Web Masters Episode #15: Scott Crosby

You already know Google Analytics. It’s on 70% of the world’s websites. But do you know how it got started? Listen to Google Analytics co-founder, Scott Crosby, tell the full story on this episode of Web Masters.



Urchin Software Corp (@urchinbiz) | Twitter

Scott Crosby:

Don’t give up, even though it seems like it’s completely flaming out. There will be some dark days, and you probably won’t be able to raise money as easily as it seems like what you read about, so just persist. I think most people can probably pull it off if they’re willing to put in the time. There’s this perception out there that you can start a company and sell it in six months, and you’re a billionaire, but it’s really a very rare occurrence. Most successful companies, they’re going to take 10 years. So give yourself that much time.

Aaron Dinin:

Give yourself 10 years to build a company. The grand scheme of things, of course, 10 years isn’t very much. But when you’re living it, 10 years feels like a long time. A lot of unexpected things can happen.

Scott Crosby:

We were actually close to raising some significant amount of VC money in 2001. We had term sheets from a couple of VCs for about a $7 million investment, and we were going to give up some large percentage of the company, which pushes the goalpost a lot further away. People seem to lose sight of that. But you raise a bunch of money, you’ve got to sell for a lot of money, or else you’re not getting anything, and so we were going to raise $7 million or so. The guy from one of the VCs was going to fly out, and we’re going to sign the papers. He was either scheduled to be on one of the doomed flights or his flight got canceled the next day or something. It was really close to him being on one of those planes. The world went to hell for a while in September 11th. The deal fell apart.

We laid off about 30% of our staff. We were pre-spending the money as one does because we were like, “Oh, yeah. We got term sheets. We’re going to get this money.” We didn’t really know much about VC. We can start hiring. We hired a bunch of people. We got another office in our same building in Downtown San Diego, and it’s going to be great. We’ve got all this new resources. Deal falls through, and we had laid off 30% or 40% of our people. It was very sad. It was hard.

But then, we just had to just really refocus on the unit economics and like, “We got to make money here. We can’t screw around. We’re not going to raise any more money anytime soon.” Rich uncles gave us an emergency bridge loan of 300 grand. We paid them back and paid them warrants for their trouble. So they got some more stock, but it took us a while. We had to really just work hard and grind it out for two years under miserable conditions of having barely enough money to keep the lights on.

Aaron Dinin:

In other words, building companies is a long and winding journey, and the biggest obstacles you face are the ones you’ll never be able to see coming. Now, as you heard, that was definitely the case for Scott Crosby. His 10-year startup journey began when he got tired of delivering boots in the back of a broken-down, old van. Instead, he decided to partner on a tech startup with his childhood best friend. That journey ended in an office down the hall from Larry Page and Sergey Brin at Google.

Along the way, he and his team had lots of twists and turns during their startup journey before ultimately and somewhat ironically helping build a company that in its own sort of way is all about mapping complex journeys. He’s the co-founder of Urchin Software Corporation. I’m guessing you’ve never heard of it, but there’s a good chance Urchin Software Corporation has heard of you. Urchin built the software that became Google Analytics. Are you ready to hear the story? Great. Let’s get dialed in.

[INTRO]

Aaron Dinin:

Hi. I’m Aaron Dinin, host of Web Masters. I teach innovation and entrepreneurship at Duke University. I study the history of internet businesses, and this podcast is, well, a way of learning about internet businesses. We do that by talking with the people who built some of the world’s most important and impactful online companies. One of the most popular tools used by a ton of those companies and millions of others is what we’re going to discuss on this episode because we’ve got a really interesting guest. His name is Scott Crosby. He’s the co-founder of Urchin Software Corporation, better known today as Google Analytics.

Google Analytics is, of course, the free website stats tracking tool from Google that lets site owners see how many people are visiting, where they’re coming from, what pages they’re looking at, and other similar kinds of somewhat creepy things. The software currently operates on somewhere around 70% of all websites on the internet, which is an absolutely crazy amount of scale. It’s certainly the kind of reach its founders never imagined because, well, website analytics didn’t even exist when they started the company, but we’ll get to all that. Before we do, I want to take a moment to thank this podcast’s sponsor.

