I do know that I do not like the ads that manage to break through my attempts to ignore them. Thus, I may not be an ideal person to talk about the long-term viability of ad-based internet sites. But I do find it hard to see a reason to be hopeful about most of the sites I've seen.
Clearly, advertising can work. Multibillion dollar broadcast TV, newspaper and magazine businesses prove there is something there. Some alternatives to advertising-supported media, such as government financial aid and its too-often associated content control, are less than attractive. Others, like the subscription-based access used by premium channels on cable TV or some websites such as The Wall Street Journal, can work quite well. But they require that the user be identified, which can be a pain and presents a privacy worry.
Internet ads have the potential to be different from most non-internet ads in that the advertiser can find out if they work. This is harder to do in most current advertising arenas because advertisers don't normally only do one kind of advertising at a time. The normal mode of operation seems to be advertising campaigns with coordinated ads in multiple forms, from bus wrappers and subway placards to TV and newspapers spots. The advertiser cannot easily find out the effectiveness of each individual form.
But on the internet it can be a lot easier to determine if the ads work. This is particularly true now, considering the move to ad fees based on click-through is what counts rather than just the number of eyeballs that see the ad. It's hard to imagine something more terrifying to an ad agency than to have its success measured on a per-ad basis based on actual results.
Ad companies are publicly salivating over the prospect of being able to produce ads targeted to users based on their individual tastes or current location. But these same companies don't seem to get that internet users are concerned about random third parties knowing too much about them. DoubleClick's almost 2500-word privacy statement is an example of how little the ad industry understands privacy concerns. It would only take 100 words to tell me what I need to know.
It seems so obvious to me that internet advertising has too many problems to be able to support things such as "free" PCs or internet access that I can't understand why anyone would ever have thought it could. But it seems many investors did. I don't envy the returns they are getting on their investment.
Disclaimer: "Harvard" and "free" are not normally associated terms, and the university has not expressed a view on this topic.
Bradner is a consultant with Harvard University's University Information Systems. Send email to Scott Bradner.