In my last post, Do you have a primary browser that you use most of the time, I suggested that many people may have a primary browser that they use most of the time and a secondary browser that they use much less frequently.  For example, we could use a browser on our home computer for most internet access but occasionally use a browser at work to read an article or find something.  Or we could use a work laptop at work and home and have an old home computer that we use much less frequently.

Those examples got me thinking about the days when we use browsers, particularly the distinction between weekday use and weekend use.  Do a lot of people do most of their browsing on a work computer during the week and a home computer on the weekend, or do they tend to do most of their weekend browsing on a computer that they also use for browsing during the week?  To find out, I counted the weekday days and weekend days when browsers saw ads and divided the browsers into three groups: 

• Mostly Weekday Browsers have more than 2.5 weekday days with ads for every weekend day with ads.
• Mixed Day Browsers have between 1.0 and 2.5 weekday days with ads for every weekend day with ads.
• Mostly Weekend Browsers have at least as many weekend days as weekday days with ads.

I found that most browsers were in the Mostly Weekday Browser group, and this was true for all ranges of days with activity.  There were fewer Mostly Weekend Browsers than I expected - there were a significant proportion of Mostly Weekend Browsers with less than one day of activity a week, but very few with between 1 and 2 days of activity a week. 

I also looked at the ads going to these groups.  Again, I was surprised by the results.  Hardly any ads went to the Mostly Weekend Browsers.  That doesn't mean that there were hardly any weekend ads - the third graph on the right shows that there was a much higher proportion of weekend ads (about 24%) than Mostly Weekend Browsers (less than 10%).  But most of the weekend ads went to the very active browsers that saw ads almost every day.

This data suggests that people who do weekend browsing do so on a browser that they also use for weekday browsing, and they tend to have more days with ads during the week than on weekends.  That's even true when you account for there being 2.5 times as many weekday as weekend days, since most browsers are in the Mostly Weekday Browser group.  There do not seem to be a lot of people who use one primary browser during the week and a different primary browser on the weekends.

That result supports my idea that many people have a primary browser where they are most often exposed to ads and one or more secondary browsers which have much less ad activity, but it could also be the case that a lot of very active internet users have a home browser and a second work (non-home) browser that both have a lot of ad activity and a lot of the less active browsers are used by people who are not very active on the internet (or who have many different browsers that they switch between on a regular basis, which seems unlikely to me).

It's hard to say which is the case by looking at browser activity alone - we need to understand the activity of the actual people using the internet.  The Pew Internet and the American Life Project is a good source of data about the way Americans use the internet.  According to their Daily Internet Activities report, 73% of American adult internet users reported going online "yesterday" (in April 2009, which is around the time of the browser results we've been looking at), and it sounds like most of the activities that they reported engaging in involved advertising.

Since we serve ads on most large publishers, we probably recorded ad activity on many of their browsers.  (These results are based on our Atlas ad-serving data, not MSN data.)  To me, this suggests that we probably reach most internet users with advertising at least once a week.  Since the number of US browsers that we reach with more than one day with ads a week is roughly the size of the US internet population, I think that most of those browsers are the primary browsers of reasonably active internet users.  There probably are also be a significant number of people who do a lot of browsing at both home and work and a significant number who don't have a lot of internet activity, but it seems like most people have a single primary browser that we reach with ads at least once a week, on average.

Wow!  I didn't expect to write all of that when I started this.  I hope we can continue on this topic in a future posting.  I'd also like to look at the home / work (non-home) distinction some more and look at browsing activity by day of week and time of day.

Erik Hanson

More from the Microsoft Advertising Institute