This is a response to an interesting request that asked how people coming to a website react based on their source -- whether those people came from a referring site... search engine... or social network. So I decided to set up an experiment to take a look.

The 1,000,000 pageview sample used for this experiment was taken from one of the blogs I run. The data is from June 2008 to December 2008. Since this data is taken from a blog with a specific design and theme the actions of visitors are 100% subjective and not conclusive towards visitor trends. But I thought it would make for an interesting hub just for fun.

• Behaviour of direct traffic
• People I referred through word of mouth
• 0.16% Goal Conversion Rate
• 6.21% adsense CTR
• 86% bounce rate
• 66 seconds on average time spent at the site

• Behaviour of traffic from referred sites
• People that came from similar "friend" sites and trackbacks
• 0% Goal Conversion Rate
• 0% adsense CTR
• 39% bounce rate
• 4 minutes on average time spent at the site

• Behaviour of social traffic
• StumbleUpon.com
• 0% Goal Conversion Rate
• 0.12% adsense CTR
• 44% bounce rate
• 68 seconds on average time spent at the site

• Behaviour of organic traffic
• Google Search Engine
• 0.3% Goal Conversion Rate
• 4.2% adsense CTR
• 66% bounce rate
• 3 minutes on average time spent at the site

• PPC traffic
• Traffic paid for through Google adwords
• 0.9% Goal Conversion Rate
• 6% adsense CTR
• 48% bounce rate
• 3 minutes on average time spent at the site

Overall I've found that social networking traffic can be the most difficult to reduce bounce rate and increase conversions for. On the other hand search engine (organic) traffic tends to perform quite well.

Terms used above

1. Conversion rate - the % of people who signed up for the newsletter  
2. Google adsense CTR - the % of people who clicked on ads  
3. Bounce rate - the % of people who left the page without looking at it