Do Windows and Mac Users Read Blogs Differently?

All the talk about Chrome OS got me thinking about operating systems and how different OSes are used by people. Also, a comment on my post yesterday by Scott Shapiro got me to segment this blog reader base by windows and mac users in Google Analytics this morning. I looked at all the visits to this blog year to date by mac users and windows users and then looked at where each set of users come from. The data is interesting and here it is:

Mac users vs windows users

The most interesting fact is that 56% of Windows users visit this blog direct or via a google search whereas only 46% of Mac users get here by those two methods.

That 10% difference is offset by a much higher percentage of visits from Mac users coming from links on Twitter, Google Reader, Hacker News, Techmeme, Facebook, FriendFeed, Disqus, and several other services.

There are a few services that weight heavier with Windows users, like delicious, stumbleupon, and Yahoo Search (which is barely used by Mac users).

This community is a leading edge geek community so it's dangerous to make too much of this data. But I think it is safe to say that Mac users are more likely to jump on new services like Twitter and FriendFeed more quickly than Windows users. And it is also true that tech news junkies who hang out at places like Hacker News and Techmeme are more likely to be Mac users than the average news junkie.

The Twitter data is particularly interesting to me. Over 10% of Mac visits to this blog come from Twitter, whereas that number is only 5% for Windows users. That's a big difference and I'm not entirely sure what to make of it. I wonder if Twitter's user base skews Mac way more than the Internet as a whole?


Reblog this post [with Zemanta]
#VC & Technology

Comments (Archived):

  1. bfeld

    I bet you’d be surprised by the very large number of readers that you have on the Mac who read your blog in NetNewsWire. It continues to be ridiculously popular, but doesn’t show up in stats like this.

    1. fredwilson

      feedburner says i’ve got 183 subscribers via NetNewsWire. the top five RSS readers for this community are Google Reader, FriendFeed, Bloglines, Netvibes, and Newsgator Online. but those numbers are just for subscribes. i’d love to see my reach broken out that way. there are 47k subs to this feed, but the daily reach hovers around 4k (less than 10pnct). I bet most of those bloglines subs are long gone.

      1. Jamie Lin

        I don’t think Google Analytics includes RSS feed downloads in its daily visits numbers. I usually read your blog from Google Reader and would only click over if I want to leave a comment. So if many RSS client users are like me, then the 4k number might be much lower than your actual daily readership.

        1. Henry Yates

          agree – I do the same from netvibes from my mac

        2. fredwilson

          The 4k is my daily reach in RSS readers, not the web. Feedburner provides that stat

          1. Jamie Lin

            Ah! So only 15% of leading edge geeks read AVC with RSS readers? No wonder the technology’s future isn’t looking up.

          2. fredwilson

            No, my daily RSS reach is about 4k and my daily blog reach is about 5-6k so somewhere between 40pcnt and 45pcnt of my readers are via RSS

          3. Jamie Lin

            Right. That was TTM number. 45% sounds more like it. Go Geeks!

  2. Peter Kirn

    Right, but you have to be careful how much you read into this. There’s liable to be a number of Mac users who read in Windows at work, meaning part of what you’re seeing is skewed toward workday (lunch break … or, well, we HOPE a break) habits. I see that in my stats — like people reading from Solaris, clearly from an office workstation.

  3. paulhart

    Seeing as you tweet links to your blog posts, Tweetie is acting as a de facto feed reader. Otherwise I’d probably be using Google Reader to keep up to date.

    1. fredwilson

      i know, in addition to many other things, twitter and twitter clients are becoming a news reader

  4. gregorylent

    high mac percentages all around, aren’t they, compared to mac/pc general population %?

    1. fredwilson


  5. theschnaz

    I wonder if even more people arrive here via Twitter than what Google Analytics shows. If I click on your link through tweetdeck, how is that tracked? Twitter? Direct?People here are early adopters and many use Twitter clients, Twitter might be even more popular.

    1. fredwilson

      totally. direct visits have gone way up on this blog in the past six months. it was 20% in the first half of 2008 and it is 38% now. some of that 18% increase is twitter clients. if all of it is twitter clients (not likely) then twitter would be responsible for 25% of this blog’s traffic, way more than google search which is at 17% for the first half of the year.that’s the power of passed links at work.

      1. theschnaz

        interesting. @fraser and I did a little research and found this…”-only 20% of its traffic comes through the Twitter website; the other 80% (logically) comes from third-party programs on smartphones or computers. So if you’re looking at Twitter stats on your website, you’re probably underestimating that source of traffic by a factor of five;”…this makes me wonder (if possible), should passed links have their own source (in traffic reports.) a direct visitor and a visitor via a passed link is not the same thing and marketers would love to get stats on passed links.

