How well do you know your interviewees?

How much can you know about your job candidates? Looking at the cover letter, resume, a couple phone interviews and in-person interviews later, you have spent less than 8 hours in total with that person. Can you really figure out if someone is great and a good fit just from this?

For start ups, it seems trite to say that you only want to work with the smartest, best people, and accidentally turn down really great people on occasion because of the high screening process. Google seems to skew heavily towards people with high GPA's from prestigious schools, and screen with notoriously difficult questions. I tend to believe that companies like Google are looking at the wrong stats. It appears that in one study, a typical job interview improved chances of selecting the best candidate by less than 2%. Sabermetrics, as described in Moneyball is really a great read on this topic.

For start ups, it's more important to know what people have done during school than what they've studied. Also, an interview can not replace or guide people to figure out the performance of a particular candidate. I think trial hirings make much more sense than locking into a full year with someone. Marc Andreessen states that you're doing great if 70% of your hires work out. Wouldn't it be better if you gave people a baby several week project to test them and make sure you're hiring the right person?

The only thing you need is PSD

There is a time and place when degrees are necessary. I have mostly seen these necessary in large companies, science businesses and consulting. However, in most disciplines, what matters above all else is execution. In these cases, having a bunch of Ph.D.'s in a room can actually hinder your performance. Come to think of it, there seems to be no correlation between advanced degrees and career advancement.

I share with you an alternative model for hiring. Instead of MBA or PhD, look for the PSD. PSD standards for poor, smart, and driven. There are winners and losers in life. At the risk of sounding cliche, knowledge can be acquired but attitude can not. The message from entrepreneurs consistently states that persistence is a key trait for success.

So how do you test if someone is PSD? Well, there are actually a couple ways to tell. What's the candidate's work history? Have they run a paper route for at least a year's time? Did they work for a family business when they were younger? Have they done anything to demonstrate hard work ethic, and a scrappy, results oriented nature?

There is nothing that can not be accomplished if you are PSD, and not afraid to make a bunch of mistakes.

Google: It's not the Search, but the Ads

Everyone knows Google from its search capabilities. However, Google has made nearly all of its money from advertising. "Advertising revenues made up 99% of our revenues in 2004, 2005 and 2006. We derive the balance of our revenues from the license of our web search technology, the license of our search solutions to enterprises and the sale and license of other products and services." -- page 40 of their SEC filing. Eric Schmidt states that Google's core competency is advertising, and hopefully they can leverage their experience in data mining to other forms of advertising.

Google's margin for Adsense is between 22% and 18%[1]. From the 2007 Q1 revenues, Google Network websites running AdSense accounted for 37% of total advertising revenues. So online advertising is a big market, but how lucrative is it? How much could a start up expect to take in this market? What is the best approach to taking marketshare? From O'Reilly:
"In addition, there's still a huge amount of room for growth in internet advertising. Paul Kedrosky's been studying the latest reports from the Internet Advertising Bureau, which show internet advertising at $16.8 billion for 2006. If I recall, tv advertising is at about $80 billion, so there's considerable room for growth as the internet continues to draw attention away from print, tv, radio, and other old media." -- http://radar.oreilly.com/archives/2007/03/the_economics_o_3.html

There are a couple key points where ad networks compete against each other. Several of these are listed by Josh Catone:
- good advertiser campaign management and bidding tools - Google has a large amount of advertisers. Keeping the ROI's for their campaigns is what will keep them coming back. Advertisers will need to be shown that their ROI is better on another ad network before they switch. Campaign management and bidding tools for advertisers are a must, and Google has this end covered well.
- good publisher management and reporting tools - Publishers want to earn money right away, and they want to know how to improve their advertising income. AdSense has done an ok job, but this is where someone can really introduce some real innovation. For a start-up, adoption from top publishers will be a turning point for success.
- great contextual ad matching technology - AdSense is good, but not great. I believe OpenAds, an open source ad serving technology is trying to compete in this avenue.
- high paying advertisers - You need to develop some sort of higher value advertisement that advertisers will pay.

Stay tuned for an analysis of behavioral advertising.

[1] Taking the Advertising Revenue from Google network sites, and dividing it by the Traffic acquisition costs, interpreted as Adsense publisher payoff.
Advertising and licensing revenues (2002-2003) page 15 http://www.sec.gov/Archives/edgar/data/1288776/000119312504074059/d1012g.htm#tx84790_13
revenues and cost of revenues 2002-2006 page 36 http://www.sec.gov/Archives/edgar/data/1288776/000119312507044494/d10k.htm#toc70021_21
advertising and licensing revenues (2004-2006) page 42 http://www.sec.gov/Archives/edgar/data/1288776/000119312507044494/d10k.htm#toc70021_21
traffic acquisition costs from page 45