APLawrence.com -  Resources for Unix and Linux Systems, Bloggers and the self-employed

How many RSS readers do you really have?


2007/11/24

Website owners take a great interest in statistics: how many hits did my website get this month? How many unique visitors? How many RSS readers?

That last figure (RSS readers) can be very mysterious, because you can't really tell what it means. For example, I sometimes subscribe to a site and only after months of ignoring it in my RSS reader do I get around to getting rid of it. In the mean time, I'm counted as one of that's sites "readers", even though I may never have read anything since the day I subscribed.

So why care about it at all? Well, I think if the count is growing or shrinking, that definitely indicates something. If growing, that certainly means that readers cared enough about your content to make the effort to subscribe: there are very few subscriptions that do not involve a real human being making a choice. On the other hand, if your RSS counts are shrinking.. well, you must be doing something pretty bad, right?

But there is a little more insight to be gleaned from RSS stats. I wrote a little Perl script today that I'll post here; it analyzes RSS visitors (that is, sites that did a "GET" of one or more RSS files) and figures out how many of them also did a "GET" of other pages.

This is very imperfect. For one thing, I have my RSS feeds syndicated through FeedBurner, so it's quite possible for someone to go direct to them for my feeds and never visit the site at all - if one of them did visit, I wouldn't know that they are a RSS reader.

Secondly, I'm not even trying to account for new subscribers: that is, someone who read an article or two and then did a "GET" of an RSS file. Technically, that overstates visitors who came from RSS reading, but since the script also radically under counts (by being blissfully unaware of FeedBurner subscribers), I imagine that balances out. More importantly, this just can't be treated as absolute truth: it's an impression of what's going on and it probably means much more when looked at as a trend rather than trying to extract meaning from individual statistics.

With all that in mind, here's the code:

#!/usr/bin/perl
%counts=();
%rss=();
# I had to break my log file into sections because of its size
# I just used Unix/Linux "split -l 4000"  to create "xaa" through "xru"
foreach(<x*>) {

open(X,$_);
foreach $line (<X>) {
    $w = "(.+?)";
    $line =~ m/^$w $w $w \[.*] "$w $w $w" $w $w "$w" /;
    $site     = $1;
    $file     = $5;
next if ($file !~ /.html$/ and $file !~ /.rss$/);
# counting uniques only, this is vistors not pageviews
$counts{$site}=1 if $file !~ /.rss$/;
$rss{$site}=1 if $file =~ /.rss$/;
close X;}
}
showme();

sub showme {
$visitors=0;
$rssvisitors=0;
$notrss=0;
$nevercame=0;
foreach (keys %counts) {
  # $_ will be the "key" - the visiting site
  $visitors++;
  $notrss++ if not $rss{$_};
}
foreach (keys %rss) {
  # $_ will be the "key" - the visiting site
  $rssvisitors++;
  $nevercame++ if not $counts{$_};
}
printf "Visitors: %s\n",commify($visitors);
printf "Not from RSS: %s\n",commify($notrss);
printf "From RSS: %s\n",commify($visitors - $notrss);
printf "\nRSS Visitors:  %s\n",commify($rssvisitors);
printf "Never came to site:  %s\n",commify($nevercame);
printf "Came to site:  %s\n",commify($rssvisitors - $nevercame);
print "=================\n";
}

sub commify {
        my $text=reverse $_[0];
        $text=~ s/(\d\d\d)(?=\d)(?!\d*\.)/$1,/g;
        return scalar reverse $text;
}
 

That's it. When I ran it earlier today, I got these results:

Visitors: 131,446
Not from RSS: 131,186
From RSS: 260

RSS Visitors:  954
Never came to site:  694
Came to site:  260
=================
 

And what does that mean? Ahh, that is the question, isn't it? It means I still get quite a few non-FeedBurner direct RSS visits (all get redirected to FeedBurner but still hit my log), and it means that about 25% of them do also come here to read. I can't tell how many visitors came from reading FeedBurner feeds that they get directly from that source, however. If I wanted to expend a great deal more effort I could also analyze the "referrer" data in the logs and look for evidence of Newsreaders, but that's completely inaccurate also. No, as I said, this is just something else to look at in conjunction with everything else. By itself, it is nearly meaningless.

For those who wonder why the Feedburner stats in the left side bar show less than the 954 counted here, well, there are several reasons. First, Feedburner has its own way of counting readers on any given day, but also not everyone who grabs feeds from here uses Feedburner. Also, because I also have sub-feeds like "Linux only" and so on, those don't show in those Feedburner stats at all. Those wouldn't add a great deal, but it all adds up.



Got something to add? Send me email.





(OLDER)    <- More Stuff -> (NEWER)    (NEWEST)   

Printer Friendly Version

-> -> How many RSS readers do you really have?




Increase ad revenue 50-250% with Ezoic


More Articles by

Find me on Google+

© Anthony Lawrence



Kerio Samepage


Have you tried Searching this site?

Unix/Linux/Mac OS X support by phone, email or on-site: Support Rates

This is a Unix/Linux resource website. It contains technical articles about Unix, Linux and general computing related subjects, opinion, news, help files, how-to's, tutorials and more.

Contact us





That's the thing about people who think they hate computers. What they really hate is lousy programmers. (Larry Niven)

I have been complimented many times and they always embarrass me; I always feel that they have not said enough. (Mark Twain)












This post tagged: