Selling Kiko

I told this story to some new people the other night, and in its retelling I realized that it might be more interesting than I thought. I've already chronicled getting funding for Kiko, the first ever AJAX web calendar, from Y Combinator, as well as how we ended up starting Justin.tv. What was left out was how transitioned out of the business. 

When Emmett and I were winding down Kiko we thought a lot about what we could do with it. Eventually, we decided we would start a new company, but had to close Kiko, which had several angel investors including YC. We had a bit of cash left in the bank, but felt a great deal of responsibility to return as much money to the investors as possible.

While brainstorming how to do this, I had the idea that we could cleanly liquidate our assets by selling everything on eBay. I wish that I could say the idea was original, but the previous year a site called Jux2 (which showed search results from both Microsoft and Yahoo) sold itself for $105,000. We figured if we could recover even $50,000 we'd be able to give the money back to the investors and walk away with a clean conscience.

We set up a 10-day auction. In the description we included what you were getting (the domain name, source code, and user base -- which had around 70,000 signups but almost no actives), and what you weren't (Emmett and I as employees). We also offered to do a week of integration consulting for $3000 of travel expenses, at the buyer's option. We listed the whole thing for $49,999.99.

We posted the auction on Reddit and, with the help of a few friends, got to the front page. Back then Reddit was much more like Hacker News in its focus on technology and startup related news. Pretty quickly the story was picked up by Techcrunch, John Batelle's blog, and some other tech related sites.

The auction turned out to be a much bigger story than we'd expected. Apparently no one remembered Jux2's sale and we were credited with coming up with the idea -- something I'd taken pains to correct whenever anyone cared to ask. Pretty soon lots of people were writing to us to ask about the auction, more details about the technology behind Kiko, and other random questions. One person wrote us imploring us to open source the code instead of trying to make a few bucks off of it. To be honest, it's probably good for the OSS community that our unintelligible spaghetti code wasn't released into the wild. God forbid someone actually try to use it...

After a few days of fielding questions we were wondering if we'd get a bid, but there wasn't much else we could think to do. But then we did! Our minimum, 50k! Emmett and I were excited and somewhat incredulous -- we had only half expected this to work and were doing it as a last ditch liquidation effort.

Then, in what was probably the biggest web failure I've ever personally experienced, eBay cancelled our auction. I was notified by this fact when Jessica Livingston at YC emailed me asking if we had already sold it because the auction page was down. Digging into what had happened revealed that eBay's terms of service only allows for a single link to be included in the description, and we had included two: one to the domain proper and one to the API.

This was unbelievable to me. First of all: multiple links, why is that so bad? Second: why wouldn't they just contact us and ask us to change it instead of wordlessly canceling the auction? Third: why wouldn't they just not allow you to create an auction with more than one link in the first place? You can do that in a single line of code! Having a $50,000 item cancelled was the worst possible thing eBay could possibly have done to us; we were convinced that we wouldn't be able to get our bidder back. 

In the end reaching out to eBay through official and back channels did nothing. I ended up creating a new auction with a shorter length to attempt to mirror the original end date, then I contacted all the people who had been asking questions and blogs who had written about us and gave them the new link. We prayed for the best and decided to take a vacation to NYC that weekend to visit friends. The auction was set to end Saturday.

We got the minimum bid back and breathed a sigh of relief. A couple days later it was the morning of the last day of the auction, and we were still at the same bid. The number of technical calls I'd been fielding had been heating up to several a day, and I still had some that morning, so I was up early watching the auction. Around 8 or 9 it jumped -- we were at 80!

Exciting times. I think it was 90 degrees in New York that day. I was sitting in my underwear in some of my college friends' shitty East Village apartment, where four of them lived and three of us were crashing. Every few minutes I would click refresh on the eBay page hoping for a miracle while everybody else watched on amused. Emmett was at another friend's place in Brooklyn, so we weren't even watching the auction together.

An hour left to go. It jumped to 100k. I ran out and yelled something unintelligible to my friends. A few minutes later it was 113. I was a genius. 150 with 15 minutes to go. I had just invented a new business model for web startups across the globe. 

In the last 10 minutes it jumped over 100k to 258. I was on the phone with Emmett literally screaming. My friends were cheering. I felt like I'd won the Superbowl. I was a rock god.

Final price: $258,100. Buyer: powerjoe1998, who turned out to be Elliot Noss, CEO of Tucows, with whom I'd had a few conversations. He called a few minutes later to congratulate us. That night I partied like it was my last night left to live.

The Aftermath

There were a few practical consequences to selling on eBay:

  • We should have had an actual contract attached to the auction. Tucows wanted a ton of warrants that we didn't want to provide, and the deal almost fell through after the fact because of it (and took a lot of lawyer-hours to negotiate). It also consumed a lot of my time.
  • Tucows bought Kiko to resell to ISPs (whom they already provided white label webmail services to). We went to Toronto to do the integration week with them. Toronto is a beautiful city.
  • If we'd wanted one, this would have been the best way to get a job ever. About 20 different people and companies reached out to us because they had heard about the auction. This included really big successful ones as well as many small startups.
  • We got to be in a bunch of press, which was probably the first time we got significant coverage for one of our projects. Lots of people wanted to write about the auction, including the Financial Times. I recall that I was obsessed with telling reporters that our next project would "change the way people think about the Internet" and trying to get that specific line in print (I was secretly referring to Justin.tv). Eventually I managed to.