that use to navigate to our site. Now, I didn’t find anything out there that actually did this, so when I was running YLighting, my only solution was to build it for myself. And when I turned that technology on that I built for myself, it gave me an incredible competitive advantage over all my competitors because I could then buy marketing channels that everybody else didn’t even see as valuable. They were cheaper and they drove incremental sales and incremental customers. So when I sold YLighting to private equity…
Andrew: If you don’t mind, let me pause there for a moment.
Jeffrey: Yeah, sure.
Andrew: You built it through yourself internally, how did it solve that problem of, let’s call the Blooming Dales question? Where she didn’t, the buyer didn’t find you via Google search, found you somewhere else. How do you track that?
Jeffrey: So, it’s a good question, typical of your interview style, which I appreciate. The short answer is, I just wanted to make sure that every time someone came to the website, I was recording into a database that all of the marketing had brought them, in the event that they subsequently purchase something, I have that historical record. That means you have to record and track all of the visits irrespective of whether they convert or not, something that wasn’t understood or easily done back when I built this. The details of it, to be completely transparent, were that we used techniques that are no longer privacy compliant. And so when we built Convertro, knowing that we were going to sell it to a bigger brands and that being compliant with modern privacy laws was critical we had to deploy different and more advanced technology, to be able to tie together the multiple cookies that an individual user was represented by.
Andrew: I hate to push on this because I want to keep the friendship, but I also want to keep your respect and my audience’s respect, when you say not complaint, what did you do?
Jeffrey: Well, there were all these techniques that subsequently the Europeans decided that they didn’t think they were good and clean ways of tracking people. But there were a series of technologies using flash stored object, HTML-5, device IDs, none of which Convertro ever engaged in but back in the day when we’re doing it for YLighting, were the easiest way to track people across multiple sessions.
Andrew: I see, okay. And so, once they finally did do something that gave you an email address you could connect it back to that ID you had over multiple sessions.
Jeffrey: No, you don’t need an email. Nope, you don’t need an email address. For instance, there’s all kinds of stuff you can find. I love your interview style but you kind of [baffle me]. I want to make it absolutely clear, we have never done this at Convertro nor would we ever do it. Back in the day, you could do some research on the net. There were companies that were deploying these objects so one of them was a flash stored object, creates a unique id when its instantiated and it’s pulled up that is separate from the cookies. So if you have two cookies, you can go and investigate that flash stored object, and there is a unique id in there so you can understand that this person, while to Google analytics they look to like two separate people, but on their computer they’re the same person.
Andrew: Okay.
Jeffrey: So those are variances of the same techniques which are widely known among the hacking community.
Andrew: I see. How did you know about them?
Jeffrey: I’ve been coding since I was a kid, I haven’t personally coded in a long time, but there were smart people in my teams always. Those people who always knew the leading technique.
Andrew: What is your process for finding those? I remember actually, years and years ago, back frankly when porn was at cutting edge of technology, and they were really pushing the envelope I would go to these message boards that the porn site webmasters would go to and talk about how they created pop-ups. What they did and so on, and that gave me great ideas that people in our world never knew. Today, I have a friend who works with me who takes ideas from online marketing, from how online marketers are especially good at using emails, especially good at convincing people to tell their friends and all that and then he takes it out to brands, like a drink. Where did you find this stuff? Where did you find the bleeding edge ideas like this?
Jeffrey: So I always had been a fan of reading Reddit, flashdoc, I certainly plowed through GitHub. But a lot of ideas, the more technical ideas, many ways came out of my head. I’ve always been someone who could visualize software and how you could put stuff together. My first company incidentally was 3D gaming, making PlayStation and Nintendo games and to be excellent at that you essentially have to be an awesome hacker. You have to know how to exploit every vulnerability that the PlayStation has in order to make the best explosions, the fastest frame rate, the most interesting stuff. So I guess, the answer is really imagination, curiosity and fair amount of research.
Andrew: I see. I see how your brand works and I see how your past experience would have allowed you to do this. The other thing I noticed in your history was that you told April Dykman, our pre-interviewer, that you didn’t have a lemonade stand so much, when you were growing up. Do you remember what it was what that you did?
Jeffrey: I did a lot. So, when I was in high school I had one of my first real software companies and I identified a problem. The problem was that manufacturers had to update their price list when they were heavily dependent on the changing commodity prices and to get that means to update those lists quickly. I fell into the problem. I was a kid, it was a summer job and I was actually assigned to work for the consultant who was hired but who was incapable of actually delivering anything. So I went to the CEO of that manufacturing company and I said, “I think I can do this.” And he offered me $5,000 upon completion. I later learned that he was paying the consultant $50,000, this is in mid early 80’s. So I built what he needed and then instead of going to find another consulting project, I started looking for other companies that looked just like him and offering at turnkey for $5,000. So by the time I graduated, I was one of the wealthiest kids in the school. I think I did a tax return in a range of $50, 000, my senior year.
