I think for those of us in new media, discussing your download stats has joined politics and money as the thing that makes you rude for bringing it up over the dinner table. The Ze Frank thing brought this to mind, but this is a lingering thought from PME. I find a high correlation between people who like to bring up their stats and people I find very tiresome very quickly. Not that I've ever met him personally but Rickey Gervais springs to mind in this regard with his interactions with our little world.

I only really see two reasons for this fascination - ego and money. Having lots of downloads is the way to prove your worth to others (easier to just tell someone a number than get them to download and like your show) and to increase your valuation when you try to cash in. Like Scoble said earlier and I've been saying for years the affinity that the users have for the show is far more important than the total downloads. Ze Frank accused Drew Baron of fudging his download numbers. None of this would matter so much if we could break the link between downloads and money. When those are tied together, you encourage fraud and jealousy.

Let's be honest here. If I could prove I had 100,000 downloads per episode and that would get me $X because of it, there is a certain rationality to seeing how much it would cost you to synthesize that many downloads. The same guys who run the spammer bot networks I'm sure could generate you actual downloads from different IP addresses that would pass any audit. If what they charge you is significantly less than what you make off it, those of a certain unethical bent could then jerk off with the invisible hand of the marketplace. Whatever the difference between those two dollar figures is, that's your margin. You could make money off a show without having a single legitimate fan yet still have verifiable and auditable stats and logs.

Over and over, you'll hear how we "really need a good way to determine actual audience metrics." Hell, that was half of the sessions at PME. I disagree completely -- we need the opposite. What we need is a way to all do our business realistically without needing them. Perhaps the pain of collecting these statistics is a clue that it is not what we should be doing. "Doctor, it hurts when I do this." "OK, don't do that."

My contention is that rather than "pay per impression" type advertising, only "pay per results" style is reasonable. Spammers and fraudsters can do a lot of things, but what they can't do reasonably is mint new money to dump into the system. If they could, they wouldn't need us. Thus, when GoDaddy has all their advertisers give out coupon codes, they use that to measure how much money flows through the system associated with the individualized codes. This is much better than relying on download stats, but has its own problems. For example, when GoDaddy sponsors a lot of different shows that you are a fan of and then you go to register a domain, which code do you use? Do you register multiple domains seperately so you can spread the love out? There is a chokepoint, and by nature the biggest shows will choke out the smaller ones in general. It's better than download metrics, but still not perfect.

When I've had sponsors, some of my fans wrote in and/or left comments on the product forums thanking them for the sponsorship of my show. I don't know how much that mattered in the cold hard economics of business decision making, but it left me with warm and fuzzy feelings. These are all tentative baby steps (not even, really) towards reaching that "affinity rating" that would be a far superior metric for deciding where to put sponsorship dollars into new media. I don't know how we'll get there, if we'll get there but I know that this is where I'd like to see the efforts going. When people say "We need to know who is listening and how to count them better" I tune them out and instead think how much better it would be if we could just measure how much people care. Both are difficult or perhaps impossible problems, but if you are going the tackle the impossible isn't it preferable to shoot for a much better impossible?