Web Masters wouldn’t be possible without the support of our sponsor and partner Latona’s. Latona’s is a boutique mergers and acquisitions company that specializes in helping people buy and sell their cashflow positive internet businesses and digital assets. If you’ve got a profitable SaaS business, e-commerce store, FBA account, content website, domain portfolio, or any kind of online work from anywhere company, Latona’s can help you sell it. If you’re interested in buying an already profitable internet business, Latona’s can help you find the perfect one. You can check out their website right now and browse their catalog of businesses being sold. It’s constantly being updated with new listings. Find them all or contact a Latona’s team to get their help selling your business. Just head on over to latonas.com. L-A-T-O-N-A-S.com.

If you’ve been listening to Web Masters for a while now, we’ve picked up on a bit of a trend. Most of the people we’d meet here discovered the web at an early age, saw its potential, and relentlessly pursued the opportunity. That’s certainly not the case for Scott Crosby, and that’s part of what I enjoyed so much during my conversation with him. His story is a refreshing reminder that building a business isn’t always as intentional as it’s often made out to be in hindsight. Even better, he certainly doesn’t hide that.

Scott Crosby:

Yeah. I always think of this whole story like a series of random fortuitous events that I would like to say were planned, but not really.

Aaron Dinin:

Sometimes, I feel like the entrepreneurs I’ve spoken with here on the show were a bit more honest, many of them would say the same exact thing. That’s a big part of what makes Scott’s story compelling. He reminds us that you don’t have to have planned out every step of your entrepreneurial journey before deciding to take that initial leap.

Scott Crosby:

Yeah. I was a computer hobbyist back in the ’80s. I was born in 1970, so the dawn of the personal computer was when I was first introduced to them. I had an Atari 80 or something. Yeah, it’s an early Atari computer. I had an IBM PC Junior before Windows existed, which was really expensive. Eventually, when Windows came out, I was appalled by that, and I got an early Mac black and white thing. So I was into programming as a hobby in high school and college to some degree. I like to make games. I was never educated in computer science per se. I just read magazines, and DAS manuals, and things.

In college, I was an English major. I did some preliminary work to do an electrical engineering degree, but I didn’t finish it. I don’t think they had computer science as a degree at UC Santa Barbara in 1988, or if they did, I wasn’t aware of it. I just did it as something I enjoyed. I never had any real school about it. In high school, they had some computer lab kind of thing with very little supervision. So you just go in there and screw around on computers for an hour. Easy way to kill a period.

Aaron Dinin:

Well, I got to say it’s a relief not to be talking to yet another person who saw computers and immediately thought, “Oh my gosh, these are going to change the world.” As someone who didn’t think that the first time I saw computers, it gives me like this tiny glimmer of hope that I’ve still got a chance of becoming a great tech entrepreneur.

Scott Crosby:

I mean, I was 9 or 10 when I got into them, but it was purely just because I enjoyed playing with them. It wasn’t that I had foresaw a career in them or anything. It was very clear from a young age that I never planned to work for anybody. I was going to start some business, and I didn’t know what it was going to be. So I didn’t really worry too much about career paths, and high school, and college. I didn’t consider that an option, so I’m like, “Well, I can do some major that’s completely impractical like English literature because it’s going to be separate from whatever I do to make money.” I like reading books, and I like writing. I like grammar. So I can stomach that as a major, but I’m not going to plan to do anything career-wise based on it.

Aaron Dinin:

So then, as a fellow English major who also eventually went into tech, I got to ask. How did you wind up building a career around it?

Scott Crosby:

I had this childhood friend. We always had planned to start something. We were post-college roommates in San Diego, and he was working just to have some kind of job in the UCSD Space Physics Department, just doing whatever work they needed to put their syllabus online and really basic early stuff like that. I was driving around in my van delivering cowboy boots because I have a friend, one of my good high school buddies who has since passed on, but he was… His dad was the founder of a company called Boot Barn, which became a major chain of Western wear stores. So as just a way to make rent money, I had this beat-up, old Ford Econoline. I would drive it around to their various stores in San Diego County and equalize their inventory of cowboy boots.