Andrew: Oh, wow.
Jeffrey: Which was pretty amazing.
Andrew: Wow.
Jeffrey: $85.
Andrew: You told April also, that you bought a Honda Prelude, with what you made from that first deal. But that couldn’t be the only thing. How could you celebrate when you’re $50, 000 as a guy that young?
Jeffrey: Funny enough I invested in stocks. So I had a dial-up modem and it dialed up to some of the first stock trader, which I can’t remember the name. I still see them in the deep files, I actually have receipts. But I would buy options. One of my first trades ever was literally, early Cisco. And so I just fundamentally believed in the future of networks and not the internets. I think I put 50 or $60, 000 to work there and made half a million, probably at college.
Andrew: So by the time you left college, you already made half a million dollars?
Jeffrey: Yeah.
Andrew: Wow. And you weren’t using this to buy anything but a Honda Prelude, you just kept [parlaying] it.
Jeffrey: I always, it’s funny. I’ve always felt richer than I imagined even when it was $50,000 when I sold Convertro, the only thing I bought was . . . I bought a European like professional espresso machine that was fully plumb, including having the people plumb it and the counters. And I think a lot of people when they sell a company for more than a a hundred million bucks, they tend to buy little more extravagant things but I’ve always felt I have everything I want. I typically, I like travel, I travel well. And I don’t care too much for cars. I don’t want to have multiple houses. I mean I’ve always been someone who is much more motivated by solving problems, working with smart people. I like the success. I like money as a measurement of success and customer adoption, but money per say, like I always said, I’ve always felt like I had more than I want. I know for instance, that I’m getting a massage chair for Hanukah tonight. It’s supposedly a surprise from my 5 and 7 year old. Of course they can’t afford to do that without my credit card so I saw the purchase go through. But yeah, my needs are always, I always thought I had everything. New computers, I always like computers.
Andrew: That seems to be it. So just the need to solve the puzzle, that’s enough of a motivator. I don’t know who would do it for me. I think I need more personal validation. Maybe I’m broken somewhere in my past. Maybe my parents gave enough validation, I look for business to give me validation. I look for my audience to give me validation. Maybe you didn’t have that, did you?
Jeffrey: When you said that I was thinking about audience. I think I’ve always skewed opportunities, to speak publicly about my success, I happen to be a particular fan of what you’re doing, and its one the reasons why I’ve responded. Multiple times, PR firms want to use me as a bait to get more publicity for companies that I’ve found. I think the audience thing somehow doesn’t fundamentally appeal to me. But I definitely get great pleasure knowing that millions of people are using EchoSign. That the videos games that I made had millions of consumers. That Convertro, as another example, the audience probably doesn’t know what it is but I’m happy to explain. But Convertro like on Black Friday was interacting with 300 million consumers flawlessly. I think I was just one of those people just really like to solve puzzles and feel like I got it right, nd sometimes audience and customers are the way that I measured whether it was right. I also when it’s worth it, enjoy winning all the contest, sell based contests. So winning is very, very important for me and losing is even more important [as always]. But the validation is not exactly the way I’m wired. I’ve known Mark Pincus of Zynga for a long time, and he had quite a bit of success before Zynga and I was always fascinated by the fact that he . . . it wasn’t until after the public affirmation that came from Zynga that he felt like he had made it. And I thought it was ironic, because as you probably know, Pincus and Reed Hoffman bought the patterns to social networking which ended up being licensed to everyone; Facebook, LinkedIn’s. So he made a quite a bit of money by being very prescient about the future, because it was 6 degree patens. To me that was enough, like being right and solving the puzzle and validating it by people like Facebook and LinkedIn to having to pay your license fee for being smart, that would have been fine. I would have dropped the mic and walked off stage. But Mark, I think Mark mentally didn’t feel personally satisfied until he was ringing the bell on NASDAQ and CNBC. I never had that type of orientation and I’d actually find it sometimes challenging sometimes to step up to it.
Andrew: That’s interesting. Actually, I didn’t think we would get you as an interviewee because you’re not very public. Because we just found you when a virtual assistant saw an article on you on TechCrunch and said, “We should get this guy on.” Alright, so then you had this tool to work for you internally. It was time for you to say, “This is strong enough that I could sell it to outsiders.” You made a sale before you actually started the company or before you created the product. How did you find that first person who bought from you?
