USC Annenberg Online Journalism ReviewUSC

Will Microads Save Online Content? (Part One)
The next big thing could be quite small

This is Part One of our look at possible ways online content can survive in today's economy. Part Two is called, 'Emmercials: Charming the Viewer.' I missed out. I didn't punch the monkey and win $20. My Internet connection is not optimized. I didn't download Internet boost NOW. I don't have a wireless security Web camera that lets me zoom in on strange women in bathing suits, standing at odd, provocative angles while toweling off after a night swim in the backyard of a home much too nice to be my own. Why? Because I don't click on ads.

Odds are you don't either. With click through rates (CTR) hovering below .5 percent, the banner ad is all but dead -- although it's lingering much longer than anyone seems to want it around. And in the scramble to figure out what's next, many advertisers are taking an even more aggressive approach. Pop-ups, pop-unders, skyscrapers, splash page ads, and even ads that seemingly come from nowhere -- completely independent of a site-to take over your desktop are fast becoming the rule of the day.

Jeben Berg, an online media planner for Publicis Dialog/Optimedia, like most in the industry, agrees that the banner ad has seen better days, calling it 'a relic that will soon fall by the wayside.' Nonetheless, banners or not, graphics-based ads still rule the day.

'We are buying all sizes and shapes,' says Berg via e-mail. 'From skyscrapers, which are anywhere from 120x600 to 750x100 to 160x600, and generally have a 15K to 20K max file size; to 336x280 with a 25K-30K file size, which are performing above the industry average of .2 percent CTR; and also tile ads which generally run around 125x125 with a 5K file size. Of course we always get bonuses of remnant 468x60 [the traditional banner ad] space, so we run a lot of those as well. However those are not always key performers and are used more for branding than actual lead generation.'

The elephant in the living room here is how these new graphic ads will ultimately differ from the 468 x 60 banner ads. Are these not simply banners with different dimensions? Won't consumers become immune to other dimensions, just as they did with banners? Didn't we learn anything from that dadgum monkey?

Somebody did, and as usual, it was a technology company. Google started a quiet revolution of text-based advertisements, or, as they are being increasingly referred to, microads. Microads are unobtrusive, self-serve, text-based advertisements. No flashing images, no pseudo-buttons, no burning fuses, and no monkeys. Just copy and a hyperlink.

Origins of the Microad

In the fourth quarter of last year, Google launched a new advertising service. The ads were no more than text and links contained in small boxes that appear on the right side of the screen when users search on certain keywords. They're unobtrusive, and highly targeted. And even better, they work.

'The goal,' says Omid Kordestani, Google's VP of Worldwide Business Development, 'was to make something that doesn't take away from the Google user experience. The idea was that the commercial results must be complementary to the standard search, be clearly set apart as ads by color, service the client to learn how to better target the ads, and that was designed around relevancy. That's the beauty, it allows fast load time, clear distinction, and tools for relevancy.'

And so were born AdWords. AdWords work thusly: An advertiser buys 1,000 impressions from Google based on a keyword. Each time a user searches with the right keywords, an AdWord box pops up along with the search results over on the right of the screen. The ads are nearly instant; advertisers can change keywords to better target their market as the campaign runs. And here's the best part: they're dirt cheap, ranging from $8-15 per thousand.

Here's the best part: They average CTRs of 'well over 2 percent,' according to Kordestani. Some independent accounts even put them in the double digits, a virtually unheard of CTR.

Greg Stuart, president of the Internet Advertising Bureau, the premiere industry organization for online advertising, says text ads on search sites just make sense.

'Most of the time when text ads are used, it's based on a search mechanism. Naturally, [users] will read all the results given because they are on a search for something. From that standpoint, text ads are really valuable. In terms of the lower production costs, the speed [in which they can be put online], the ability to change things; you can develop a text ad for every single keyword at a low cost in a way that you can't do with graphics.'

The idea caught on, and all manner of variations -- Httpads, Pyrads, TextAds --began popping up across the Web. Several of these services are hoping to become the next DoubleClick, and even Google is talking about syndicating its AdWords. In fact, microads have caught on so quickly, that a LinkExchange style service-dubbed BlogSnob -- has already sprung up that uses the microad format to cross-link Weblogs. There is even a Yahoo! group dedicated to the topic.

One of the first sites to adopt microads was Metafilter, the community Weblog built by Matt Haughey. Google proved that microads could work on a search site, but MeFi (as its users adoringly call it) seems to demonstrate that they might work anywhere users come to click through.

'A lot of people thought it wouldn't be successful because it was too cheap, 'you should look into getting a couple of thousand from Apple,'' says Haughey. 'Well I don't know about that, how to talk to those people. This was something I could do easily. It's totally self-service. Someone makes the request, pays the $10, and it's in the database. All I have to do is say yes or no. And I accept the money with paypal.'