        1. fredwilson

          There is so much that needs to be done in this area of analytics and will be done. There’s money behind those links and people need to know

  6. Phillip Baker

    Your community skews towards Mac users more than the Internet as a whole so I think it’s a fair bet that Twitter might skew as well. According to this link Mac OS X has a 9% market share whereas fractionally over 1/3 of visitors to your site are Mac users.http://marketshare.hitslink

  7. blipsman

    Could the 5% difference in Twitter users be because of iPhone users accessing Twitter? Or is iPhone data reported separately?

    1. fredwilson

      i don’t know. it’s a good question

    2. Mark Essel

      Was just thinking of that Benjamin :)i’m on my iPhone now, but later I’ll get here from my windows desktop. How am I counted?

  8. Kiran

    I guess this data has to be collected across multiple blogs with a cross-section of vc blogs, tech blogs etc.

  9. needcaffeine

    I’d be interested to see if you could break out *nix, Windows & Mac users. I use all 3, so I’d throw the #s off. Taking geeks/techies out the field, I’d say Windows users are less Internet savvy. People who’ve used computers long enough will tend to move to *nix or Mac; after having gotten too frustrated with Windows.

    1. fredwilson

      I was going to say something like that in my post but thought better of it!

      1. billc124

        I disagree with this. I am very internet savvy and use Windows everyday both at home and work. I own a Mac, but it is part of my home theater system and not used for anything but A/V content. I work in the IT industry at a help desk company and all our stuff is Windows based. I have used Ubuntu and countless other flavors of Linux and talk about frustrating, I just don’t care to invest the time to get all my devices, etc working under Linux, it just isn’t worth the hassle.There are tons of internet savvy Windows users, trust me.

        1. fredwilson

          Of course there are. But as a percentage of each user base, I think mac users trend more savvy and windows users

          1. billc124

            I find that to be totally opposite in my group. My friends that own Macs got them because it was too hard for them to figure out how to do stuff in Windows so they got something easier to use. They are not computer savvy at all and they like the Mac because they can figure things out easier. I know, because my tech support help calls from them went from once or twice a week to not at all. I guess it depends on the circles you run in.

  10. Cheap Ed Hardy

    I like Windows so much ….So i will follow the Windows!!!!!!!!

  11. disqus_v4XrgpXgPt

    Am not sure if the two OS’s have any influence in how people discover information, except for availability of the software used to discover information. Many people using Twitter, do it via Client app’s, so does that show up in the log files?I believe very strongly on the power of passed links, only because I could observe that in my usage pattern. My usage pattern has shifted from seeking information via web, rss etc to discovering that through my ‘trusted’ network. My usage of RSS reader has dropped by over 60% versus usage of twitter & fb. What twitter should do is provide every user a ‘trends’ info which tells me how many links I have clicked from twitter. That would be interesting to track.

    1. fredwilson

      It sure would

  12. Josh Fingold

    Internet Explorer’s default is to subscribe to feeds using the browser as the reader program. There is also extended support of these feeds in Outlook. Perhaps that is the reason for the higher percentage of direct traffic from Windows users.

    1. fredwilson

      That’s an interesting hypothesis

  13. Maurice

    I think your assumptions about a techier-than-average audience using macs, and also using newer services like Twitter and Friendfeed are probably right. It would be interesting (and also significantly more complicating) to add in some mobile operating systems. iPhone would certainly dominate, and likely lean more heavily toward new services.The real question is…. what does all this mean for the future of the mass market OS? And, for media consumption services like google search, twitter, etc.

    1. fredwilson

      That’s a great question and its a lot of what I think about and blog about. Do you have any thoughts on the answer?

      1. Jens

        If twice as many Mac users use a new app as PC users do, even among an early adopter audience, does this mean that developers of new apps should be targeting Macs first?Does it cost 50% to acquire a Mac user compared to a PC user?

  14. Jamie Lin

    Another interesting comparison is between IE users and Firefox users. At our properties, we found FFers on average spend 2x the time on site than IE users. If you further segment it, you find FF+Mac users are the top power users followed closely by FF+Win users.

    1. fredwilson

      That’s me FF+macI wonder how chrome+win comparws to FF+mac?

      1. Jamie Lin

        We don’t see much difference between Chrome+Win and FF+Win, in terms of time on site. But then again Sosauce is a travel site and it makes some sense if Chrome users are not more interested in travel than FF users.