Jeffrey: I was asked by VC to consider, to go and meet with this company, sort of a mid-stage start up. For their benefit, I used that and as I often do, as an opportunity to pitch what I had in mind. And when that client, that customer said, “Yeah we would definitely use that.” Then I said, “Alright, I got something.” Its conventional wisdom now, to get customers first, but that’s always what I’ve done. I don’t really, I don’t care if the VC is excited about it. What I want to know is, are our customers excited, how many customers can I address, and how much will they pay for it. And that’s really my analysis. I often think that the things that I’m excited about and other things and lot of other people going to be excited about. But I certainly had, I quashed good ideas simply by trying to find that customer, and not finding them.
Andrew: Do you have an example of that?
Jeffrey: Kind of, I think very recently.. Let me try to think of it. That one will waste the hour talking. I think I had some, the irony is that one got funded along those themes but I had this idea sort of uber for car mechanics where the car mechanics comes to your house. Now, there is one called my mechanic, I think. But when I went and talked to people, I knew they still pay for it, people expressed their concerns that they didn’t like the guy coming to their house to the quality of the service. And since I couldn’t facilitate even one introduction to a mechanic I backed away from that idea.
Andrew: And it was you personally calling up people and saying, “Do you have any needs for a mechanic? Can I send someone over to your house? Will you pay me and I’ll send someone?”?
Jeffrey: Yes.
Andrew: So interesting, because you mentioned EchoSign, we use EchoSign here at Mixergy. When I hire a new employee and I have someone come in that I need to 1099, I quickly send a template over an EchoSign and they sign it and have it forever. The reason I say this, you’ve created this really impressive company and quite impressive career. Why are you personally calling? Why can’t you hire some young hot shot and say “Here’s my thinking. We have to sell it before we get. Before we get funding. Before we know we have something here. Go try to sell it for me.”
Jeffrey: Fundamentally, I am curious, and I’m interested in people’s reactions to my ideas. I want to be the primary source. I’m certainly, later on in the company for Convertro I hired four MBA’s and had them call, I’m thinking I had them do 75 interviews with the agency decision makers, which is a non-trivial endeavor. But when I’m starting at the ideation stage, aside whether this is the right allocation of time, as well as my relationships and reputation, I really want to hear it directly. And at YLighting, I have a long time business partner Dave Feldman, we’re constantly listening to our customers, talking to our customer service reps, we listen in all the time to sales. Not the problem ones, just to randomly take a temperature of what customers are saying, what kind of problems they’re having. I’ve always want to be able to read emails. I think it just a fundamental curiosity. But I also think, that when people don’t engage in that level of detail, they risk bad decision making. The filter is . . . it can be very distracting. You can certainly have some smart people do a lot of good work for you but there’s nothing like the interactivity that you get from talking to the person directly.
Andrew: Directly with the customer.
Jeffrey: Yeah, exactly. And by the way, I’ve often . . . I enjoy meeting every prospect, for hiring as well, and it’s often near, having sold all my companies at the point where they just about big, usually right before that it becomes impossible to meet everyone. So I often tell anyone we hire after that, I want to have coffee with you. But yeah and I’m able . . . I’ve been very successful at post acquiring whether I’m going at the coffee or I’m coming back in confidence and walking with manager and say, “I got a but that says this one is not going to work out.” But I’m pretty accurate. So I can’t get that by looking at a resume. Can’t get that hearing a synopsis from the hiring manager for me. I literally have to pretty much look him in the eye and listen to them.
Andrew: You know, that’s something that came so many times in my interviews that the entrepreneur with the idea talking to potential customers first. And there’s something to nuance of the conversation that helps you take the product, or in your case the decision not make a product. So you started talking to customers and one of the things they told you was. “This isn’t clear enough, what you built for us.” Right?
Jeffrey: It was really fundamental challenge which we experienced with Convertro. The original vision of the company was that we’re going to take away data of what was working, in terms of observing the conversions and non-conversions and then used that too programmatically by more media against what we’ve learn from that database. Already in a close loop. Automated path. Now we’ve built that at YLighting and it worked really, really well. And created incredible results. But when we built that for Convertro and we would start calling on people to see if they would like to be our customer, the question they would always ask is, “Let me see your dashboard.” I would say, “Well that’s the beauty there is no dashboard. “What do you mean there’s no dashboard.” All the other guys shop with a dashboard. And we’re like, none of this is an automated platform for buying media smarter. And they’ll make you a lot of money and then we eventually realized after the 30th time, people said, “Where’s your dashboard?” “We can’t buy this without a dashboard.” We had to go build a dashboard. So . . .
Andrew: What did they expect to see in a dashboard?
