After my Jangl experience, there were several criteria I had for my next venture:
*Keep funding amount and team small, growing only with validation
*Solve a real consumer problem
*Make sure it’s both web and mobile day 1
*Make sure there’s an efficient way to acquire customers day 1
*Have a low cost structure
*Use open source technologies
*Hire only the best
*Naturally viral
*Take investment only from people that want to validate the thesis with me
I graciously joined Venrock as an entrepreneur-in-residence in the Fall of 2008. I didn’t mean to time it with the world’s financial crisis, but why not. It’s a good time to start a company when most people aren’t looking right?
I spent the better part of three months thinking and mapping out over a dozen ideas for new companies. There was Chadr which was a web and mobile topic-based chat room. There was Smash which was free mobile messaging. There was MyBurbs was a neighborhood, trusted Craigslist. There was Swarm which was location-based deals on the go. There was Tkrew which was teens doing online work for brands they love. Then there was Cc:Betty, which was a smart email service. (That links to a demo at DEMO). Disclaimer: The night before I ended up in the hotel bar pre-pitching press until about 3am, so I was in fine shape for my 8:30am demo...
Email hadn’t really improved materially in forever, and there was a particular set of problems with group email. For example, I email a group of people, and someone replies-all. So far so good. But then someone replies only to me, leaving the rest of the group out of the loop. And then someone else forwards the message to others, disconnecting the conversation even more. Then there’s the poor bastard that has 21 messages out of this parked in his inbox. Where to start? Who said what when? What do I really need to know? You get me.
Enter Betty. We came up with a crafty “onramp” to using Cc:Betty. Email. So in a group email, simply add “[email protected]” to the message. We would take that email, it’s contents and recipients, and re-create it in a shared “mail space”, and Betty would email everyone a notification saying “Heya I’m Betty and I’m tracking this conversation for y’all...[link]”. Anyone receiving that message can login and have a very quick glimpse of the entire context of the conversation. Any attachments are in their true form and actionable, i.e. photos, videos, music, calendar dates, etc. It was sort of an awakening of the whole conversation. You could filter on people, content, to-do’s, links, votes, and locations. We took it to market in an open beta and got our first 100,000 users. It was a great sampling size to learn from and nurture. With a seemingly effective onramp in place, a built-in network effect, and an elegant UX (at least for those days), the next box to check was distribution.
We engaged both Yahoo and AOL to see if they’d be willing to let us integrate Cc:Betty into their email platforms as an app. Combined there were a few hundred million people in addressable market. If we could pull off a seamless integration, these could prove to be customer acquisition engines for us. The good news was, they were both planning 3rd party programs, so we inked a deal with both of them to be pilot partners.
With a couple distribution deals in hand, a strong beta with compelling engagement patterns, and some validating press, we closed seed funding with Venrock leading, Hillsven and Seraph Group following.
We built the AOL integration into an AOL beta environment, getting lots of good feedback very quickly. AOL seemed to like us too, sort of courting us in seemed. They’d show up with doughnuts and just want to hang out and learn more. In parallel we built a proof of concept integration with Yahoo, but we were awaiting some more evolved APIs to do what we really needed to to. But then the startup gods pumped the brakes... Both AOL and Yahoo were in the midst of management change-overs, and both deals were put on ice so new leadership could evaluate the business and set course for a new strategy. Realizing these were both going nowhere fast, I emailed Brad Garlinghouse, who had just joined AOL, and I killed our deal. And since we didn’t have anything in production at Yahoo, I just let that fizzle. With our distribution outlets dead, we went back to focus on our own property.
The good news, was we had a very small cost structure, and were only 4 employees, so we could experiment and turn on a dime. We kept at it, grew the user base and grew engagement. Over the course of a few months, we grew to several hundred thousand users. We learned that 30% of our users were using Cc:Betty with friends/colleagues/teams, and 70% were using it in the workplace. Of the 70%, the average number of “mail spaces” was 9, the average number of people in them was 7, and the average number of messages posted per mail space was 39. On some level this was a private Facebook, but on other levels it was an email replacement, and at least a Basecamp replacement.
People who grocked Cc:Betty loved it, but others totally missed it. Some thought Betty was a spam bot and others just don’t play nice with group technology tools regardless. It turned out that crafty onramp we had was too clever for its own good. We needed to simplify, and we needed to play to the heat in the workplace user base, and fast.
