Users can digg — or bury— sponsored messages. (Zoom in)
Social news-sharing site Digg is testing a new advertising model in which sponsors pay less the more popular their ads are. It’s the latest in a slew of social media sites to let users vote on ads, a practice that has the potential to challenge the clickthrough as the standard metric in online advertising.
Digg Ads – which will flow through the site’s main content stream like other stories while being identified as sponsored — will allow users to vote their approval or disapproval of each ad and comment on it the way they do regular stories. Popular ads will rise to a prominent place on the page while unpopular ads will be buried. And, most interestingly, advertisers will pay less if their ads are popular.
Digg isn’t the only social networking site experimenting with this kind of democratic advertising model — StumbleUpon and Reddit both allow users to vote on ads, as does Facebook. Some reward popular ads by letting them hang around longer, effectively a discount since more ad time for the same price means a lower cost per impression. But Digg’s model is bolder, since it gives popular ads both a price discount and more prominence.
It will be interesting to see whether users chime in more for ads that they like or ads they hate. Remember, there’s long been a way for viewers to vote for ads they like: It’s called a clickthrough, and it’s getting rarer all the time. Although click-through rate is the biggest determinant in the quality ranking that Google gives to sponsors in its ad auctions, click-through rates on display ads today have dropped to about 0.1 percent, according to Doubleclick. One could argue that people are even less likely to vote on ads than they are to click through them. But ad voting is less intrusive; it doesn’t take people to another site. And falling interest in clickthroughs is a good reason to start testing the viability of voting as a supplemental data source.
Digg’s system will give advertisers more feedback on their ads, a feature they might try to sell as a sort of free focus group. But it will be an unusually public focus group, and advertisers might not appreciate being defamed in the comments section below their ad (which might tempt social networkers to sensor comments about ads — a move that would de-legitimize the system).
But what is a “better” ad, anyway? One man’s treasure is another man’s trash. The problem with Digg’s system, as I see it, is that even if an ad for chicken costumes for cats is right up my alley, the mob might well overrule me and bury it before I even lay eyes on it. In that sense, Digg’s ad-voting system could seem like the opposite of targeted advertising. That doesn’t mean it has to be. The greatest promise for ad voting may be its potential as a less-intrusive tool for mining the user preference data that enables targeted advertising.
That potential seems more obvious when Facebook, with its treasure trove of profile data, culls votes from individuals than it does in Digg’s mob rule. But Digg says its system could ultimately harness the same power. “Our algorithms will be able to identify compelling advertising content that may only appeal to smaller segments, and as we build out the platform, these ads will be targeted by topics and other criteria,” Mike Maser, Digg’s chief strategy officer, told me in an email.
For now, incenting advertisers to make better ads is not nearly as important to social networking sites as incenting users to vote on ads. Companies like Digg should focus hardest on that, encouraging users not just to vote away what they don’t like but to vote for what they do like. They should try to make the experience as fun, simple and unobtrusive as possible while leveraging people’s natural inclination to opine. Facebook has tried asking users why they don’t like something (Irrelevant? Inappropriate?), but too many choices requires the opening of a new window.
Instead, I recommend just a few options, similar to the “Like” buttons Facebook already uses. Make them conversational, like “Cool!” and “Not for me.” And allow opinion-sharing among friends – maybe a number next to each choice tells you how many of your friends voted that way, and you can see which ones by clicking on the number. Overall, though, the social experience of group voting shouldn’t preclude the ability to use that voting data to target ads toward each individual’s unique tastes.