Last week I described a coming revolution in the delivery of television programming that will fundamentally alter advertising, with profound implications for political campaigns.

Within 10 years, viewers are going to be able to control the ads to which they are exposed. Unless voters and consumers want to see an ad, they won’t.

Responses by commercial advertisers to this challenge focus on three dimensions: ubiquity, quality and targeting. Ad makers are struggling to make their work interesting enough that viewers actually want to see it. They are then putting it in every possible channel from the Internet to word of mouth. In addition, marketers are utilizing the superior targeting capabilities of cable to focus their resources on audiences most likely to yield results. Sophisticated targeting also enables advertisers to alter ad content to make it more enticing to particular segments.

Perhaps the most celebrated success of the new marketing paradigm was BMW’s short films. The auto company hired top directors to make five short films staring BMWs, which became a rage on the Internet. More than 14 million downloads were recorded, and BMW sold more than 200,000 cars in the United States for the first time ever that year. “The wave of the future,” cheered the marketing world. But think about the differences we in politics face in cost, audience and scale.

BMW spent $15 million making its films appealing. Campaigns spend $18 thousand dollars or less on their average positive ads. No one I know of has spent more than $70,000 making a candidate spot. The average 30-second spot you see on TV costs $375,000 to produce.

A few years ago, top creative agencies offered to make some spots for the party. With “everything donated,” they estimated a spot could be made for as “little” as $1 million. Chanel spent $10 million-$20 million on its recent ad.

When candidates point to a $5 million Coke ad and ask their media consultants, “Why don’t mine look that good,” I want to scream, Sam Kinison-style: “BECAUSE YOU SPENT 10 GRAND AND THEY SPENT 5 MILLION!”

It is possible to make compelling films cheaply (see my cousin’s forthcoming documentary “Mad Hot Ballroom”), but it doesn’t happen very often. Indeed, most of the advertisers who have tried to replicate BMW’s success have failed, even with big budgets and the world’s top directors.

While political spots are vastly less expensive than commercial spots, our audience is also different. Product marketing, whether on TV or the Web, begins with an audience that wants a car and is at least interested in a BMW. The parts of our base we need to mobilize because they vote infrequently are, almost by definition, not interested in politics. The swing voters we must persuade, likewise. How successful would the BMW films have been in a group of hardcore cyclists who weren’t planning to buy a car?

Finally, the scale on which we need to communicate has not been matched, even by BMW. If each BMW download was a different registered voter, those voters represented about 11 percent of those who cast ballots in the last presidential election. And how many of those downloads were in any one state or congressional district?

Of course, it will be very useful, and very expensive, to target ads in a more fine-grained way. But higher-level targeting may not get political ads all the way over the hump to compelling.

The advent of television inaugurated a revolution in political campaigning, with technology driving change. During the next 10 years we will likely see more changes in campaign communication than we have in the past four decades combined.

Mellman is president of The Mellman Group and has worked for Democratic candidates and causes since 1982, including Sen. John Kerry (D-Mass.) last year.