Wenda Harris Millard is leaving Yahoo. That’s as good as an excuse as any to talk about what it means to sell advertising to big advertisers.

Sooner or later, just about every media plan (especially those online) includes the line, "and we’ll sell advertising." More often than not, this is a pipe dream. Here’s why:

There are two kinds of advertising and this leads to two kinds of ad sales.

The first kind is the rational kind. Yellow Page ads, direct mail and Google AdWords fit into this category. This is advertising that works, if ‘works’ is defined as, "pay $3 and make $4." With measurable direct advertising, you can count on profit-minded small organizations to give it a try (small buys) and if it obviously makes money, to buy some more. While it helps to have a salesforce, it’s not essential (if the ads are really and truly money makers). Most media companies end up with this sort of advertising, and most of them fail, because the advertising just isn’t effective enough to be this obviously profitable.

[Most advertising takes skill to turn a profit, and most advertisers don’t have any. So, there are people who make a profit with AdWords or with Squidoo offers, but not everyone. It’s just too hard to build a medium so effective and so well-priced that anyone with a typewriter can turn a profit.The challenge is to have enough people make money that others are drawn to the medium and invest the energy to actually get good at it.]

The second kind of advertising is the glamorous kind, the kind that people think of when they think of the Super Bowl or Time magazine or of profitable ads that are worth selling. These are the ads that built Yahoo, the ads that built NBC and the ads that so many entrepreneurs and media moguls are counting on. These ads don’t sell because they work. They sell because they are sold.

Let me be fair: they work if we define ‘working’ as: pleasing the client, pleasing the agency, increasing brand goodwill, and building, over time, a groundswell of awareness and brand respect that ultimately leads to profits. A Chevrolet ad on the Super Bowl doesn’t generate 2 million dollars in profit. No way. But, perhaps, over time, $100 million worth of Chevy ads turns into a billion dollars of profit. It’s vague, it’s tricky and it’s political.

Jerry, who I used to work with, and Wenda and others understood the investment, patience, malarkey and magic that goes into selling an irrational, expensive item to people who pretend to measure but really don’t. I never really figured that out, and my guess is that most people with that line in their business plan haven’t figured it out either.