Andrew Chen: The common perception is that a long recession would cause consumer spending to drop, and advertisers dependent on that spending will run into trouble. However, unless they want their revenue to completely fall flat, they can't eliminate their entire advertising budget. Instead, they will focus on advertising that is more of a "sure thing" and which is often described as a flight to quality. For online ads, those who have the ability to deliver consistent, accountable revenue will continue to thrive - most analysts believe that Google will continue doing well because search marketing is some of the highest ROI, most accountable spend out there. Similarly, other direct response ad inventory including remnant ad networks, affiliate programs, some verticals within lead generation should all be fine.
On the flip side, the publishers that are seen as cool but experimental will have trouble attracting dollars. For example, advertisers may have less of an incentive to be the first to be the first to try out a new social medium, virtual worlds, widgets, or other types of ads which are perceived to be in their infancy. Because these types of media have no track record, lack established standards and thus have higher transaction costs, even though they are great branding opportunities they may seem riskier than before.
But just as the percentages of some online brands drop, the wildcard is how much that is offset by having brand TV dollars migrate to the web at the same time. My expectation would be that the biggest beneficiaries of this will be branded publishers selling banner ads and brand integration opportunities, however. Think more MTV and less VampireFreaks.
BNET: Currently there's a plethora of ad networks -- can the market sustain so many?
AC: In my opinion, the increase of ad networks over time will probably continue over time, in part because it's so easy to start an ad network and because there are only moderate network effects. Because of the continuing automation of the ad industry, it's becoming easier for individual ad networks to support inventory from thousands of publishers and advertisers, based on ad exchanges like Rightmedia and the commoditization of once defensible technologies like ad servers. Similarly, it's very easy for a couple key individuals to leave an ad network like Valueclick and start a new one themselves, bringing clients along. In this way, one can think of ad networks more like brokers in a multi-broker exchange rather than as the exchange themselves. As buying and selling ads becomes more efficient, the number of parties who will execute these transactions are likely to increase, not decrease.
BNET: Will social networks ever become a good place for advertising?
AC: I think the answer to this question is kind of a nuanced one. The underlying question might actually be: Will social networks dominate advertising in the same way it dominates traffic?
The reason why I'd clarify this is that MySpace is making $650 million a year and is growing 100% per year. That's a ton of money for any startup, yet it's still far behind in terms of the fact it's also one of the top sites on the internet. So clearly for some set of advertisers, social networks are already working as an advertising vehicle, but it's taking time as brands get comfortable with the medium. I think that within a 4-5 year timeframe we'll see a lot of money in the social network space, but it depends a lot on recessionary spending, among other factors.