Feedburner analysis to Determine Real # of Subscribers
by blogrdocI’m a web newbie, but I’m no stranger to interpreting data. If someone who *knows better* can tell me that what I’m about to say is complete BS, please let me know. (I should add that this goes for all my posts.)
One thing I noticed about feedburner is that you can’t trust the number of subscribers that it reports. So I setup a fake website with no content and let feedburner track it. Sure enough, it reports a dozen or so ’subscribers’. To determine my real number of subscribers, I simply subtract this number from my # of subscribers.
So here is the “equation”:
# of real subscribers = (# subscribers of real blog) - (# subscribers of fake blog)
This is probably only applicable/useful for small-time bloggers like me. If you’ve got hundreds or more subscribers, you obviously don’t need to worry about this.
I should add that this is an application of my theory of applying variation where it counts. In engineering/scientific experimentation, we call these ’splits’.
So, if I wanted to be fancy, I’d say,
I ran splits on my feedburner feeds to determine the “signal” from my blog.
In the life-sciences, they call these ‘treatments’.
Popularity: 12% [?]
2 Responses to “Feedburner analysis to Determine Real # of Subscribers”
By Alik on Mar 21, 2008 | Reply
you might have not taken into account the “subscribers” who are actually content scrappers - there are plenty, they suck up your content through RSS and show it under different domain name adding AdSense - hoping someone will subscribe to theirs and they will get free $$ “reusing” your content for free. Once you become A-List blogger you may consider complaining to Google here:
https://www.google.com/adsense/support/bin/request.py?search_ask=1&subtopic=&contact=rpv&main_topic=other&contact_type=11&contact_topic=Report+a+policy+violation&Action.Search=Continue
They have this section:
“This site is distributing my copyrighted material without my permission. “
By blogrdoc on Mar 21, 2008 | Reply
Agreed. My ‘fake’ website actually looks a lot like this one, so the non-descriminating eye might also try to follow the fake one, in which case, the equation is correct. However, obviously - it is possible that those scoundrels have made a distinction. The equation still would provides a 1st order approximation and is very simple to calculate.