Email clients usage statistics

[hadaso]

I wonder if mail providers that are represented here have some info on email client usage statistics that they be willing to share with us.

[bramhall]

Do you mean something like this, which lists "top 10" only and gives percentages, not absolute numbers?

[George_B]

I was about to post something similar, except that we see Outlook having an even bigger share. Outlook 2003 is really bad though, it has all kinds of limitations and throws very confusing error messages. It's the IE6 of e-mail clients

[hadaso]

Originally Posted by bramhall

Do you mean something like this, which lists "top 10" only and gives percentages, not absolute numbers?

Well yes (I saw that one), but I hoped for something that doesn't put Outlook Express and Outlook in the same number (they are not the same client).

[janusz]

...there is enough information available to make some clear statements about the popularity of different services.

The big three are (in descending order of size) Microsoft's Hotmail (now called Windows Live Hotmail), Yahoo! Mail and Gmail.

Despite what many people think, the number of active Gmail accounts still falls short of either of its two largest competitors.

Full article.

[janusz]

... and more statistics on email clients proper....

[n5bb]

That only shows which email clients or webmail systems don't block web bugs (embedded tracking images) by default. And which don't have good enough spam filtering to reject those messages before the user saw them. And it only counts messages which were opened (and not just ignored). And it's biased by the demographics of the mailing lists used by their users. For example, it may ignore all email users under the age of 21.

So in my opinion that site provides an imperfect measure of what their users find in their advertising campaigns. It may be dramatically different from what one particular email service provider finds for their general customers.

Bill

[hadaso]

Originally Posted by n5bb

... which email clients or webmail systems don't block web bugs ... only counts messages which were opened... And it's biased by the demographics ...

... imperfect measure ... It may be dramatically different from what one particular email service provider finds for their general customers.

These are all very good points. I was thinking of info an email provider can obtain by (automatically) examining headers in incoming mail, that can reveal the user agent, things like:

X-MimeOLE: Produced By Microsoft Exchange V6.5

But then I am probably more interested in demographically biased information, such as what user agents are used by people who send email to me, so perhaps a better question is how I can perform such statistics on my own mailboxes. The reason I was thinking about this is that emails sent from different systems at the Open University (Israel) seem to be displayed as intended on Outlook and nothing else (well, at least in FastMail some of them are next to unintelligible, just because FastMail sticks to mail standards. Also in Gmail many of them are quite unreadable, and Gmail seems to be very popular with my students, though it's only a feeling, I haven't counted, and I never checked how many of my students use their Gmail account through IMAP on Outlook. Of course Outlook is the client installed on every PC at out offices). So I was wondering how many of our students are using Outlook (not Outlook Express), and that got me interested in any kind of statistics about email client use.

[Berenburger]

Originally Posted by janusz

Full article.

This is cool! (html5 browser needed.)

[elvey]

Originally Posted by n5bb

That only shows which email clients or webmail systems don't block web bugs

That's what I was thinking.

Plus, those numbers don't pass the smell test, IMO.

I didn't look at how the 'Fingerprint' tech works, but my Thunderbird install is configured to block everything but IMAP and SMTP to mail servers by default, so fingerprinting that doesn't look at the mail I send is very unlikely to work.