  15. Norrin Radd

    A few months ago to check users, browsers and OSs using different microblogging UI I obtained amazing results. Doing analysis about user statistics. reading your post I decided to share that experience. I’m sure that you or other readers will check this targeting your sites or doing same tests.I planned a two phase test: first phase push the same post over two different kind of systems with completely different UI ( and, second phase pushing posts targeting users from their Browser/OS preference.Before starting the tests I verified that all posts was listed in google/yahoo/msn results. And looking at Google Analytics I realized that the user of IE preferred to visit Tumbler (74% IE users) insted Posterous (76% NO IE users and 60% NO WINOS users). OS and Browser push users to choose different UI and have feeling with different websites.At that point I started to target posts with contents and keywords to see how users react and in that way I realized an interesting phenomena: users seems like fans! You can push away or obtain different type of users working also about their deep feeling. Working on their day by day or spare time passions or tools. Like the browser, OS and so on.In the same way that keywords can drive traffic to your post or website it will be possible to drive traffic working on professional/personal preference in using PC stuff.Kind RegardsRenato

  16. Scott

    I wonder what percentage of “direct” visits are actually referrer-less visits from referred sources?For example, links sent in e-mail messages, instant messages and even twitter desktop apps would probably show up as “direct” visits even though they were passed links. But since those links generally don’t appear as anchor text on a web page (many exceptions such as webmail, meebo IMs, etc.) the referrer is blank. Probably wouldn’t change the numbers too much, but I see this as an area where referrer credit needs to be established, eventually.

    1. Scott

      Actually, I take that back. and other URL shortners would imply with a good probability that referrals came from twitter and other passed links, but i’m not sure how Google Analytics handles those without creating a campaign a la:

  17. Scott

    Great point about Mac users more likely to try new services.I think the overall mentality of having a Mac makes one less hesitant to install new software. Compared with PC software, Mac software generally has less virus/trojan risk, is easier to uninstall, has less risk of a blue screen of death and won’t slow your computer down as much. So even though web services generally have the same benefits/risks on the Mac as the PC, the application mentality makes trying new web services on a Mac less intimidating.Also, my blog reading methods haven’t changed since switching from PC to Mac (for 80% of my computing) a year ago. But I do feel far more comfortable trying new software / services.

  18. Z

    have you considered that many people may have windows at office and mac at home? it may not make sense to completely segment by “who” instead of by “when”. especially when were only talking about 10%

    1. Michael Altschul

      Agreed. I’d love to see a similar report run based upon time of day AND OS. I fall into the Windows/Chrome daytime camp, Mac/Safari evening camp.Regardless, your audience clearly shows a preference for Mac (relative to general population), who show a preference for passed (vs. sought) links. Conclusion: when I’m on my Mac, it tells me what to do; when I’m on my PC, I tell it what to do.

    2. Mike Altschul

      Agreed. I’d love to see a similar report run based upon time of day AND OS. I fall into the Windows/Chrome daytime camp, Mac/Safari evening camp.Regardless, your audience clearly shows a preference for Mac (relative to general population), who show a preference for passed (vs. sought) links. Conclusion: when I’m on my Mac, it tells me what to do; when I’m on my PC, I tell it what to do.

  19. Z

    have you considered that many people may have windows at office and mac at home? it may not make sense to completely segment by “who” instead of by “when”. especially when were only talking about 10%

    1. fredwilson

      I have that setup actually and my habits are roughly the same on both machines

  20. Mark Essel

    Can you add a link contractor to get separate analytics, obviously asking readers to click it may not succeed. Feels like this type of info should be a known quantitiy to bloggers.

  21. Mark Essel

    Can you add a link contractor to get separate analytics, obviously asking readers to click it may not succeed. Feels like this type of info should be a known quantitiy to bloggers.

  22. sdelmont

    What if you exclude “search” from the numbers? I’m sure many of those visits are “accidental” searches from the general universe of internet users, including grandmothers, kids, etc… that population tends to have higher Windows market share, hence skewing the rest of the segments down.

    1. fredwilson

      Search is broken out

  23. Carl Rahn Griffith

    In the corporate world you will typically find users to be enforced to use WIndows machines and by definition more ‘locked down’ regarding what they can access online and when.Mac users will often be freelance/independent or if in a corporate with a Mac culture working within a much more flexible business environment and so have no restrictions, or ones much less draconian.Just a theory.

    1. fredwilson

      A good one

  24. BmoreWire

    I’d also be curious to see which users are at startups vs big companies vs independant and how the mac vs windows users break out by that and then by source. Using that you might be able to segment your audience by startupers who are looking to sell or get their product used by a big company and big company users and influencers who will be able to use, influence, or buy a startup.

  25. Jan Schultink

    The totals are stunning, giving the average Mac market share. 50% Mac in your case. Take out Google search (people at work trying to find stuff and getting to archived posts) and the difference becomes even less.Not sure whether you can draw conclusions about the average Mac or PC user. What you can say is that people who are interested in things FW is writing about use Macs..It also shows that startups with a service aimed at “geeks”/early adopters need to make sure things work on a Mac to get initial market traction. 50%….Thought experiment: what type of sites would be “all Windows” 🙂

    1. RichardF

      Jan, that is a great point about making sure things work on a Mac/Safari from an early adopter perspective.

  26. Andrew Warner

    They even make decisions differently.When I interviewed the founder of Magnify360, which helps increase conversions, he told me he could increase a landing page’s overall conversion rate by showing a different page to Mac & PC users.

    1. fredwilson

      I believe it

  27. ShanaC

    The real question seems to be the following. If I did a Big Five personality test on various people, what would their tech habits look like, and how could they be shifted?It’s not just apple/pc or how one reads- its a whole spectrum of behaviors that seem to be based on a whole slew of things- personality and culture. So you want an in depth personality test and a cultural read of people in order to figure out their tech habits. Who wants to be first? Let’s see how levels of Openess concur with their tech usage and likelyhood to use a Mac or twitter with a client?

  28. sgauria

    I am surprised no one is slicing the data by price. The avg mac costs 2x of the avg PC. From your stats, I would conclude that your blog audience is pretty high-end, which is not surprising given the VC focus. I would be very wary of extrapolating that data to say that “geeks”/early adopters are likely to be on a PC to the same extent.

    1. Carl Rahn Griffith

      I’m on a Mac – I wish i was ‘high end’ …. ;-)The demographics implications are pretty fascinating but I think a lot of it will need to be conjecture.I do know that in our village Macs are virtually unknown, Windows being common-fare, and Netbooks are becoming incredibly popular – Chrome focused on the sub-laptop (ie, Netbook) market could be a very astute move indeed.

      1. ShanaC

        It is why I think it is a personality thing. What types of people despite price are buying Macs.

  29. abarkett

    Re: the assertion that “Mac users are more likely to jump on new services like Twitter and FriendFeed more quickly than Windows users.”This is probably true, but it may have nothing to do with the Mac platform, or even really with the type of people. Mac people may be quicker to jump onto ANY trend. That wouldn’t surprise me at all, because they tend to be younger and more image conscious. The really interesting question would be, “Do Mac users tend to jump on trendy things like Twitter MORE quickly than they generally jump on everything?” For example, I bet Mac users jumped on the iPhone bandwagon WAY faster than they did onto the twitter bandwagon. That difference (difference in adoption rate of iPhone vs. twitter BY mac-heads) is actually the more interesting thing to look at.

  30. stuffaboutlife

    I have also noticed that the Mac user market tends to be an ‘early adopter’ market for many things. There are also quite a few very well done Twitter clients for Mac OS X that have spilled over to the iPhone.Did you happen to get stats for other operating systems as well? I’m wondering where the Linux visits were coming from, if there were any.

    1. fredwilson

      I can do that but didn’t

  31. Phil H

    News just in: People outside Barnes & Noble are more stupid than people inside!The above is as true as your statement. Don’t say ‘Windows users this’ or ‘Mac users that’. Generalisations are pointless when you are looking at a statistical sample. Instead: ‘a greater proportion of Mac users are early adopters’. It may well be that the majority of early adopters are Windows users, but you aren’t looking at it that way, and you could easily mislead.You ask if twitter’s users are more Mac than Windows, but 23.5k is via Mac+twitter, and 24.4k is PC+twitter. So, yes they are more even than usual, but twitter is probably still a majority PC crowd.Sorry to be a stats Nazi, greatly enjoying the blog.One thing that could be interesting to see is what proportion of readers read at work (as I’m doing) or at home. The vast majority of business machines are PCs, so you may find that you have an even higher proportion of Macheads than you think, just that they prefer to read at work. Me? I’m a Linux, but work machines are all Windows.

    1. fredwilson

      I think you are right about barnes and noble 😉

  32. Perry

    You wouldn’t be surprised by this because you’ve been at this so long, but I’m surprised at the total ratio of Macs to Windows. If the total ratio of Macs to PC’s is roughly 1/20 +/-, Mac users would seem to be much more curious about topics related to VC stuff.

    1. fredwilson

      I think its that geeks prefer macs and they also are interested in the stuff I write about

  33. tokerud

    I think it is the other way around. The forward-looking computer professionals and enthusiasts use Macs. Has been true for a long time but now the Mac has a higher market share and runs Windows so that effect is magnified. Why? Partly because if you care a lot about computing and internet, you want the best tools available.

  34. paramendra

    I think most of your readers are techie types. You don’t get too much general readership. That explains, I think.

  35. Stephen Purpura

    Excluding the obvious that you have (in aggregate) more of something, it’s tough to draw any conclusions about individual traits of people that also happen to be reading your blog from this information. If you want suggestions about how to better understand the microdemographics, send me email.

  36. Lovely Rose Quintos

    Based on your statistics,the two topic or keyword were the top visited sites everyday.Random sites that talks about windows and mac.I think people are always excited for the updates regarding windows and mac.