So I know that wasn’t a permanent long-term solution. Then, one day, my friend, Paul, who I grew up across the street from in Tustin, California, which is Suburban Orange County like bedroomest of bedroom communities. He lived across the street from me. I knew him since I was four years old. So he comes home from work one day and just says, “Hey, we can network these two computers together that we have with this cable, and we can do HTML. Have you ever heard of that?” I’m like, “No, show me what that is.” So he showed me what a webpage looked like. He figured out how to get a modem to work, and we networked our two Macs together. He showed me the website he was building for UCSD, which was your, just gray background with blue text, early, early 1995 website. I’m like, “That’s awesome. Let’s do that.”

Aaron Dinin:

That, listeners, was the start of Google Analytics, a guy transporting boots out of the back of his beat-up Ford Econoline van, seeing a crummy early ’90s website and saying…

Scott Crosby:

That’s awesome. Let’s do that.

Aaron Dinin:

By the way, Paul, the friend he’s referring to in this story is actually Paul Muret. He’d go on to a pretty successful career as Scott explained.

Scott Crosby:

He still works at Google, by the way. He’s a VP of engineering. He’s a big deal.

Aaron Dinin:

He definitely is a “big deal.” He currently leads Google’s artificial intelligence and health teams. So 20 years from now, when we all have Google microchips implanted inside of us detecting early warning signs of heart attacks and strokes, we’ll have Paul to thank for that, which actually, by the way, means we’ll all owe a huge debt of gratitude to Scott’s uncle who gave Paul and Scott their first round of seed funding.

Scott Crosby:

I had a rich uncle, this biotech founder guy that made DNA sequencing equipment in San Diego. I guess if anybody was my entrepreneurial role model, it was my rich uncle. He built this pretty successful local… probably had a couple hundred employees, and he always told us that if we wanted to start something, he would help us fund it. So we took our little one-page business plan to him and said, “We’re going to make websites and host them.” It was essentially what it said, and he said, “Okay. Great. Here’s $10,000, and you can use a corner of my office in Solana Beach.” So we’re like, “Okay. Cool. Done.” So we incorporated and set up shop. We had our two old computers, and then we bought a Sun SPARC 20, which is this pretty ancient Unix machine by today’s standards. We set up an early version of Apache web server on that, and we just started trying to get business.

Aaron Dinin:

How did you get your first customers for… What would you call it? I guess an early web development shop?

Scott Crosby:

So we just tapped into whatever networks we had and just asked people if they needed websites. A few people did. We got bootbarn.com, and we eventually got Sharp HealthCare, which is the biggest health system in San Diego, and of course, my uncle’s company. We have people that know people, and there weren’t a lot of people making websites back then. So my brother’s girlfriend worked for Honda. We eventually got in there and all these other places just by virtue of knowing people or plotting our way along. So we started getting some paid web hosting customers and developing websites. Then, eventually, we got our own office down in Old Town San Diego.

Aaron Dinin:

So how does that, a company building websites for people, eventually become an analytics software company because those two things obviously look nothing alike?

Scott Crosby:

We were hosting websites, and bandwidth was really expensive back in the ’90s. So we had to pay hundreds or I don’t know, thousands of dollars a month for our higher traffic websites just for the bandwidth. So we developed a really basic tabulating software to figure out how much bandwidth our customers were using on our web hosting server. My friend, co-founder Paul, built a web interface for it, and I said, “Hey, that looks pretty cool. I think we could show that to customers, and they might be interested in how many hits they’re getting and how much bites are transferred, which pages are accessed,” because that was all pretty easy to see from the server logs.

He built this cron job basically that would just every day process the logs and figure out how much bandwidth our customers were using, and then we would send them a bill at the end of the month. Then, a little while later, he came up with a nice-enough-looking interface that we could start giving the customers logins to it. One day, I don’t know if it was one of us or both of us, we just realized, “Well, this is actually a better business than a consultancy.” When times are good, you have to hire people. When times are bad, you have to lay them off. You’re always hustling for business, and you know why it sucks. It’s a pain.

Aaron Dinin:

Yeah. I think every tech entrepreneur gets their start by building websites for other people, and it’s always terrible.

Scott Crosby:

Yeah, totally. Clients are a pain in the ass. So early on, maybe in 1998 or so, we brought on my brother, we consider him one of our co-founders, and another guy named Jack Ancone. So we had four guys essentially that we consider co-founders, and so we developed the very first version of Urchin just as an accounting tool. Then, we had this epiphany that that was really the business we wanted to be in, and so we had a board meeting. We decided to jettison our real business, which was web hosting and design, which 100% of our revenue came from. We decided just to determinate that, give all our customers to a friend who also did web hosting, and concentrate 100% on the software side of things.

