My Europodcast two cents
Om is stoking the controversy about podcasts, centering on this report from Forrester. Here's some data from the terra incognita known as "not North America" to add, for what it's worth. Look at the statistics (page 4) from the BBC's podcast trials, which show, among other things, that the 8:10 interview on Radio 4's Today show got 330k downloads in February, I presume mostly from within the UK or UK ex-pats around the world, as the content is typically domestic and frequently political in focus. Then again, it's a daily feature, so it's unclear what the unique user figures might be. The other thing that's noticeable in looking at the detailed figures published in the last three months is the volatility in downloads of individual programs. Only In Our Time, Sportsweek, The Weekender, and Letter to Gaelic Learners have seen two consecutive months of growth in downloads, so the momentum is not evenly spread by any means.
Still, looking at the total download activity in February (1.73m), average downloads per day were up 2.3% month-on-month, following 5.6% growth in January. Again, it's very difficult to get a grip on the listenership split between UK and non-UK (I'm sure the BBC has logged all the data, but it's not public), but for some of this stuff I would think the users would overwhelmingly be UK residents or nationals. I guess initially we could be tempted to say that the user base looks like it might be pretty small, but has a respectable growth rate.
Maybe it's more relevant to look at the addressable potential converts to podcasts, i.e., the users who are already doing on-demand listening to BBC programming. I have been tracking this data, which runs from October 2003 to present, and it shows some interesting things. In February, all BBC radio channels collectively had 11.9m on-demand listening sessions (+32% YoY, +167% since Oct. '03) from a pool of 8.8m "unique user agents" (a fancy way of saying that we don't know how many individuals are represented by a single cookie, though I guess RFID might fix that in future), logging 6.3m hours of on-demand listening (+42% YoY, +244% since Oct. '03).
That's about 44 minutes of on-demand listening per day (normalized) per unique user agent, which is up 14% YoY and 47.6% since November 2003. In contrast, average live listening per day is higher, at 75 minutes, but it's only up 11% since November 2003. So, unsurprisingly, there's an apparent shift to on-demand listening (now 37% of total online listening) and away from live, and the number of podcast downloads is at a level which is already equal to 16% of on-demand listening sessions per month. If we remove the mystique of "podcasting" and view it as just another mechanism for a behavior which is already widespread and gaining momentum, perhaps its potential as a force to be reckoned with becomes more visible. And of course, if doesn't hurt that the BBC is promoting the hell out of it on radio.