Jeffrey: All the same things that every analytics type company wants to show. You know they want to see the things that they can in 100 other ways. Who are they coming, when did they come. What did they do?
Andrew: And what you want to do is to say, “Don’t worry about it, we will make sure it’s handled right.”
Jeffrey: I want people to measure us by the performance and the results, not their qualitative appreciation of data. I find it . . . it goes straight to a lot of problems with data driven startups which is too much focus on the dashboard and not enough focus on the insights or the action-ability of the data that comes into the dashboard. And so at Convertro, by the time we released our first real dashboard, we probably had at least eight top engineers working on the data pipeline, another eight working on sort of the algorithms to make clients more money and only two or 3 people on the dashboard. But to a prospect who’s evaluating a solution based on the dashboard there’s no reason to put the guys on the data pipeline, you might as well throw everything in the dashboard. And I think that’s an advantage for people that can make great dashboards that doesn’t have anything under the hood. It also was a disadvantage for us because people were not in the best position to stand and evaluate what it is we really did. How will this make us the money? Because what they were using as the criterion for determining to whether to go forward with us was how pretty our dashboard is. And the dashboard beauty is really irrelevant when you’re coming, when the product is meant to help you buy media better. The key is they’re not going to know this is worthwhile spending money on, unless they actually try it and not just look at the results but act on them. And we knew, and we still know and we see that every day, when clients go all the way to buy media with it, programmatically or manually, they do see incredible results and then that’s what causes them to renew and that’s what causes them to hire us as their next job. But that was the challenging thing. Convertro is that you had to get them to go so deep before they even knew if it was working.
Andrew: So how did you get them to buy it before you had a dashboard to show them, this issue that I meant on top the interview? How did you deal with that?
Jeffrey: I mean the pre-dashboard ones were in many cases were reputational.
Andrew: They say, “He knows online businesses. He knows how to increases sales. He obviously has done really well at YLighting. Let’s buy what he’s going to give us.”
Jeffrey: Yeah, I mean, in the case of True Car, True car came to me and said,” We want you to be, CMO.” And I said, “I just raised money for this startup from [Bessemer] and I have my own thing, I can’t do it.” They said, “No, no. We really want you to meet CMO.” And I said, “This is not going to happen. But I have this great product that could actually be like a CMO proxy.” And so in the early days at Convertro I turn some of the inbound recruiting leads into customers, by promising them I would personally provide them two to four hours of consulting per month as part of the contract.
Andrew: I see in my sense having. I have this conversation with you is the consulting also helped inform you about how they use your software what they were doing internally. So you were learning more about your customers while they paid you to learn more.
Jeffrey: Yeah. I mean, I get a lot of people who come to me and they’re . . . they have some startup and anytime their target customer is a large enterprise I always tell them, “You don’t need to raise money. If you can’t go and sell that large enterprise on being an alpha design partner, etc. then there’s probably no business.”
Andrew: What was the Alpha design partner? How would that work?
Jeffrey: It’s a well-known play. You know, many of the Fortune 100 have innovation labs, they have multimillion dollar budgets for them to participate in startups to bring that startup, creativity and attitude into these large stodgy companies. And so, all you got to do is solution that resonates with a problem that they have and for them 500,000 or 700,000 is peanuts in terms of . . . they’ll that risk all day, they’ll take those risks a year. And all of a sudden not only you have all the funding that’s non-dilutive, but you actually have an engaged customer who’s going to tell you, what it is that they would pay more money for. You can’t beat that when your target customer is a big company. It doesn’t work for SMB, it doesn’t work for consumers, but if you’re selling to the bigger enterprise, it’s really easy.
Andrew: This is the labs that you said inside the enterprise?
Jeffrey: Yeah, like innovation labs at Kraft, and Proctor and Gamble. All the fortune 100 have innovation lab.
Andrew: I had no idea. And so they will hire these startups to solve the problem for the bigger company and work through the lab with the startup to help them solve the problem that the bigger company can’t do on their own, and they do it as a client not as an investor.
Jeffrey: That’s right. I know one thing to note is that having been through a couple cycles and bubbles, I’ll say cycles; one thing that does change is the maturity of the startup needs to be higher as the bubbles get bigger. So you going to see like a data logics, in those innovation labs at the height of return for advertising technology startups now. Whereas, if you’re building a company when there’s not as much attention to the startup world, a smaller company, like its pre-funded, can actually get those contracts. Some of that is reputational too.
Andrew: The other thing that I said at the top of interview was people considered it snake oil, I use that from you pre-interview notes with April to try to get people into the interview and to say something that I know could identify with. But did they really consider what you were selling, what you’re selling to be snake oil? Did it sound that suspicious to them because they didn’t know how it worked?
