The glut of advertising on the web

Eric Clemons, Professor of Operations and Information Management at The Wharton School of the University of Pennsylvania, recently published a short essay on TechCrunch about why advertising is failing on the internet. The piece is largely qualitative, and ignores some of the quantitative side of the equation, though none the less I do think he has some valid points regarding the long term evolution of advertising as it moves online.

However, despite my support for his premise in some areas, I disagree with him on others. I'll admit I'm a little biased, especially when it comes to his remarks on search advertising (I was a full time search marketing manager for three years), but overall I feel that his assertation that online advertising revenue is dropping largely because of its ineffectiveness is at least partially flawed because he seems to ignore some of the basic economics of the online ad industry.

Content (supply) is expanding much more quickly than the demand of online advertisers.

This is more applicable to ads that are sold on a CPM basis, but it can also affect CPC ads as well if there is a lack in relevancy (more on that below). It's a basic economic principle: when there is more supply than demand to fill it, prices drop.

Because of the lack of quality targeting or evaluation of influence, many advertisers still skew toward 'reach' and want only big sites, even if they're just pageview traps with low user engagement and an unfocused user base.

Also, many webmasters have gotten very good at creating what I call 'pageview traps'. For example, needlessly splitting articles into three pages, or creating a Top 10 list on ten separate pages. This drives up the pageviews and artificially skews other user metrics that can be used to sell advertisers on strong user engagement (avg. pages viewed per visit, bounce rate, time on site, etc) making sites appear to be good choices for an advertiser.

In reality, those users are not likely not paying any attention to the ads - they just want the next piece of the content they're viewing. While this ups pageview counts, which might help the content provider milk more CPM out of its advertisers, it often decimates ad engagement, so ads sold on a CPC (cost per click) model are likely to see clickthrough rates plummet.

As more and more sites do this, the end result is massively expanded inventory, dropping overall CPM/CPC rates.

Go a little further down the chain to smaller niche sites with a focused userbase and relevant content, and the problem then becomes filling the 'reach' side of the equation. On these sites, the topics and readers are usually pretty highly focused, but the only way they can scale to appeal to advertisers is to aggregate together via ad networks. The problem is, once dozens of sites are clumped together, then the network of sites as a whole are just as likely to suffer from the same problems noted above.

So while I do agree that the current drop in ad revenues is partially due to the ineffectiveness of ads, I also think that overall, the online ad economy needs some tweaking. Currently the incentives for content sites are to maximize pageviews and clicks on ads, often by whatever means necessary, and this is greatly devaluing the quality to both advertisers and consumers.

When a model is developed that can target more effectively, with pricing based on performance, then the incentives will shift away from intrusive ads, pageview traps, and misdirected clicks, and instead be about creating messages that actually fit and are worthwhile to both advertisers and consumers (for example, ads in lifestyle magazines are often just as worthwhile to many readers as the actual content).

I'm not sure how we'll achieve this, but I do agree with Mr. Clemons that it won't be through banners.

//read more at TechCrunch
//photo via Darren Donahue