Having listened to MIKE-FM (actually WMKK-FM, but their marketing department never ever actually uses that) for long enough, I’ve realized that they don’t actually have “less DJ time”. They just TiVo, or timeshift, their DJs.
From their perspective, it makes business sense. Instead of having several DJs on staff, they simply have a handful of voice talents on retainer. DJs have to be on during certain hours, since they’re generally live (allowing weather, news, traffic breaks with banter). Voice talents, in this modern age, can work anywhere and anytime, as long as the .wav or other files get delivered to the server before the broadcast. MIKE-FM seems to have two main voice talents (the main snarky guy and the older, goofy comic foil), and a couple of other occasional ones (some community ad and other longer spots).
The thing that gets me is that Mike doesn’t really seem to have less talk! I’d love to see someone do a time comparison of Mike vs. competitors for a certain timeslot. While Mike’s automated voice talent bits are typically pretty short, they come along very frequently – every other song, usually; and sometimes they have a couple of blurbs together. They have a fairly good repetoire of blurbs: they clearly record plenty of blurbs and then randomly stick them in. They even have some blurbs that are very topical: they clearly shuffle the voice lineup at least weekly. But they still have just about as much talk as other stations, overall. So while it may be a “Voice Actor” they hire, they’re still doing the same job as a DJ – just timeshifted, and without any of the actual human connection. Too bad.
If they could actually go the next step, it would be interesting to see some radio stations do that. Imagine the Web 2.0 or Semantic Web or whatever applied to this model. Timeshifted DJs, but where a music director actually choreographed entire sets of music (or styles of music) along with some semi-topic-specific blurbs. Scatter a handful of daily-topic blurbs throughout (daily voice talent recordings of the previous days scandals, for example), and it would feel pretty ‘live’. This needs a very rich semantic catalog of music (for automated style selection and matching), as well as improving current beatmatch technology (which I thought some radio station had a big proprietary way to do interestingly but I can’t find it right now) to really cross styles and be more accurate.
Although I still like my live DJs on some stations, please.