So it seemed like, in some ways, the obvious thing to do because it’s a scalable business with no real fixed costs and you can make an infinity of money if you get it right, but it had zero revenue. On the other hand, we had all these customers that were paying us. I don’t know. Maybe we’re making $50,000 a month on average or something like… decent money. We could have a few employees, but we knew that it was going to be a hard slog forever to maintain those customers, and get new ones, and grow that revenue to a meaningful net number.

So we decided, “Let’s just get rid of that business,” and we raised a little bit of money. It’s the only real money we ever raised, which was consequential. We asked our uncles for some more money. I actually have another rich uncle that is a furniture magnate in San Diego. So we raised some money from our two uncles, some friends and family, and one little VC called Green Thumb Ventures out of New York that was a friend of Jack’s. We’d raised a total, 100%, all-in I think at $1.2 million or something, and that was that. That was the end of our fundraising.

Aaron Dinin:

So in my experience, fundraising usually requires some sort of pitch or vision, but you all had no revenues, at least not for the analytics product. So what were you pitching? What was your proof that this was worth investing in?

Scott Crosby:

I think it was mostly just our word that we thought it was a good business, and we had some positive feedback from customers. I don’t think we were charging more than a handful of them for it yet. But in the scheme of things for rich uncles, it was a small money, so they’re like, “Yeah. Whatever. Here’s a couple hundred grand. Just don’t bug us for a while and make sure that we get it back someday.” It’s a little fuzzy. It was a long time ago, but they basically took our word for it.

We showed them the product. They’re like, “Well, yeah. That looks pretty cool.” My more tech-savvy uncle was like, “Yeah, I see that that’s valuable,” and there are other companies doing that like Webtrends existed at the time and a couple others, but it was small business. I don’t think there was anything close to proof that it was going to be a good business, but there were some evidence, and we were asking for such small amount of money, they just took us at our word.

Aaron Dinin:

Just to be clear, even though Scott is shrugging things off as just his rich uncles giving him a few bucks, there’s nothing typical about being able to raise a million dollars like that. Also, worth noting. By that point, Scott and his team had already built a successful consulting company. It wasn’t called Urchin yet. It was called Quantified Systems, Inc. According to what Scott just told us, it was generating more than half a million dollars a year in revenue. It had clients like Boot Barn, and Honda USA, and a huge healthcare system.

Now, if I’m an investor looking at that, even someone’s rich uncle, I’d probably be more than a little intrigued by what the team had already proved capable of doing. In other words, I think Scott is being a bit modest here, and that seems to be part of his personality, but let’s be clear. They’d already gotten some impressive traction, even if it wasn’t explicitly in the form of a software analytics company. All they really needed at that point was, well, a new name.

Scott Crosby:

We’re in San Diego. There’s the ocean theme and all that. We had a really hard time coming up with a name, and we had this one customer. My brother will get mad at me for telling this, but we had this one customer. It was a division of Pioneer Electronics, and they were called Laser Visions Direct or LVD. They sold 12-inch laser discs. Even back in the ’90s, it was an old format that was being phased out. But they still had their hardcore base of customers, and so they put this business online, and they were one of our customers for a long time.

There’s this guy there that would call us all the time. It’s one reason we got rid of this business. He would call us all the time and just talk forever. He would never shut up. He would call sometimes. I would say, “Oh, hey. Okay. Hold on a second,” and I would say, “Hey, Paul. It’s your mom,” and then transfer him over. It’s just ridiculous. His name was Uchida, and we ended up calling him the Sea Urchin because he would stick to you, and he would never let go, and you would just talk forever. So one day, we’re like, “What are we going to call this thing?” It was just a term we said a lot, “That guy’s an urchin.” It became a generic term for us for anybody that would just latch onto us and waste our time. It doesn’t make any sense, but we just said, “Let’s just call it Urchin.” “Okay. That’s cool.” We got the URL, and that was that.

Aaron Dinin:

Are you kidding? I feel like that makes it even better. I’ve spoken with a ton of founders, and a lot of them seem to have had almost like quasi religious experiences about finding their company’s names. Now, I got to tell you, that was never my experience.

