Chris Cree spotted today that Friendfeed subscribers are now counted towards Feedburner stats.
It can make quite a striking difference with Feedburner if you have a few followers there.
But even this doesn’t really account for the shifting sands in online attention.
The latter half of this post was originally published Mar 21, 2008 @ 20:38
Since then Twitter has for many people emerged as the primary way they read RSS feeds, combined with various forms of lifestreaming.
The first time I see tweets and blog posts often is also on services such as Blogcatalog’s dashboard or even Mybloglog (though that can sometimes lag a little on updates these days)
Current Calculation Problems
- Blogcatalog & Mybloglog numbers are just as relevant as Friendfeed
- Twitter numbers are probably more relevant than any Lifestreaming service
- Facebook subscribers are still not counted
- Aweber & Feedblitz, along with Feedburners own RSS to Email service are included, but they are the only ones I know about. Where is the Getresponse support Simon? Infusionsoft should really offer something as well, though they don’t offer RSS to email – I am not sure about Mailchimp
The onus really is on the developers of these other platforms to report numbers to Feedburner, but I have no idea how that can be managed with Facebook and Twitter.
With both Facebook and Twitter there are 2 significant problems
- The data doesn’t always come from an RSS feed – individual Facebook apps programmers use various methods to pull data, and often Facebook items come from other sources. The source on Twitter could be any one of the many RSS to Twitter services, but equally could be a WordPress plugin.
- Items get shared – shared items in Google reader have often affected Feedburner data in the past, how does this work with retweets?
It is good to see services like Postrank now feature within Feedburner stats, as they provide various ways to filter RSS content on multiple topics, and then include only the best items for you to read.
(Niche marketers will probably find a way to make best use of that)
Note: I do have specific strategic reasons why I don’t currently display any RSS subscription options, or even an email subscription box.
This post used to have lots of comments, but Disqus hasn’t managed to sync them after 2 days.
It is amazing how long it takes for things to catch up, the following was written over 15 months ago, and the numbers are really still totally inaccurate
Originally published Mar 21, 2008 @ 20:38
I have been digging around in my Feedburner stats to see how various social streaming and life streaming applications I use are reporting data to Feedburner.
Currently reports as:-
Section:- Feedreader & Aggregator
Mozilla/5.0 (compatible; FriendFeedBot/0.1; +http://friendfeed.com/about/bot)
Currently reports as:-
I am not sure which Yahoo service they are reporting as, so I am listing a few possibilities
Currently reports as:-
Section:- Feed Readers and Aggregators
A web-based newsreader that allows you to select and manage RSS headlines within a My Yahoo! account.
There is also Yahoo! Slurp and Yahoo Test Bot – both listed as bots
Is Lifestreaming Subscribing?
Here are some reasons Lifestreaming should count as a subscription
- Personal – with most applications views can be attributed to individual users
- Selected – unlike meme trackers, someone has made a specific choice to read your content
- Trackable – if necessary it would be possible to identify only active users
- Traffic – traffic from lifestreaming is quite visible, though it is hard to determine if it comes from a RSS subscription link, or when someone tweets about you, or maybe from being Stumbled or dugg
- Email – Friendfeed sends subscriptions by email too – does that make it 2 subscriptions?
- Active – people are actually using these services more and more, and subscription data would thus be a useful representation
There are some aspects I am not sure should be counted, but are probably more valuable data than from many feedreaders
- Profile views – MyBlogLog, Blogcatalog and Friendfeed all allow visitors to view content before deciding to subscribe to it in some way – whist no long term commitment is made, a lot of this activity can be attributed to individual unique users, thus could be counted as a subscriber in some way.
- Shared Social Media Links – as mentioned before, when links to your site appear having been dugg, stumbled or shared in Google Reader – whilst this can result in traffic, it might not be something that can be counted as it is not necessarily related to the RSS feed, but to the permalink.
- MyBlogLog Topics – this is based upon their tagging system (I have wanted it to link to content for ages – make sure you update and cleanup your tags) – I don’t think it would be appropriate for this data to end up in RSS subscription stats
- BlogCatalog Social Search – also recently introduced and whilst it doesn’t have RSS yet (nudge Daniel) I can see this happening in the future – again this is a search much like you would have on Technorati or Google Blogsearch
Depending on how you use these sites, they can add or subtract to the total information overload you are subjecting yourself to. I am not sure whether my own usage patterns are typical, but I find I am using Social and lifestreaming more than RSS Readers.
I have 1000s of unread items in my RSS readers, though on a lifestreaming service I am not reading every item either.
If the conversation is moving away from blogs there needs to be a way to measure it, track it and possibly respond to it.
I still am not sure how to react to the new item sharing feature introduced today on Blogcatalog, where someone can share items to people following them on the Dashboard, and to their Shared widget, and leave a comment.
When I first heard that this was going to be coming just a few days ago, I immediately thought that I would be vocally against it, but it is like a Stumbleupon review or a Delicious bookmark – it is not trying to start a new conversation, just tell someone why you are sharing the link.
The problem is that people will only share content using a certain number of different methods. Isn’t it best to use the one that is most likely to be seen across multiple networks?
Disclaimer: I consult a little with Blogcatalog