Google Reader Stats are Bulls**t (With Proof)
Google Reader stats, in case you don't know, are bulls**t. In fact, all Feedburner stats for most top blogs are bulls**t due to the effect of default feeds. Want 80,000 free subscribers? How about 200K or more? Read on.
Context & Ripple Effects
The argument lands one day after TechCrunch published its ranking of top blogs by Google Reader subscribers — a leaderboard built entirely on the numbers this piece calls inflated. The mechanism, per the article, is default feeds: readers that ship subscribed to popular blogs out of the box, padding Feedburner counts with people who never chose to follow.
It also reopens an old wound. Jeff Jarvis was complaining about publishers not controlling their own audience numbers back in his 2005 post demanding ownership of reader data, and Google's own Webmaster Central blog ran a September explainer on how it counts subscribers — meaning the metric's methodology was already under discussion before this takedown.
First-order effects
- Any blog-ranking or comparison built on raw Google Reader subscriber counts — including the TechCrunch list published October 14 — is skewed toward blogs bundled as defaults, misstating who actually has the bigger audience.
Second-order effects
- Feedburner's subscriber number is the de facto currency for blog advertising and acquisition talk; if top-blog counts carry hundreds of thousands of phantom readers, advertisers and buyers pricing off those figures are paying for reach that doesn't exist.
Third-order effects
- If default-feed inflation goes unaddressed, RSS subscriber counts lose credibility as an audience metric, pushing publishers and ad buyers toward click-level or server-side measurement instead of reader-reported tallies.
The trend: Feed-reader subscriber counts are decaying as a trustworthy measure of blog audience size, as bundled default feeds inflate the numbers that rankings and ad deals rely on.