Google Analytics Benchmarking Newsletter 2011, Volume 1, July 2011

1. Introduction

Welcome to the first volume of the Analytics Benchmarking Newsletter!

This month, we are replacing the standard “benchmarking” report in your Google Analytics account with data shared in this newsletter. We are using this newsletter as an experiment to surface more useful or interesting data to Analytics users. Data contained here comes from all websites which have opted-in anonymous data sharing with Google Analytics. Only those website administrators which have enabled this anonymous data sharing will receive this “benchmarking” newsletter.

You may be wondering, how many websites are in this “anonymous data sharing” pool? Currently, hundreds of thousands, and we’ve endeavored to make all of the metrics here statistically significant.

The date range of comparison for this newsletter is from November 1, 2010 – February 1, 2011. Comparison is done with data from November 1, 2009 – February 1, 2010. Absolute metrics such as total # visits, pageviews, or conversions for all opted-in websites are not reported.

To simplify the prose, the phrase “websites” will represent “websites which have opted into anonymous data sharing with Google Analytics” for the rest of this newsletter.

2. Site Metrics

Compared to a year ago, websites have seen reduced pages / visit, average time on site, as well as bounce rate.

 

11/1/09 – 2/1/10 11/1/10 – 2/1/11 Difference
Pages/Visit 4.9 4.5 -0.4
Bounce Rate 48.2% 47.0% -1.2%
Avg Time on Site 5:49 5:23 -0:26

 

2.1 Breakdown by Geography

Our anonymous database has aggregated geographic breakdown at the country level. Here are a few representative countries and their respective aggregate metrics. The first number in each cell represents the metric for the date range 11/1/10-2/1/11. The parenthesized number is the Year over Year delta compared to a year ago.

 

Country Pages / Visit Bounce Rate Avg Time on Site
United States 4.7 (-0.1) 42.5% (-6.1%) 6:06 (-0:10)
United Kingdom 4.9 (-0.3) 41.5% (+0.2%) 5:38 (-0.27)
France 4.4 (-0.4) 49.7% (+1.4%) 4:40 (-0:08)
Brazil 4.1 (-0.1) 47.8% (-2.9%) 5:20 (+0:03)
China 4.1 (-0.1) 58.2% (+1.0%) 3:46 (+0:37)
Japan 3.9 (-0.1) 48.6% (-9.0%) 3:47 (-2:59)

 

For bounce rate, the distribution by country is plotted below:

 

 

The distribution above is annotated with some countries — which seem to indicate a story of leisure and stage of economic development. For a related metric: average time on site, the distribution by country is plotted below:

 

 

The type of countries annotated in the average time on site graph above seem to be in reverse order as those in the bounce rate distribution.

2.2 Breakdown by Traffic Sources

Traffic sources below are identified by how the “source” and “medium””” parameters are received by the Google Analytics collecting servers. Here is an article describing what these designations refer to.

 

Traffic Sources Pages / Visit Bounce Rate Avg Time on Site
Direct 4.0 (-0.5) 47.2% (-4.0%) 5:21 (-0:07)
Referral 5.0 (+0.1) 43.1% (-1.1%) 6:36 (-1:48)
Organic Search 4.9 (-0.1) 47.9% (-1.1%) 4:43 (+0:06)
CPC Search 5.6 (+0.0) 41.4 (-1.7%) 3:57(+0:07)

 

2.4 Conversion Rate Distribution

Many marketers’ favorite metric is conversion rate. Here is the worldwide distribution of Google Analytics “goal conversion rate” by country.

 

 

Would anyone have guessed that states which are known for conversions are also high for their citizens’ goal conversion rate? Note that for some states with few population, the statistical significance of the conversion metric comes into doubt.

3. Traffic Sources

Traffic sources below are identified by how the “source” and “medium” parameters are received by the Google Analytics collecting servers. Here is an article describing what these designations refer to.

 

% Visits from Sources 11/1/09 – 2/1/10 11/1/10 – 2/1/11 Difference
Direct 36.5% 36.8% +0.3%
Referral 21.0% 19.4% -1.6%
Search Engines 27.0% 28.0% +1.0%
Other 15.5% 15.8% +0.3%

 

4. Operating Systems

Browsers and Operation Systems (OS) are identified by the “referrer” string sent by users’ browsers.

 

% Visits from OS 11/1/09 – 2/1/10 11/1/10 – 2/1/11 Difference
Windows 89.9% 84.8% -5.1%
Macintosh 4.5% 5.2% +0.7%
Linux 0.6% 0.7% +0.1%
Other 5% 9.3% +4.3%

 

5. Comments

This is the first volume of our Analytics Benchmarking Newsletter. We hope that it provides useful insights. If you have specific comments or suggestions on how to improve this newsletter, please send your feedback to: analytics-benchmarking@google.com.

Leave a Reply

Your email address will not be published. Required fields are marked *