Monday, November 27, 2006

First days at Rockyou

I'm finally moved out of my eBay cubicle and into Rockyou's new offices in downtown San Mateo. It's a nice top floor space - one studio-type room where all 9 of us sit (no cubicles), one conference room (a cardboard packing box is our conference table), a mini-kitchen and a cramped bathroom. As you can tell, still settling in :) We have direct access to the building's roof deck and are within walking distance to my favorite pizza joint on the peninsula, Amici's. So far so good.

A few thoughts from a 4-year corporate guy going back to a startup.

I forgot how all-encompassing startup life could be. Rather than being one of the last to leave, here I'm usually first. Lance and Jia (Rockyou's cofounders) were understanding about my family/kids time requirements (7:30-9:30). That said, with the kids asleep, I'm usually back on the laptop working through deal terms, ripping through MySpace profiles to verify our competitive position or Skyping with Jia/Lance on the week's upcoming activities. I've always been a night owl and don't go to sleep before 1AM or so... except now its spent working/thinking rather than watching Conan or playing poker. Given my poker playing as of late, that's a double bonus.

A large user base is a good thing. Rockyou occupies a very specific niche, personalization widgets for social network, blog and personal website users. As we close in on 10MM unique users in November alone, we're realizing that this niche is more canyon than ravine. While the absolute number is great to have, a large customer base gives us more flexibility to innovate across multiple paths, vs living and dying on one specific product bet (whose focus is to get users in the door in the first place).

This is a lesson learned from eBay and other consumer Internet plays in general... scaling to massive user growth/usage as a primary goal vs monetizing the user base is an absolute requirement these days. Focusing on monetization first gives the competition the opportunity to undercut with a free product. The continuing growth of online advertising ensures that strong traffic will generate $$ at a base level at least. (PlentyofFish's $300K a month from Adsense is a clear example of this). Unlike 1995-2000, when the majority of online advertising spend came from highly volatile, unprofitable web startups (which caused the bubble to pop when they ran out of cash), these days advertising online is being driven by the Fortune 1000, hundreds of thousands of brick/mortar businesses AND online retailers/players. This base of business won't disappear overnight, which helps secure the future for online plays that are able to cross millions of users and keep them engaged. So, for those folks looking to scale users quickly, send me an email ;)

True wireless Internet access is another good thing :) I'm loving the Sprint EVDO wireless hookup on my laptop. We went down to Disney Land this weekend, and I was able to surf the web all 6 hours down I-5. The connection is pretty cheap ($30/month) and fairly fast -- about as good as my old DSL line. Anyone who does serious travelling should get one and no I'm not being paid by Sprint ;)

Lunch with Princesses is expensive... One of the biggest rackets around is Lunch with the Disney Characters at California Adventure. We paid $70 for a burger, fries, mac n cheese lunch for four, but got to meet Princess Belle, Snow White, Cinderella, Ariel and Sleeping Beauty. Woot! I'll post the thrilling pictures tonight. As a result, I'm thinking of opening up a local chapter of Lunch with the Disney Characters in Danville. Honestly, it's an almost perfect business model.

Wednesday, November 15, 2006

Alright, I was planning on waiting until I made the formal jump, but my friend Cindy outed me. In a couple of weeks I'll be joining the team at Rockyou as their VP of Business Development. Blogging about Internet plays over the past 1.5 years invariably got me riled up about what's happening out there. The disintermediation of the Internet, the increasing personalization and relevancy for individuals online, the mass market acceptance of Web 2.0 sites... it was too hard to ignore.

Now, let me address something right off the bat. I love eBay (especially the eBay Motors team which has a huge vision for the future and a super-talented team to get us there). My last 4 years at the company have been both super challenging and rewarding. For all the naysayers out there, simply put, I firmly believe that there is no better large/public company to work for, especially in the Internet space. eBay cares a ton about all its stakeholders, whether its the community of buyers/sellers, eBay employees or its shareholders. Ultimately balancing all three can be quite a dance, but eBay has a ridiculously strong team across many levels to see this through.

Of course there are challenges and competitors, but if you had a chance to see the day to day here, I'd think you'd walk away impressed by the caliber of folks trying to accomplish the near-impossible - making eBay, Paypal & Skype the universal platforms for online commerce & communications... something which I personally believe will come true, if it hasn't already. Have I had my momentary disappointments? Sure. But they pale in comparison to the positive experiences I've had and the support the company has given me in my professional development. I'm a big supporter of the company and will always be. Semper Fi.

So why leave? Well, it came down to finding a small company which is the leader in its space, doing all the really cool, cutting-edge things that I've been posting about for months. Rockyou is the leading widget provider for social network users (including MySpace, hi5, Friendster, etc) bar none. The Rockyou site had over 8.5MM unique users come through in October (Alexa is so very very wrong) and has way over 100K flash widgets embedded a day on blogs, websites and user profiles. This network of widgets is now in the millions, generating over 100MM views a day. Lance and Jia, the founders of Rockyou, have a great vision of the company's future which mirrored my interests and thoughts about the Internet experience of tomorrow. Better yet, they understand Rockyou's customers, the social network user, and have built some of the most dynamic and coolest widgets on the web with them in mind. The team is also completely geared to work with partners looking to capture that viral marketing magic that propelled YouTube and others to success. Something unique is happening at this company, and to cut the chase, I wanted to be part of it.

So I'm going to miss the heck out my good friends and peers at eBay, but look forward to working with partners and cutting deals in this dynamic online personalization space. More posts to come about going back to startup land :)

Tuesday, November 07, 2006

A lot's been written lately about the poor results generated by Alexa. (Here's a rundown on wikipedia) Frankly, I agree with most of them. Alexa can be fairly good when measuring 'like' sites with similar user demographics or capturing directional trends. However, by not supporting Firefox, over-representating Windows users, and with fairly simple gaming of Alexa (Webmasters can ask users to download the Alexa toolbar en masse driving ranking upwards), results can be real spotty.

All that aside, plenty of folks use Alexa as a key reference point for advertising deals, business development, and even fund-raising. Simply put, it was the best available free alternative for web traffic measurement. That's until launched their 'Snapshot' function this October. Check it out. Compete runs sampling over 2MM users (including Firefox users ;)) and its results are simply a LOT more accurate. I've been looking at two sites in particular that according to Comscore and Nielsen Netratings have over 2MM uniques. On Alexa, their traffic rank and reach are below 3,000 and 300 respectively. Since Alexa bases their reach number out of a universe of 1MM and knowing that total Internet users equals about 1B, that would equal about 300K uniques... hmm. That doesn't ring true.

On Compete, these sites are measured at between 2MM to 3MM users. The other nice thing about Compete is that they offer actual estimates of users (vs an estimate based out of 1MM user universe). Now, there's a few things I'd like to see improved, such as multiple time ranges and pageview estimates, but the accuracy of Compete vs Alexa has convinced me to switch. Still, I'm going to miss Alexaholic :(
Listed on BlogShares