Scott Crosby:

Well, that’s the reality of it, and there’s not really a good reason for it at all. It’s just a term we used in the office for somebody that wastes your time, and it has nothing to do with the product at all. It was just something we said a lot. So it just came to mind, and we all sat around and said, “Oh, okay. Cool.”

Aaron Dinin:

So if I’m being honest here, you’re not exactly describing yourselves as a crack team of super entrepreneurs. It sounds a lot like you were just, I don’t know, like going with the flow.

Scott Crosby:

Exactly, a crack team. None of us had any experience starting a business at all, and we’re pretty much just out of college with little to lose because we didn’t have any kids at that point. We always had planned to start a company, and we thought we’d probably be successful at it for no good reason, just that we thought we could do it. There was no The Lean Startup books out and all those kinds of things. Maybe some Jim Collins books were out at the time, but we didn’t read any of that.

The only one that we had any exposure to, which I admit was helpful, was The E-Myth. I don’t know if you’re familiar with one. It’s The Myth of Entrepreneurship. I remember exactly one thing from that book, which I still find helpful today, which is over-communicate organizational clarity, and over-communicate it again and again. That’s the biggest, most helpful business advice that I ever got, and I will always remember it. There is so much internal confusion in every place I’ve ever been like, “Who does what?” Just over-communicate it. That’s incredibly helpful.

Aaron Dinin:

That is definitely good advice. I’m going to start using it with my students who… I swear. I tell them things a million times, and half of them still don’t do it. Okay. So you switched to being a software company, and it’s not like a web-hosted software at this point, not like what we think of Google Analytics as. It’s more a piece of software that people are downloading. Is that correct?

Scott Crosby:

Yeah. I mean, our very, very first customers, we ran internally for them and gave them a login. But back then, there was no such thing as SaaS software. It wasn’t a thing. I think we sold our first copy of Urchin in 1998. I had a framed check on the wall and everything. It was like this $200 version, and it didn’t really even occur to us to do a hosted version. People bought software like shrink-wrapped software off the shelf, and we didn’t sell it that way, except for in Japan. Everywhere else, they would just download it from our site and run it on their own servers.

That was the way you sold software. So we weren’t breaking any new ground in that arena because it didn’t even occur to us. We would sell software. You download it. It’s a one-time thing. You could buy a support contract. That was our first foray into recurring revenue somewhat later. But yeah, we would sell it. We’d give a certain amount of free upgrades, and then you could buy it again if you liked it. Our business model was to sell traditional licensed software like Microsoft, and that worked really well for us for a long time.

Aaron Dinin:

What was your customer acquisition strategy? How are you growing during this time?

Scott Crosby:

So we had a few strategic directions. We focused on low price per sale, but lots of users because we thought that’s a free way to get lots of exposure. So all the web hosting companies, that was half our effort. Institutional customers like universities, the Defense Department, and all kinds of government entities. So we would go after them mostly at trade shows, A lot of those entities, because they’re hard to call them directly. So we would just go to all the trade shows, Search Engine Strategies, those kind of things.

Probably the most important early sale we had was with American Honda. We found out that they were using Webtrends and it couldn’t process a day’s worth of logs in 24 hours. So they were falling further and further behind every day. We said, “Hey, let us prove that we can do it in a fraction of the time, and we can keep up with the logs or servers you’re generating and actually give you timely web analytics.” We didn’t call it that back then. It was called web stats, or web statistics, or something. So we got into Honda that way, and we crunched their logs in 10 minutes. They were suitably impressed, and then we got them as a customer. We just leaned heavily on milking our customer success stories, and doing case studies, and splashing our logo everywhere. So we got Honda using it. Sharp HealthCare used it. A bunch of major universities were using it. We would just trumpet all that.

My favorite sale that we ever got though was really just a cold call in which was to EarthLink. Back then, in the early 2000s when I think we first got them, they were the biggest hosting company in short of AOL. You think of them more as a web access company, but they also did a ton of hosting. So my brother, being very tenacious, just filled out this web form for partnership request, and he filled it out every day over, and over, and over again. Probably a few dozen times. Eventually, their VP of web hosting emailed him back and said, “All right, all right. Come on up to Pasadena and show us what you have.” Got them as a customer, and then that opened up all the doors at the other web hosting companies, which turned out to be critical from a strategic point of view because Google wanted to give it away free to the whole world. So the fact that we had a proven track record of being scalable to tens of thousands of websites was attractive to them.