It's the E-Courtesy, Stupid

Although the self-service and low cost aspects of microads have made them a hit with advertisers, it's their discreet format that seems to have been a hit with users. Jean Ichbiah, CEO of Fitaly, a Palm-application development company, decided to try out microads precisely because they are non-invasive.

'The thing I like most is that it's not intrusive, so there is no way anyone will feel offended,' says Ichbiah. 'I really liked the idea that it would be discreet, so you might actually look at it. It's a paradox.'

But one that works. As previously mentioned, microads claim to boast much higher CTRs than their flashier, eye-grabbing, graphics-based counterparts. The research I did seemed to confirm this. The test microads I bought for my personal site on Blogger and Metafilter ranged from a low CTR of .778 percent, to a high of 2.03 percent. But why?

'The nature of the Internet is a text-based information medium,' says Bryan Eisenberg, CIO of FutureNow, who writes a weekly column on conversion for ClickZ. As evidence Eisenberg points to usage patterns, 'banner blindness,' and eye tracking studies that show the way people scan Web sites is inherently different than how they scan newspapers.

'In newspapers, the eye goes from pictures to the headline to the body. Online, it's different. People look at the headline, then the text, and then the images. The average CTR for banner ads today is about a third of a percent. Banner ads can be used very effectively, but for branding. Pop ups have horrible CTRs. Most people find them intrusive and a pain in the you-know-where,' says Eisenberg.

And pop-unders?

'People are disabling them. X10 annoyed everybody. We have seen some pop-under effects, but it's a very big numbers game. You have to have millions of impressions to make it worthwhile. It's a cost-per-action deal where [X10 is] only charged when people bought. It's slightly higher than .5 percent for the more effective ones. I personally have never clicked through any of them.'

'The Web is more like radio than TV, the power of words. Obviously, success comes from good copy, and targeted copy,' says Eisenberg. 'There is so much data and nobody knows how to mine it. What works is an effective ad. Nothing will beat an effective ad. We can reach the right person, with the right message, at the right time.'

And that seems to be the essence of microads, reaching the right person with the right message. Michal Wallace tried taking out microads for his Web-hosting business on both Metafilter and Google, and found that it was all about targeting.

'Metafilter is highly targeted towards indy publishers and the blog community. I felt I got my money's worth,' says Wallace via e-mail. 'On the other hand, I bought a too-general keyword on Google and spent $20 or so in about a few hours with no results. Back on the first hand, since they're so cheap, it's a good way to try out different things and see what sorts of responses you get.'

'My overall average for Google was 0.77 percent-21 clicks on 2718 ads. No big deal. But I had CTRs as high as 10 percent on specific keywords. At MetaFilter, my worst performing ad has a 1.48 percent CTR. The others were 2.3, 1.84, and 1.53 percent.'

Where Will They Work?

Haughey was skeptical that microads would work on content sites, sites that people do not call up with the express intention of clicking through-as they do at Google, MetaFilter, Fark, and on most Weblogs. Yet Eisenberg is optimistic that microads can work for content sites, as long as the ads stay relevant and don't take away from the overall user experience. Yet for content sites, the key may not be a radio advertising model at all, but classified ads in newspapers.

In many ways, the microad is merely an updated version of the classified ad. The comparisons are obvious. Indeed, when the Chicago Tribune announced that it was putting self service classifieds online, it touched off a debate as to whether or not these constituted microads.

Like the classified section in a newspaper, microads are hyper targeted, you aren't going to see an ad for, say, a wireless Web camera when you're looking for a new car stereo. They are also very low cost, making them perfect for peer to peer, rather than business to consumer, advertising. And classified ads, like microads, are booming online.

'Classifieds are becoming a huge growth category,' says the IAB's Stuart. 'The migration of classifieds from print to online is having a tremendous effect.' According to the latest report from the IAB, classified advertising now counts for sixteen percent of all online advertising, up from seven last year. Meanwhile, keyword search terms, another form of text advertising, is at three percent, up from just one last year.

But the real question is probably one of dollars. Namely, is there any money for site operators in microads?

You do the math. Suppose your site has 1 million page views per month. If you set aside one spot on your page for a text ad, and sell 3,333 impressions for $10 each, you'll earn roughly $3000 if you sell all your ads. Microads certainly seem more appealing than micropayments, which although much discussed have yet to truly materialize.

$3,000 per million obviously isn't going to save Salon's bacon. But for sites like MetaFilter, Fark, FuckedCompany, Indymedia and other bastions of independent content with small staffs and low overheads, that kind of number can be a life-or-death difference.