Jeffrey: Absolutely, I mean working backwards we had solve for the perception of what we were doing but it wasn’t snake oil by partnering with Nielsen who’s like leading media measurement firm in the world. To validate, we end up parting with Stanford, UCLA, Nielsen because people fundamentally believe what . . . it was great. For the early part of Convertro they well. I don’t understand that this is something that Nielsen would be doing. And if Nielsen can’t do it, how could some little startup do this. And I’m like, “No, no, they can.” “We got it all worked out.” For having Nielsen come in and endorse us was a key decision to do that. But the problem . . . there’s a quote that everybody knows in marketing, “50% of my marketing is working, I just don’t know which half.” Convertro at its core, was saying, that, “We actually know something between 80-85 percent of marketing that’s working. And so when you challenge something, so mathematic, as that old quote, there is a lot of disbelief from people who have been in marketing. Someone one time said to me in a panel, they said, “What does a guy in the lighting business know about marketing?” That’s why I said, I know more about lighting than I do about marketing. But the reality is that our mission and what we’ve accomplished, what is considered as the holy grail of marketing, and so showing up with a laptop in a presentation to the biggest vendors in the world. And saying, “Yeah, we’ve solved this thing” It did raise a lot of eye brows.
Andrew: I wonder how you did go pass it. One thing you said is you partner with a bigger company whose reputation was trust. And so then they’re okay, was Nielsen. What else do you do when they don’t understand your product and they don’t trust you? And you actually have a solution because not proven yet.
Jeffrey: So, I definitely think it’s dependent on what you’re doing. In this type of algorithm driven space, you have to come up with a test that you can present the results and they can see with their own eyes that your solution . . .
Andrew: You mean, their company. There’s a test, you mean externally for your company or for how?
Jeffrey: But the first thing is yes. If you can pre-contract, take a client’s data and show them how your algorithm is going to make them more money, that is the best case. The problem is that it’s very hard to do that for certain types of business challenges. This one that’s impossible for it. The reason is that we have to be the primary data collectors, so that the data can be normalized in a way that our machines can eat it. And there’s no way to do this out of the box. In fact the structural dynamic in marketing today, even if I said “I’ll give it away for free and show you its value”, there’s so many people that need to engage in the process of implementation as well as following up on the recommendations, that when you give you give it away for free, they never actually implement, and if they do implement, they don’t actually act so you still don’t achieve that result. So you have to charge them the money from the get go, in order to implement, to get the data and then to process the data and give them the recommendations. And they have to have enough money at stake in order for them to actually act on it. And then they be like, yeah, this actually worked. But in order to pre-sale to get to a contract, I think case studies are critical, references and logos. And we took a strategy of what we called, “The climbing up of the ladder.” And so in the beginning we worked with little startups and then we got to work with Dollar shave clubs, and then that got us to Square. And then Square got us to Eddie Bauer, Eddie Bauer got us to Gap, and then Gap got us to Williams-Sonoma. So, most companies would like to see some companies that they see as equal as or bigger than them and that’s not always possible. But that’s a critical way of overcoming disbelief.
Andrew: I noticed a bunch of Los Angeles companies already mentioned in this conversation. How much does it help you in Southern California?
Jeffrey: It doesn’t help at all.
Andrew: Not at all.
Jeffrey: Not at all, half of my company is in Israel, and that’s because there’s sort of a dearth of top quality talent in Los Angeles. The same reason that I went to Silicon Valley when I graduated, UCLA. It’s so close, that if you’re good and this is what you’ve aspired to do, you’ll get in your Honda Prelude and you drive up north, and you can come up home on weekends and holidays with no problem. So the proximity to Silicon Valley is real challenge to startups in Los Angeles in particular, that are technology driven.
Andrew: Strange. Why didn’t you get in your car and come to San Francisco?
Jeffrey: Why didn’t I do that?
Andrew: Yeah. Why didn’t you get into Bay Area then?
Jeffrey: No, I did, I spent 17 years there. EchoSign was built in a garage in Palo Alto. YLighting was in Palo Alto.
Andrew: And Convertro is not in L.A.?
Jeffrey: Convertro is in L.A.
Andrew: Oh, so . . .
Jeffrey: Coming back to L.A. I grew up in Los Angeles. My parents got sick with cancer I came temporarily and . . . one thing led to another. Our daughter was born, ppreschool was started…
Andrew: I see.
Jeffrey: And life is pretty good here. The one thing I will say is that the weather it’s just unbelievable.
