Some of you might have noticed I set up my clicktracking script to track clicks on feeds. If you click a link to get my rss feed, for instance, then a little statcounter tells me someone [anonymous] out there clicked the link to subscribe to my blog. Yay! Thanks :). From a business perspective though, this solution is way less-than-optimal.
First, it doesn’t correspond well with actual feed usage. It seems that many people add RSS feeds to their readers … then proceed to ignore said feeds. I have to admit I’m as guilty as anyone, as I dislike using my clunky RSS reader (Feedreader.com). Whatever the reason though, there’s an enormous disparity between clicks – indicating intent to subscribe – and action – people actually accessing the feed within their readers once they’ve subscribed.
How do I know? I compared my stats with Linda Bustos of Get Elastic Ecommerce Blog after seeing that I’d had a few thousand people click my RSS subscription links. Turned out they were getting multiple times the traffic, while having approximately the same amount of suscribers.
While I could attribute that to other factors, like superior use of social media, I don’t think that’s the answer. I don’t have raw numbers to back it up, admittedly, but I’m pretty certain that the difference is at least partly related to subscriber base. They probably had multiple times more clicks on their subscribe links, so that the actual usage numbers of their feeds came out to several thousand – not the intent numbers, which must have been far higher.
Second, click-tracking sucks because it only gives me intent numbers, not usage numbers. And while that’s actionable to an extent – I can tell what content I might develop follow-ons for or what categories are better at gaining new subscribers – it still leaves me flying partly blind as to what folks read.
In theory, the categories should correspond between intent and usage, but who knows. Some categories may be skewed towards folks who subscribe a lot but read a little.
Another reason I’m at a disadvantage as a result of not having usage numbers is that those are the de facto standard for tracking subscribers. Therefore, trying to sell this site might be that much more challenging (let alone finding a buyer who’ll value the tens of thousands of dollars of links I’ve earned this site…). And it also limits the social proof I can use, since I have no Feedburner badge.
I’ve yet to see anyone else raise this issue of a difference between intent and usage, for RSS feeds. In the email world, it’s subscriptions vs opens. My understanding is that open rates vary from 5% to 30%, on average. If I had to guess, I’d say my open rate is on the low end of that.
(My memory might be faulty, so on that note, go buy yourself a copy of Sitepoint’s Email Marketing Kit yourself – it’s terrific and worth the investment in your future education. And no, that’s not an affiliate link. Note: Like other Sitepoint Kits, it’s geared more towards beginners, not seasoned vets.
Do any of you know alternatives to using Feedburner or Google Analytics to track subscriptions? Massive dofollow link love, a free 1 hour consultation (worth $200) plus an hour of linkbuilding to anyone who recommends an adequate solution to get me equivalent data without sharing more of my info with the Borgle.
Update: Willy Franzen – whose entry level jobs site you should check out – has made some excellent headway on solving this:
“Use a WP plug-in that allows you to have content that is only shown in the feed. Then add a unique tracking image to every post. Your server logs will show all of the times the image is downloaded.It’s the same way that e-mails track open rates, I believe.Make sure that you disable hotlinking of images, so that if your feed gets scraped your stats don’t get screwed up.”
The next issue to resolve is to adapt the pixel tracking solution to the various feedreaders out there, since some of them request the feed (and tracking pixel) whenever a user does, while others request it once and serve it to all their users. So for those feedreaders, all the users would appear as a single subscriber. How can that be worked around?
p.s. Hat tip to Avinash Kaushik for helping me resolve this difference between clicks/usage.
Oh, and of course, if you liked this post on analytics – get my rss feed, pronto ;).Tags: Analytics, RSS, Subscribers