(FORTUNE Magazine) – Is advertising dead? It sure seems that way, at least in my own life as a consumer. I rarely click on a Web page banner, I assiduously avoid ads in consumer magazines, I use my TiVo to fast-forward through television ads, and, when I'm driving in my car, I wish I could fast-forward through commercials on the radio.

This is not just an Internet thing. Everyone hates Web banner ads; that's why investors have decimated the stocks of businesses from DoubleClick to Yahoo. More broadly, though, there's a sense that technology is in the process of killing mass-market advertising. Last summer, for instance, Michael Lewis generated lots of buzz with a piece he wrote for the New York Times Magazine speculating that the personal video recorders from TiVo (for which I am a director and in which my firm was an original investor) and ReplayTV would spell the real end of the mass market--because they allow people to fast-forward through television ads. Radio, too, faces a technological challenge. Local radio stations are digitizing their broadcasts and streaming them over the Internet, while two new all-digital broadcast networks will deliver broadcasts to special radios in new-model cars next year. (For more on the latter, see "Satellite Killed the Radio Star" in this issue.) Both digital developments threaten to turn radio into a national business--but currently more than 90% of radio revenue comes from local advertising. Newspapers and magazines also face the technology assault. While they haven't been seriously hurt yet by Web competitors, their own Web offerings have largely been failures. The result: They can't be confident that their key source of current revenue--selling static ads that sit near static pages of editorial copy, like this one--is a reliable source of future revenue.

The gist of this, of course, is simple, persuasive, and unavoidable: Interactivity, the hallmark of computing technology, fundamentally calls into question the viability of a model based on being able to force people to watch, listen to, or read advertising. Besides, what could be more appealing than to imagine the death of advertising? We hate the stuff, and we're especially sick of having completely noninteractive ads shoved down our throats.

In spite of everything, talk of the death of advertising is nonsense. Despite its many flaws, Internet advertising has grown in just four years from virtually nothing into a $5 billion business. Radio advertising still generates more than $15 billion each year, magazine advertising about $12 billion, and television $35 billion. This is big business. And while it might seem nice to imagine a world in which we could just skip past ads, I can't. Companies will still want to use advertising of some kind to communicate with customers, and customers will still want ads that inform them of new products.

The real problem may be that our entire system for valuing advertising is completely screwed up. In every major medium the basic benchmark for pricing ads is CPM, short for cost per thousand. In magazines, it is the cost per thousand subscribers (or readers), as measured by an accepted auditing group. In television, it's the cost per thousand viewers. In radio, it's the cost per thousand listeners. In very general terms, consumer advertisers expect to pay somewhere between $5 and $30 per thousand impressions, while business or specialty advertisers expect to pay $20 to $100 per thousand.

That's in old media. On the Internet, everything is even vaguer and more confusing. CPMs have dropped to a dollar or two. But no advertiser really knows what thousands of what it is paying for. In theory, a thousand impressions on the Internet is some approximation of a thousand pairs of eyeballs looking at the Web page that carries the banner. But despite the best efforts of the likes of DoubleClick, there's no practical, reliable way to know who is looking at your ad. For that matter, there's no guarantee that you are buying a thousand human impressions, given that computers scan Web pages all the time as they automatically hunt for various kinds of information. And then there's an additional problem: No major company has had a huge success thanks to a Web advertising campaign.

But here's where I think it gets interesting: Can anyone say for sure that any of those "thousands" defined in the old media--television, radio, magazines, etc.--are more reliable indicators than Web "thousands"? I don't think so. The supposedly reliable benchmarks of those media were actually just mutually agreed-upon illusions. No company expects that by paying a magazine $40 for each thousand subscribers, it will actually get all those people to read its ad. But this CPM metric has seemed like the best one available. Now, with the Internet, and technology that allows us to measure things more precisely, like whether someone actually buys something after he clicks through on an ad, those imprecise measures seem more Stone Age than ever. No, advertising isn't dead. But the way we measure its value is going to be radically different in the future.

STEWART ALSOP is a partner with New Enterprise Associates, a venture capital firm. Except as noted, neither he nor his partnership has a financial interest in the companies mentioned. He can be reached at alsop_infotech@fortunemail.com. His column can be bookmarked online at www.fortune.com/technology/alsop.