Andrew: I know it. I actually was looking in my inbox to see . . . did we have any conversations and apparently we were at Ryan’s house. I don’t know which . . . Oh, I know which Ryan it must be . . .
Jeffrey: Cause Cast.
Andrew: Cause Cast. Yeah, Cause Cast, Ryan we were having a conversation about cycling, and you’re part of a cycling group that I asked you about and I remember in L.A. you can really plan on going cycling on Friday morning. And know that it’s going to have been and look forward all week for the Sunday ride. And know it’s going to happen. Can’t do that in [inaudible 00:39:18] as well.
Jeffrey: I still do cycling, it’s true. Even on a rainy day in Los Angeles, the sun comes out, the trip, the roads get dry and you can go for a ride so good memory by the way. Sounds like you were aided by Google.
Andrew: Yeah, always.
Jeffrey: But the weather is just incredible, I love the beach. And I think there is starting to be a covert weather people who are migrants from Silicon Valley. Coming down here. And there . . . I think there’s almost soon going to be young people who don’t leave UCLA and Caltech to go up north because there’s enough opportunity here, enough of the eco system and of course great weather and great surfing. By the way, just to give you an interesting anecdote about how crazy this migration is, at Convertro we engaged on campus hiring at UC, UCLA and Caltech. And we identified all the candidates that we were excited about in the process who were all at Caltech. So we spent a lot of time sorting, prioritizing this resume and that resume. We got to 12 resumes that we really wanted to spend some deeper time with. All 12 of them had offers from Google and Facebook and all 12 of them went up north.
Andrew: I remember actually being really excited, there was just this great company in Southern California. They were building an office there and everything was going great and then suddenly they disappear and then they go to San Francisco. And that’s . . .
Jeffrey: By the way, ZipRecruiter is probably going to be one of the first huge successes in Los Angeles…
Andrew: Your current company.
Jeffrey: Yeah, I just started. And they’ve really, definitely attracting engineers out of Google in California to come here. They have a significant number of Ex-Google, L.A. people there. I bet you know you but they raise largest series A in Los Angeles history. And a lot of that comes down to their manacle about building a product in engineering culture in Los Angeles. And they have great problems and the benefits are extraordinary. When I first heard them I said, “Why are they doing this?” And the answer was because that’s what Google gets. And I was like, that’s really ambitious. Fortunately, they have the metrics to support it so they can do it. I’ve never been in that position before in that company to produce so much revenue that you can have those kind of benefits. But they have insane benefits. Insane.
Andrew: What do you at ZipRecruiter?
Jeffrey: I’m the chief strategy officer, it’s my first time not being the boss.
Andrew: I would say one of the things that I’m proud of is that people trust us with information. And you did say months before, “I’m investor in this company and I just don’t want to hear a couple of things that our coming up, but I don’t want you to say until this day.” The date was the 26th of August.
Jeffrey: Yeah.
Andrew: Come back to . . . I’m looking at your face . . . Did I say anything that we shouldn’t have said. No, right? We didn’t reveal. I didn’t reveal anything. I think everything now that I have in my notes is public. But going back to how you created a first version at Convertro, I understand what you did with YLighting. Now that it was a brand new company, all you had was the licensing of YLighting. What are the first versions? How did it look like? How did you build it up in a way that it was actually useful to other people?
Jeffrey: Well, by the way, I see the sun breaking through on the beach right outside my windows. Do I need to adjust the shade?
Andrew: No, we’re fine but I can see that you might lean to get it out of your eyes. So adjust the way you want to. We’re fine here.
Jeffrey: It’s raining when we’ve started. Now there’s sun so that’s why I love LA. the first version was really focused at the most fundamental product. Today we have hundreds of features. A user guide that will put you to sleep, but the fundamental idea was, tell me for each sale. Let me look and see how that sale is distributed across all the marketing events that led to that conversion. So if a person clicked on a search ad and then viewed a display ad and then clicked on an affiliate and there was a hundred dollar order, show me how exactly, divide that $100 appropriately. We eventually developed much more sophistication around how we divided that credit.
Andrew: How did you do that first?
Jeffrey: Equal.
Andrew: Equal. I see so then at least each aspect of marketing got some credit to bring in the sale. How did you track where it all was coming from?
Jeffrey: We built a tagged server. So you had a JavaScript to your website and then collected the data. We used Google Analytics, yeah…
Andrew: Okay.
Jeffrey: The Google analytics has a very sophisticated way of collecting data and we back solved for that.
Andrew: So what you were doing was telling them here is the marketing channel worked specially well noticed it because tis not getting referral link, it’s not the last point of touch. How . . . that’s how you do.
