Echo Statistics - Where The Referers Are
It's been a while since I did some detailed Readership Analysis, so I went and checked over my logs to get some fresh data. I wanted to answer the question: "What's the best I've done recently at being heard, and where?".
For all of September and October, I had only one regular blog post which, again cumulative over the entire two months, registered over 1000 readers, specifically 1112 total unique IP address accesses. I know, I know, people will tell me not to complain, there are bloggers starving in Africa (or enslaved in Sudan). But this is my absolute personal best for over two months - and note it's less than one popular Slashdot comment, or the daily readership for some not even A-list.
The stand-out item was the BNETD / Blizzard entry, which concerned the horror of that court decision for fair use. The echoes broke down from major sources as follows:
Total = 1112 (excludes constant regular blog audience)
| lwn.net | 290 | 26.1% |
| unknown | 258 | 23.2% |
| freedom-to-tinker discussion | 102 | 9.2% |
| Wes Felter echo | 85 | 7.6% |
| Ernest Miller echo | 71 | 6.4% |
| Google searches on keywords (later) | 87 | 7.8% |
| weblog.foopee.org | 59 | 5.3% |
| Groklaw discussion mention | 46 | 4.1% |
| sethf.com (i.e. later blog entries) | 39 | 3.5% |
| digitalconsumer.org | 23 | 2.1% |
Not exactly a Slashdot effect, but instructive in the distribution.
No echoes == no audience.
By Seth Finkelstein | posted in statistics | on November 14, 2004 11:59 PM (Infothought permalink) | Followups
Keep in mind, though, that you're syndicated here and there via RSS. Most of the bloggers I read I read almost completely via RSS, so it's doubtful that that will show up in the same sort of readable fashion as a straight page hit.
Some needs to figure out how to meter RSS, I think.
Yes, of course. The RSS audience I'd call part of the "constant" audience, and I know that's around 300. Given that I run my own domain, instead of using a community service such as livejournal or blogspot, I can do very detailed log analysis. So I have direct access to the number of feed retrievals, for example, as well as the notes Bloglines puts in its referer string saying how many subscribers for a feed.
What I was doing here was specifically looking at the "spikes", the echoes. That is, how much audience I ever get beyond the core fans. It's relevant to know that the answer is "rarely" and "not a lot" (that may be depressing, but it doesn't change the numbers).
Note, again, the 300-reader estimate isn't perfect, it could be off by some margin. But it's not off by a factor of 10.