Aaron Dinin:

This right here is the big distinction between Urchin and their competitors at the time because it’s not like there weren’t other companies providing web analytics, or as Scott notes, they weren’t even called analytics yet. They were called web stats, but those other companies were all focused on the top of the market and the enterprise space. Whereas urchin just wanted as many users as possible.

Scott Crosby:

Website Story. I don’t know if you remember them, but they were a local competitor that actually went public and were worth some real money. There was Webtrends. It was public. There were a few obvious candidates that were already either public or acquired, but they all had a different focus than us. They were more into the enterprise sale. That was never our main thing. We considered it too painful and too intensive. So we’d rather just do lots of little sales or a sell to the… My brother called them the fishing boats. You get all the acquirers of customers. Go to them rather than the customers themselves.

Aaron Dinin:

By luck or maybe it was actually a skill and Scott is just too modest to admit it, having a web stats platform that could scale across thousands of websites was exactly the value proposition that a certain fast-growing search engine company out of Mountain View was looking for.

Scott Crosby:

We did a lot of trade shows back then, and we went to the Search Engine Strategies in San Jose. I think it was spring of 2004. A couple of Google guys approached us and just wanted to talk about doing a partnership of some sort. When a big company talks to you about a partnership, it usually means they want to buy you. It seems to me in my future experience. So we said, “Yeah. Great. Okay.” Then, we had a couple meetings with them there, and then eventually had some on-campus meetings. But before or roughly concurrently with that, the Website Story guys actually made an overture to us and then made us a cash and stock offer. They come out of the woodwork. We were starting to do pretty well by our standards at that point. We were really profitable and growing at some double digit percentage every month. When you’re doing well, people just sniff it out. It’s funny how it happens.

So Website Story made an offer to buy us, which was very generous offer. There was nothing wrong with it, but we felt like we were going to get one from Google as well. So we pushed them off a little bit and tried to delay it. Eventually, Google came through with slightly lower in terms of dollars, but half of it in stock. We felt like even though Google had just gone public themselves in 2004, it was by no means a foregone conclusion that they were going to be a trillion dollar company someday. We could see the writing on the wall that Google was probably the better horse to bet on.

Aaron Dinin:

Yeah, I’d say you made the right bet. Why do you think Google acquired Urchin instead of maybe some of the other companies they could have gone after at the time?

Scott Crosby:

A couple of the significant players in the space had already gone out, Webtrends and Website Story. But I think more importantly was our focus on scalability, which we always just innately thought was a better way to do it. They were about the enterprise sale. We were about large quantities of sites, whether big or small and high-performance log analysis. So tons of credit goes to my co-founder Paul, who really is a genius. He wrote a great piece of software that just did this job better than the other ones.

I think Google does a very thorough analysis of your code base and your engineering team, and they gave it the thumbs up. So they said, “Under the hood, this is a good piece of software, and it works as advertised.” It really is better suited to the Google way of doing things, which is all about scalability. “Can we serve hundreds of millions of websites with this?” They felt like it could, and it pretty much did. When given the sufficient infrastructure, it handled the load.

We just had a different focus for better or for worse. Maybe we could have sold for a lot more money if we were like Omniture and doing hundred-thousand-dollar deals to individual companies. They ended up going public, and it was a big exit and everything, but that just wasn’t our thing. We were just more about backend infrastructure and planning so we could handle whatever load was thrown at us. That really fit the Google mindset better, I think.

Aaron Dinin:

Can you talk a bit about what happened after the acquisition? I mean, getting acquired by Google, that’s the big dream, right?

Scott Crosby:

So I was there for five years. My brother was there for 10 years. Our other founder, Jack, was there for seven or eight years. He went into a business development role. At first, I got slotted into the sales organization as we weren’t really selling software per se, but it was really important to Eric Schmidt and Larry to a lesser degree that we get some big name customers using it for respectability, so they could do case studies and that kind of thing, which Google was big on doing.

So I had my own sales team from Urchin, which was five of us. I think it’s five or six. They gave us another four to six Google salespeople that they felt like would be a good fit for going out and doing more like enterprise-y sales just to get some marquee names, even though that wasn’t from a backend point of view. Not really what we went after. Some of our milestones were based around getting a certain percentage of the Fortune 500 using it. I don’t remember what it was. Maybe it was 10% or something, so fairly big goals, but they were achievable.