Jeffrey: Let me try to clarify that. The first pass was just let me know what happened. I told people; don’t worry about who got credit, or which one’s mo,re important just simply knowing what the top of the keyword funnel looks like. Now you can do that in Google Analytics today and there’s another solution out there, but when we came out with that, we were the first.
Andrew: You just said, the last thing the person did was Googled our name and bought. But what’s the first thing that they did that got you and then you come back to the see?
Jeffrey: Because otherwise you think that the keyword they clicked on wasn’t valuable because you misinterpret it as non-converting. Whereas, actually it’s incredibly converting, it’s the thing that brought them to the top of the funnel. I’ve always argued that data quality is more important than algorithm at the end of the day. Now the businesses we are in today. The data quality challenge is table stakes but in the algorithm is where the action is. So instead of equal distribution of that conversion value, now we have lots of people and lots of machines figuring out, what’s the appropriate distribution or attribution, but in the very first thing, all I did was say, collect data and feed it back to them. And I had it embedded into the software that people used to buy keywords, keyword, bid management software. I was involved with GO-TOS which became Atlas one point, the first 3rd party bit management software. So I embedded into that and then our customers put that on auto-pilot.
Andrew: How many developers did you have building that first version?
Jeffrey: The very, very first version was built by three developers.
Andrew: How did you find them? How did you recruit?
Jeffrey: So my co-founder/CTO of EchoSign, it was his best friend from college, and they all went to like Russian immigrants who grew up in Chicago and went early age to University of Illinois Urbana, where half of the internet seems to have gone to.
Andrew: What are your proudest moments, after the Super Bowl? What happened there?
Jeffrey: Well I mean now I’ve upgraded that since I spoke to you. Now my base is Black Friday but in Super Bowl we had three clients for Super Bowl TV commercials. And we were able . . . we were eventually learned that we have to provide attribution, not just for digital channels but for television, radio, catalogue. And we had three of our client had Super Bowl commercials which are typically believe to be waste of money by digital natives. GoDaddy and TurboTax, and have anything in your notes . . .
Andrew: No, I am looking that up right now.
Jeffrey: NFL was the third.
Andrew: Two, I’m sorry.
Jeffrey: NFL.
Andrew: NFL.
Jeffrey: The football league, it was fantasy football, they had a few spots running. And by 8’o clock in the morning the following day everyone had that our expected value of those commercials, as they would manifest over a long period of time, calculated to the penny. We did all of that without a hiccup, without anything . . .
Andrew: How could you know the value of a Super Bowl ad? What did you do to measure that?
Jeffrey: So, there are a couple of pieces that go into it. But the heart of it is, algorithms figures it out, what a baselines of visits to say GoDaddy’s website would look like against every channel as long as conversions. And it looks at what actually happens after that spot ramp and then the figures like out, oh, was the difference or the incremental is attributable to the spot. That’s the simplest way to understand it. There’s a lot more that goes into it, time decay. How does it that effect things over a long period of time? Life time value. The latency or the time between the spot and when people typically make a purchase decision. So we got a lot of people, 38 engineers, who are some of the smartest people I’ve ever had the pleasure of working with, who all pretty much are on the bleeding edge of everything, and media measurement and media science.
Andrew: You decided to . . .
Jeffrey: Right now, I’m in my conference room I’ve all things like three of the probably the top 5 professors of marketing science in the world. Two of them are editors, co-editors of the journal of marketing science but these guys are Berkeley, UCLA, Stanford and Chicago. I feel like I’m dropping a lot of school names suddenly, there are lots of school names in this interview. But we bring in all kinds of people, some of those people are full time members of our team.
Andrew: Speaking of the way, one of them apparently gave you a jacket. We worked to make sure we got. You asked me, “Andrew what kind of jacket should I get?” “Wear this one or that one.” I said, “Go with the logo. And I’m looking at it. And we can’t even see it on camera. We might as well at least show you one time.
Jeffrey: There you go.
Andrew: You’re still in the Convertro offices today?
Jeffrey: I am. I’m here through the next week or so.
Andrew: And what happens after that?
Jeffrey: Then I walk down the street and join the ZipRecruiter. So ZipRecruiter is also in the same neighborhood. I’m improve the view. So I don’t get peekaboo view of ocean. And I get a full view but I can’t walk out onto a patio outside . . . or can be outside anymore.
Andrew: Oh, that is a nice part. At least you be able to see outside. Why did you sell your company to AOL?
Jeffrey: So I was like either raising say another 30 to 50 million dollars. We’ve only raised 12 million on Convertro to build an enterprise sales force, to put a product in the hands of the largest advertisers in the world. It’s a highly consultative, very enterprise sale. I know how to do that but I didn’t think that the 50 million dollars would ensure me success. Instead I was looking for a platform of sellers who had existing relationships with advertisers, who can walk my swat team of domain experts into every company that could potentially buy our product and let that be the way that it gets in the hands of most people.
