Panel by Tantek Celik, Jonas Luster, Joyce Park, Ernie Hsiung , & Danah Boyd.

Social relationships only require single directional relationships. Social networks place more emphasis on relationships than personal information. Many of the things which are currently being touted as social networks are community based sites which place greater emphasis on individuals than on relationships.

Danah stated some of the more common problems with social networks are

• social akwardness

• articulation problems
• problems with being public

Joyce also stated another recurring problem is that many networks are created by men and do not take women into account. Another common problem is that some social networks fail to adapt to their audience.

A couple interesting facts:

• Orkut took off in Brasil for two main reasons
  1. Brasil had two large cities which for a long time did not mix much but recently started mixing more frequently on a larger scale.

  2. Orkut originally had many flags on it. This reminded many people in Brasil of the World Cup. Some people joined because they wanted to beat out the other countries. Now the innovation at Orkut is driven by the Brasil community.

• For this next statement you will have to pardon any cultural ignorance I display, but I think this is what they were trying to say :) Friendster is popular in some Muslim areas because with certain segments of that cultural background you have to show how and when you met someone, you can't just meet someone and instantly marry them.

They also talked a bit about XFN and recommended a few books. Jonas recommended John Scott's Social Network Analysis. Danah recommended Robin Dunbar's Grooming, Gossip, and the Evolution of Language

Online panel with Philip Kaplan, Jason Calacanis, Gokul Rajaram, Henry Copeland, & Bill Flitter. To get an overview of the size and compare the various blog advertising networks (and Google AdSense)

# of salesmen:

• Google has a ton of salesmen.

• BlogAds has 2 salesmen
• Pheedo has 3 salesmen
• AdBrite has 6 salesmen. 3 for inqueries and 3 for calling advertisers. AdBright also recently accepted $4 million dollars in VC funding.

service & fees:

• Google AdSense places contextually relevant ads on your site. While they have the largest distributed ad network they do not state the profit share percentage. Google introduced Smart Pricing a while back, which works to automatically discount the value of a click if they believe the site has low quality traffic. He stated that the publishers were not given a smaller share of the cost per click in these situations, but that the overall cost per click is lowered.

• BlogAds allows you to sell flat rate ads. They keep 20% of the ad spend.
• Pheedo allows you to track your RSS feeds and also places contextually relevant cost per click ads in your RSS feed. They keep 35% of the ad spend.
• AdBrite allows you to sell flat rate targeted ads. They also allow you to sell bulk untargeted ads if you chose to. They provide a great overview of stats including repurchase rate. They keep 25% of the ad spend.

Business models for blog networks:

Jason stated that he expects many publishers to stack the various advertising networks on their sites to achieve maximum profitablility. He stated that when he started his network he wanted to pay bloggers 50% and that did not work. He later said that he gave the bloggers the first $500 and then 50% and then that did not work. Now he stated that he pays bloggers a flat monthly fee. He stated Gawker media pays $2,500 per month. One of the major benefits of blogging is that the time spent editing normal media is not usually required for bloggers.

He stated that at this point the bloggers still usually need to do something else to get to self sustaining but he thought that the business model would allow bloggers to just blog for his network sometime next year.

Jason also stated that for the people who run multi channel blogs that the more profitable channels are used to pay for the less profitable channels. The channels which lose money can still be ran to offer a wider selection or offering to people who like the network. His goal was to run 700 channels. Recently I believe they launched their 75^TH channel.

Criticsms for this panel:

1. For this panel being about making mone with online ads it was a bit excessively focused on blogs. If it was to be exclusively about blogs it should have said that on the schedule.

2. I can not fathom how there was a panel about making money with online ads and there was exceptionally little coverage given to the concepts of:
   • selling ads directly

   • partnering up with a merchant or creating your own product
   • affiliate programs
   • selling links
   • sponsored posts
3. I asked about low quality screen scraper sites participating in AdSense and got a fairly canned response from Google AdSense's product manager. He stated that they use technology and people to review sites and disapprove low quality ones. If they do that then they are doing a lousy job.

   Discounting the click prices a bit and still leaving sites in the program still encourages people to cloud up the web with crap sites. Granted a long time ago I made a few crap sites myself, but Google could at least be honest about not caring about that issue.

4. On that same question Jason Calcanis stated that click fraud is a non issue. He stated that people were just looking to make up ways to find problems with the business model. He then pointed at Google's AdSense product manager and stated click fraud is not a problem, and he of course agreed.

   Funny that this is the complete opposite answer to what George Reyes, the CFO of Google, gave a while back.

   "I think something has to be done about this really, really quickly, because I think, potentially, it threatens our business model."

5. As a person who has probably bought and sold around $100,000 worth of various ads in the last year I know that fluctuations in bid prices or click volumes can have significant effects on ROI and ad spend.

   Even if the fraud averages out over time it still:

   • tears apart consumer trust

   • increases account management time. If it takes me more time to manage an account then it costs more money for me to manage it, and that means less money is going to Google.
   • causes more inefficiencies in the market
6. This panel should also give tips on how to increase site exposure and ad click through rate, since those concepts directly relate to ad revenue.

Bonus Rant:
Jason Calcanis used his speaking gig to rip on Gawker media no less than three times.

• talking about finding Michael Jackson news on Gawker

• talking about finding a Lindsay Lohan nipple slip on Gawker
• saying "I read it on Defamer. It must be true."
• words and tone matter. apparently I was not the only person who noticed Jason using the microphone for the wrong purposes
  

  post: Sometimes panelists need to be reminded that they are there to talk about the topic at hand, not to trash people they don't like and then admit they never heard of the topic of the session till they were asked to be in it. Could have been a really interesting session, though.

  comment: i agree. i think two audience questioners put it best when they basically said you ALWAYS have to consider the authenticity, knowledge and motivation of ANY source. After sitting through this session, I question all three for panelist Jason Calacanis.

For a person strong on the blog ethics concpet you would think he could give a speech without doing that crap.

Panel by Nick Finck, Jeffrey Veen, & Kit Seeborg.

Jeffry Veen shows Haynet as an example of good usability.

He believes that to perform well it is best if he imerses himself in the topic of the clients site. He jots down lots of notes on yellow stickies and covers the wall with the various ideas.

During project definition and research many possible solutions can be quickly eliminated, thereby lowering the overall cost of a project.

To find some of the core ideas behind the sales point of the website Veen collects market research data from people in the real world, often while failing to mention websites, as he feels that the needs overlap. The research ends up averaging out to about $200 per person and most smaller tests usually test about 6 to 8 people.

Veen also feels that analyzing data from server logs also tends to fall off track. Since you are only tracking where people are going wrong with the current system it may not bring about the most effective solution.

I totally agree with this concept. Often people have self fulfilling prophecies about their sites. The first day I talked to a marketer for a major cable and high speed internet provider. He stated that since most of their referals to their website came from direct type in traffic or bookmarks that he did not feel they needed to do SEO. If you don't know what traffic you are missing out on then it is kinda hard to see the need to go after it. This is why only analyzing your referal logs can give you bad or incompelete market data.

Seelong states that it is a good idea to follow the money trail before you get too deep into a project to ensure you will be OK to do whatever you desire. She also likes using survey services such as Survey Monkey.