After we achieved all our goals, and got our earnouts, and milestones achieved, I transitioned into managing the engineering team that was continuing to develop the installed software version of it for a while. Did that for a couple of years, and then it became clear that Google wasn’t going to put a lot of resources into it. You could tell it was going that way, but it was still a significant business. It still brought in several million dollars a year. So they didn’t want to just turn it off, but I didn’t feel like it was… It wasn’t the way to climb the ladder at Google. So eventually, I just decided I was done with corporate America.

Aaron Dinin:

So I feel like lots of companies that get acquired disappear or get dramatically reshaped. But aside from the name change, Urchin grew into one of the most important parts of Google. At the same time, it can’t be cheap to give away analytics to 70% of the web. So what’s going on? Why does Google invest so heavily into this free analytics tool?

Scott Crosby:

Yeah. The server storage was not insignificant. When we did sell to them, they said, “Okay. We’re going to keep the traditional license software side of it alive,” which I managed for a while because we have all these customers that want to have updates, and new releases, and so forth. But they really back-burnered it and put 99% of their effort into the SaaS version. Within a few months, they released it for free, and it got such overwhelming demand that they shut it down for six months and retooled the infrastructure because all the Google SREs were mad at us for using up too much software server resources. It was a positive disaster for a while. It would have been worse to have the opposite happen, crickets, but it was an infrastructure problem for a few months at Google.

Aaron Dinin:

Yeah, that. Why would Google do that? Clearly, they’re spending a lot of money on something that’s free.

Scott Crosby:

They did want to see a return, and they did quite a bit of internal studies to figure out like, “We know how much we’re spending on the infrastructure. How much are we actually making from this?” Fortunately, the thesis proved correct, which was Eric Schmidt and the product managers that actually did the acquisition was that if customers had real visibility into AdWords and the results of the AdWords, that they would then spend more, and that was indeed the case. I don’t even remember the specific numbers, but it was something like, “We spend X, and we make 10 X.” It was like an order of magnitude. So it was every dollar we put into this, we make 10, and so it was internally considered a pretty solid success.

Aaron Dinin:

Okay. I’ve got to ask, and I’ve always wanted to ask this question to someone who might actually have an answer. Having analytics on all these websites means Google has visibility into most of what people do after they click links from a search result. So in other words, they can actually see what people are doing even when they’re not explicitly on Google. I mean, that’s a little weird, right? Is Google using that data?

Scott Crosby:

I don’t know the answer to that. I think so. Google, if they’re doing anything behind the scenes to just be all knowing and omniscient, I don’t know. I honestly don’t, and I think that they probably do, but they generally live by their dumpy evil mantra, and they don’t do a lot of stuff. But having been on the inside for a while, I would be like, “Ooh, that’d be embarrassing if that got out.” They generally do stuff above board and are cool about it. Not that there’s anything necessarily nefarious with knowing everything that’s going on in the web, but do they use that to tune their algorithms? I don’t know. They probably do.

So Google Analytics is on… I don’t know. Last time I looked into it, maybe 70% of URLs that are active on the internet. Yeah. Clearly, they know exactly what’s happening out there. On every single website that they got the tracking code on, they know how many visitors are going there. They know what they’re clicking on. They know where they’re coming from and going to, if they want to know that. Do they actually pay attention to any of that? It seems like a good thing to train your algorithms on, but I don’t know.

Aaron Dinin:

How do you feel about that? In the current world of data profiles and all sorts of online tracking, what does it mean for a company like Google to have access to all that data?

Scott Crosby:

So obviously, they know that all that individual visitor history drilled down as we call it. It’s all there. They could look into it if they wanted to, but I think they felt like anything that feels creepy is creepy, so just don’t do it. I mean, when you would have meetings with Larry and Sergey, they shut down stuff they thought was sketchy. We had some features that Google even shut down because they felt they were a little bit too much like individual user tracking. It was anonymized, but it was still… We had these features where you can see what an individual user had done on your website. It was like, “Some alphanumeric string had gone to this page, and this page, and this page. This is what their profile looks like.” I mean, like what computer they’re using, browser, all that sort of thing.

