The State of the Seanachai

I thought I’d take a moment to type about the podcast and how it’s doing.

By the numbers.

My father is a very quantitative kind of guy. And I have sometimes had difficulty understanding that point of view. Any attempt to codify or quantify writing just seems to wreck it. (Hello Microsoft Grammar) But I shouldn’t beat on Microsoft, it’s been going on in English ever since the Chancery in Merrie Auld England. But recently, I’ve found great comfort numbers.

Especially this number – ‘209’

30,000 episodes in May.

The Seanachai transfered 209 gigs in May. At an average episode size of 5.9 meg that’s 34,711 episodes transfered. With a weighed average size (just the last 5 episodes) of 6.6 meg an episode that’s still 31,030!

Here’s what the traffic looks like since we launched the new site.

(anybody want to guess about the mystery spike that happened right after I posted a Change in Orbit?)

So how many listeners is that?

No clue. Here the numbers become amibigous. The numbers only show who downloaded – not who actually listened. And of the people who downloaded and listened, how many episodes did they download? Just the two I posted this month – or did they go back and listen to more?

No clue. (Sure feels like numbers are turning on me once again.)

If you say only two episodes. 30,000/2 = 15,000 listeners. Let’s cut the episode number in half as a “fudge factor” and I’ve got 7,500 listeners. Or maybe each listener listens to 20 episodes a month for the first three months and then listens to them as they come out.

How do I find out? Maybe a survey? Seriously, all ideas are welcome.

5 replies on “The State of the Seanachai”

  1. Patrick,

    Somewhere on your site is probably a bunch of log files that tells you what someone downloaded and when. You can probably poke around in one and see the IP address, time, and what they downloaded. That might give you an idea of what’s going on. If you do it by hand I suggest looking in a log file where not very much is going on. That way you can find the logs for the same person closer together.

    If you had a spare code monkey lying around, you might give it the access logs to see what patterns fall out. It could filter out all the downloads, then count how many downloads per unique address. It might even tell you if the downloads were a few seconds apart (catching up) or several minutes (savoring each one). Just for curiosity’s sake, you could send it logs from 3 or 4 days to get a good sample.

    Congrats on your upcoming book! I can still hear your voice, though, when I read the lines.

    Joe

  2. PS. You can only call them ‘code monkies’ if the aren’t around. Otherwise you have to call them ‘code ninjas’.

    Joe

  3. A couple of ideas:

    1. Not definitive (but maybe more likely to get a response than a survey; folks seem to have an aversion to them). In some podcast I heard awhile back, possibly Rob at PC411 interpreting Feedburner-published stats, I think I heard that most podcast listening is still done at the pc rather than on ipods et al.

    My suggestion: post an mp4 version. Give a verbal buildup to this “special experiment”. Relate some of this info and your curiosity. A graphic shows up saying “click the button to let me know you’re here. Click only once”.

    If they’re listening on the go, encourage them to visit the site and click the button there.

    2a Post a chapter of How to Succeed in Evil either password protected (MightySeek’s WordPress addon–PodPress? makes this easy to do).

    b. Force anyone who wants the chapter to let you know we’re out here. Maybe a couple questions, but DON’T make it scary like a survey) We then get the password either at your site or via autoresponder. I for one would be happy to “register” like this.

    c. Maybe offer a bonus story too. I’m so keen to get all your output that, again, I’d gladly register.

    That’s my instant response. May come up with more later.

    BB

  4. The idea is to do something that results in as-near-as-possible 100% participation by listeners. Most people don’t care enough to take action. So you’ve gotta be firm/agressive and the response must be super-easy for listeners to do.

    ie they’ve gotta be strongly motivated even in order to do not much.

    ttfn

    BB

  5. There’s a program called “wusage” that can answer a lot of those questions, if you’re willing to sell your soul to set it up.

    Or, I think Bob, down the hall, has a something for you …

Comments are closed.