With an arsenal of data, we spent late Fall and Winter of 2009 rebuilding Betty into a consumerized enterprise looking product, with all the bells and whistles. We added real-time, presence, threaded messaging, a whole new blue UI and UX structure, nixed the cc’ing part, and rebranded as Threadbox. The 70% workplace users loved it. We grew about fivefold the first month, and engagement grew about threefold. Over the course of a few more months, we were deducing the notion of a premium service. Our data told us companies would pay for hosted versions, for enhanced security, self-branding, and additional storage.
Unlike my last go around, we had a very supportive and engaging board. We were close enough that I could call them with the best or the worst of news. That makes all the difference in the world. But I’ll never forget the board meeting where we presented the data and the plan for a premium service. We had about six months financial runway, so naturally my concern was - can we build the premium service and grow it fast enough to matter (meaning, fast enough to get enough market data to slam dunk a follow on financing round and keep building)? We would need to add resources and spend against billing, customer service, etc. The news and plan was met with some skepticism.
While the board were pleased with the pivot results and our earnest execution of Threadbox, it kind of felt like that awkward moment when you first realize the girl wants to break up. I really loved Venrock and Hillsven though, and they loved us. So we were all keeping it real. Bobby Lent from Hillsven was the first to speak up...
”You. You guys.. You’ve done some really great work here. But man, you gotta ask yourselves, is this what you really want to do? I think you’re capabilities are better than what this road has cut out for you. Michael, when we first met you I thought you were the best bus dev guy in our portfolio, and now I see you as the best product guy too, and I mean that. Do something bigger, it’s ok.”
Bobby was right. We weren’t enterprise guys. We were consumer guys to the bone. Our thesis, however well thought through, didn’t play out as big and as soon as we needed.
Brian Ascher from Venrock added, “We liked our bet here and think you guys are great. But if you look around this consumerized enterprise space you have Box.net, Dropbox, Yammer, Xobni, and a whole cast of others. They’re all raising upwards of $40 million in funding, and spending years to build out. And guess what, none of them have exited yet, so we don’t have an obvious proxies to deduce. We got to ask ourselves, do we want to spend the next 3-5 years doing this... Michael, what do you want to do?"
I got the memo. And being honest with ourselves, we all completely agreed.
Over the course of about 3 seconds I recollected having dinner with Aaron Levie four years prior, when Box.net was just a prototype. I recollected a conversation with George Zachary, a Yammer board member two years prior. I recollected various conversations with Matt Brezina from Xobni. I recollected a lunch with Drew Houston from Dropbox at SXSW. All these guys were swinging, raising lots of money, spending lots of time. Is that what we wanted to do? Is that what the investors wanted to do? Could our captable even tolerate the dilution we'd need to go the distance? (Lots of times when you're an EIR and the VC you're an EIR with funds you, there are captable implications that make it just a little more complicated).
Regardless, the problem we solved wasn’t big enough for enough people to warrant a venture business, but it could be a great addition to a larger SaaS offering or social property. We would spend the next six months improving the KPIs and we’d go on to sell the company.
We quickly engaged and pitched a leading CRM company and found a really nice product fit with them. At the time they were getting ready to go to market with a social product that we were becoming experts in. We started negotiating and had a seemingly good rhythm, but they started dragging their feet and trying to buy time. Clearly they were hedging their bets with another acquisition candidate, or they weren't really as serious to begin with.
We pitched Facebook on unifying all messaging into one real-time service. Although they didn’t buy us, but they built it - sort of.
Google had launched Google Wave to great hype, but users had a difficult time getting their heads around it. We talked to them about a potential deal, but that discussion involved moving to Australia and spending four years on an earn out. We kept pedaling.
Then I reached out to Mike Jones, the CEO of an aging Myspace at the time. Mike was a nice guy I had known from his days at Userplane/AOL and my days at Jangl. I gave him the same pitch I’d given Facebook. We would overhaul Myspace mail, instant-messaging, groups, forums, and create a whole new real-time conversation piece: Myspace Threads. Mike was very receptive so we started working on a structure. He got some GMs involved, corp dev, legal, etc, and within a couple months, we would be the last acquisition Myspace ever made. In July 2010, I would become a product and technology head and my team would be a ninja crew that would start chipping away at Myspace Threads, and in hopes of a full stack rebuild of Myspace. But that's all a story for another day.