Andrew: And did you start going out looking for either funding or for an acquirer or was it something that happened to a competitor that led you to the sale?
Jeffrey: So, there’s two competitors in our space that are part of most if you look at for us straight, we’re all in the top quadrant. A dormitory in visual IQ. In visual IQ, I went and hired bankers and went out looking for someone to acquire them. We us in a dormitory, we’re both very intent on building our own large businesses. But incidentally, Visual IQ had the best customers in the business. They had American Express and JP Morgan Chase. Although they didn’t really have a real technology, they were mostly a consulting firm. So they hired bankers and went around with a book to sell the company. And as every potential acquire looked at the book and that with these guys, that resulted in them looking and saying, “Who are these competitors? We’ve got to call and talk to them to get up to speed.” And in so in week I got eight calls from big companies, corporate head departments. The following week I had another 10 so by the third week I had 20 people wanting to come and meet with us and talk about Convertro. So, at first I didn’t take those meetings but I learned, that was a very large social network that I didn’t take the meeting with and the way it was interpreted by some prospects is that I didn’t that have the guts to show up and talk. So then I knew it was going to hurt me on the sales side, I better start taking these meetings. And so we did and that was the original reason for . . . started to spend a lot of time on with some folks. Eventually, the Dormitory and Convertro were acquired and Visual IQ ironically, no one bought.
Andrew: The one that started it all was the one that nobody bought. It’s interesting actually, it’s kind of like high school in a way. You’re saying because you didn’t go up and show off, you had people assumed you didn’t and that hurt your ability to sell.
Jeffrey: Yeah. That was definitely a challenge. I think you learn something on every company. I think a lot of what I learned on this one was, we didn’t sell ahead and that hurt us. We didn’t engage in exaggeration. I built a company of truth seekers and really smart people who are typically humble and only want to say what they have. But in this particular space we were competing against people who had no problem exaggerating what they had. So for us not to step up actually hurt us a lot. In fact, there’s been two Forester reports in the space. The first one we didn’t do so well on, and lot of it had to do with . . . we didn’t exaggerate what we did. We told them what we do. We told them what we’re working on. We told them that we’re excited about that. But our competitors exaggerated what they did at that time, and they stacked higher in the rankings. The second time around, I wouldn’t say that we exaggerated, but I hired professional analyst relations folks and we killed it. We won, and I think that’s particular microcosm of some of the challenges in particular that technology startups will have. They need to . . . unfortunately; they need to sell really far ahead or else the market’s going to discount them.
Andrew: Almost vaporware is what you’re saying. This is what we . . .
Jeffrey: You call it vaporware you’re never going to get your employees to believe in this mission. But you definitely need to sell the roadmap, not the reality.
Andrew: Interesting. Yeah, I can see how that might be different from you personality.
Jeffrey: Yeah. And not only that, but different than the personality of the types of folks that I attract and retain.
Andrew: Right. But if that’s the game, you’re saying, you have to recognize the rules of the game and play it.
Jeffrey: Well, like I said, I play to win. Don’t hate the player, hate the game.
Andrew: And you have played to win, and you have won so many times. I could have done this whole interview on EchoSign. I could have done this whole interview on YLighting. I’m so proud to have done it on Convertro. Was Tech Crunch right? Actually a few different places said, you sold the businesses for $101 million?
Jeffrey: A hundred and ten.
Andrew: So they were just a little bit off. Wow. And were they right about 80 million in cash?
Jeffrey: No it wasn’t all cash. We just achieved our [earn out].
Andrew: I see.
Jeffrey: A couple of weeks ago, and we got the whole earn out, paid up.
Andrew: And that was the difference when people reported and where it ended up.
Jeffrey: Yeah, we were paid, actually it’s all public information. I just don’t want to mess it up. I think if you afford me the liberty of possibly having people in the comments say that I got it wrong. I was just going to say, I think it was 80 cash on close. Actually I can’t remember, but anyways it’s a 110.
Andrew: Okay. Congratulations on the success and the way you that you build it. I’m so proud to have you on. One of the ways I measure success is through people whom I admire. And more importantly whose work I admire respect that kind of interview that I do here. It’s so good to see the feedback coming in from you. Thanks for being a fan. Thanks for doing this interview and I’m looking forward getting better and better and leaving a legacy here. Thanks a lot. Home of the ambitious…. Congratulations. Bye and thank you everyone.
Jeffrey: Keep it real.
