Hooolaaaa! I think that Clay has some good points. I'm not sure that there is a conclusive answer to them either, especially since Clay's arguments have been focussed on the quantitative aspects of the debate, whereas the responses (somewhat at cross-purposes) have been seemingly focussed on the qualitative.
I guess that, rather than trying to rebutt, I'd simply make a couple* observations:
1. GM describes and to some degree anticipates, a trend whereby broadcast advertising loses efficacy, precisely because it has no inherent connection to, respect for, or engagement with, it's audience. (You could even say target market. I for one have never wanted to be anyone's target - except maybe Cher's, but then I'm just perverse that way!)
This is not to say that broadcast advertising suddenly ceases to work at all - as the saying goes, Rome wasn't burn't in day - but it's efficacy relative to other approaches suffers, particularly as its audience/target market grow increasingly aware. This awareness encompasses self-awareness, awareness of others experience, as well as awareness of the practices and deceptions of traditional broadcast advertising. All of these growing awarenesses are nurtured and fertilised** by the nature of the web.
Not to demonise advertising, where I am sure that honesty does exist, but full and frank disclosure is not exactly the first thing that springs to mind when someone mentions the industry.
2. Further to other more learned responses below, I'd factor in the notion that no-one is an M of only one MM. Each of the M's of the hypothetical 100-M MM's is also an M of any number of other MM's. Cross fertilisation thus adds a fractal dimension to micro-market conversations that tends to make a mockery of the one-dimensional view of target markets adopted by traditional broadcast marketing methods.
Yes, every student of economics (even a dillettante like me) should understand that simplifying assumptions are often necessary for human comprehension. Still, if you simplify too much reality out of the equation you run the risk of ending up simply irrelevant. Does that ring any bells? Does it perhaps explain why most people feel simply insulted and revolted by most such advertising?
3. Speaking of which, you mention costs of acquisition, but then you discuss products like Coke (I assume you mean Coca Cola? ;-)) where these can hardly be measured? It's not like anyone is a passionate Coke customer. Not anyone sane, anyway? People drink when they're thirsty, and usually whatever that's to hand that is wet and preferably not immediately lethal. Yes, preference plays a part in the selection, but I'd argue that any existing preference owes far more to the efforts of the production people than to the advertising/marketing folks. And if such a preference is already established then acquisition costs are no longer relevant.
Perhaps GM is more relevant to customer retention? I know quite well, for example, how the profitability of a customer grows as the length of the relationship grows. At least in the financial circles where I have spent some of my former working lives.
Constantly acquiring new customers, and particularly where it is your method of acquisition and its resultant disappointment that pisses them off, is NOT the best way to spend your marketing budget. This is why churn is such a major consideration for companies that want to do business with you more than once.
4. I think that you've also missed one of clockies' points, which is: that broadcast advertising doesn't work on the web. The attempts to make it do so, on a medium which is inherently bidirectional (as TV and print are not) is what has made the vast majority of it (the web) into crapola.
The net is inherently open and conversational, by intent, by design, and in its operations. Trying to force-fit it into the exclusive (not to mention exclusionary) uni-directional model of broadcast advertising cannot be the height of wisdom.
Complaining afterwards that "no-one knows how to make money on the net" is also suspect, since it assumes that everyone is a clueless*** broadcast advertising wonk.
*OK, four is more than a couple, but it started out as only two - honest!
**Did you like that? I thought it was kinda cool, since the web is admittedly mostly shite.
***Where the net is concerned
Well it ain't what you dance
It's the way you dance
It ain't what you move
It's the way you groove.