Aaron Dinin:

Okay. So Google actually had Urchin remove the really personal tracking details? That surprises me, and it makes me feel a little better about Google. Urchin being wrapped up into this big company created more protections for consumers. I wouldn’t have thought that to be the case. I would have thought it would be the other way around. Google would demand more data.

Scott Crosby:

I think they just considered it too much of a political hot potato. It was just creepy. I even joked with our development team that we could do a composite sketch of the user like they do for court. When you see it on the news, the people that draw the court scenes. Obviously, it was a non-starter, but it was hilarious to think about it. So Google, I’m sure that they can see where this is going and like, “Nah, we just don’t want to go there. We can show plenty of monetizeable information without getting anybody’s hackles up over privacy too much,” because they’ve always got problems with privacy anyway. So they’re just trying to keep it at bay to the extent they can.

Aaron Dinin:

Since we’re talking about things Google killed off from the company, how do you feel about ultimately losing the Urchin name?

Scott Crosby:

I thought Google Analytics was a good name, and Urchin lives on and goes to the machine with UTM. The acronym is Urchin Traffic Monitor. So I feel like it’s still there.

Aaron Dinin:

For those of you who don’t know, UTM codes are a feature in Google Analytics that allows website owners to better track where traffic to their website is coming from and what it does once it’s on the site. For example, the next time you see a link to say a New York Times article on Facebook, if you look closely at the URL, you’ll probably see a question mark and a bunch of stuff after it that says things like, “UTM_source=facebook.” That’s part of how the optimization people using Google Analytics at the New York Times are going to know who came to the site via Facebook. Then, they’ll be able to answer questions like, “How many people who come to our site via Facebook eventually purchase a subscription?” which, of course, is a very important business question to be able to answer.

Scott Crosby:

Anytime I see a UTM in a URL, it delights me a little bit because I remember when we thought of that, we were just sitting around our conference table saying, “Okay. We need to have these name value pairs in the URL. What are we going to call it? How about Urchin Traffic Monitor?” We called it tracking module at first. But then, we came up with modules, so we changed it. The official last name of it was Urchin Traffic Monitor, and so that lives on. It was nice of Google to allow that to stay. I figured they must have thought it was too much trouble to change it.

Aaron Dinin:

Imagine what it must feel like to be Scott. Two decades after randomly deciding to name your company based on an inside joke that made fun of an annoying customer and everywhere you look, you see that name impacting just about everything anyone does online. It’s got to be a unique feeling, almost as unique as discovering you actually invested in that company and completely forgot about it, which also apparently happened to one of their early investors.

Scott Crosby:

We had this one little VC out of New York who after he wrote the check to us, we never heard from him again. 250 grand. He didn’t hear from us again, and so we told him we sold the company. Probably like the most surprising thing he’s ever heard in his life. It’s like, “I completely forgot about that investment, and now you’re telling me I made like 30X on it or something?” I don’t know if that moved the needle for him or not, but it had have been some welcome news.

Aaron Dinin:

Yes, I’m sure it was very welcome news. Unfortunately, I’ve got some unwelcome news for you, which is we’ve reached the end of this episode. I want to thank Scott Crosby for taking the time to share the story of Urchin and Google Analytics. If you want to see what he’s up to these days, you can find him on Twitter. His handle is @impunity. I-M-P-U-N-I-T-Y. I have no idea why that’s his handle. You might want to ask him. I feel like there’s probably a good story in there somewhere. I’m on Twitter too with a much less interesting handle. It’s @AaronDinin. That’s A-A-R-O-N-D-I-N-I-N. I also post lots of articles about startups and entrepreneurship over on medium.com. Just search for my name, and you’ll find all of them.

If you’ve got questions, thoughts, concerns, or feedback about this episode, you can share those with us on Twitter. We’re @WebMastersPod. If you’re interested in buying or selling an internet business, be sure to visit our sponsor’s website, latonas.com. If you like the audio for this podcast, you can thank our sound engineer, Ryan Higgs. If you want more stories about the people who helped build the internet as we know it, well, we release new episodes just about every week. You can get them by subscribing to Web Masters on your podcast app of choice. While you’re there, please consider giving us a great review and share a link to this episode with your friends. No UTM codes required. All right. That’s it for now. I’ll be back soon with another episode of Web Masters. Until then, it’s time for me to sign off. Goodbye.