Links? Phooey. Eyeballs, that's what advertisers want.

17 August 2005

ClickZ has reported that the Feedster 500 list of most influential blogs is of "dubious value as an evaluation tool for media buyers". Well call me surprised. A list made up largely of inbound links from blogs with some other factors rolled in does not satisfy advertisers? The surprising bit was that the concern was because the list does not rank blogs "according to niche or topical focus". Steve Rubel says it "sounds like an opportunity for somebody else". It is indeed an opportunity but not based on a metric that is more important to bloggers than anybody planning advertising spend.

Inbound links are easy to measure and give a reasonable idea of how bloggers view other blogs. But that is a poor metric for something that advertisers actually care about, unless they just want to reach active bloggers. There is probably a reasonable correlation between inbound links and visitors. But what advertisers really want to know about is traffic. How many people look at each page? Or, if it were possible, how many people look at this page are looking to buy a consumer durables with a value of $1000 this month? Here's an idea. Why don't bloggers simply make that information available instead of complaining about how inaccurate Alexa is and the flaws in lists based on blog 'influence'?

Boing Boing, for example, puts up its stats in Awstats form. I appreciate not everyone can use the same approach but getting bloggers who care about this to provide stats in a reasonably raw form would go some way toward demonstrating what sort of traffic each one gets, if advertiser friendliness is what they want to demonstrate.

Traffic data can be faked, but the odd spot check here or there would help keep the playing field level. I'd be interested to hear why traffic data should be kept secret. I don't plan to have advertising on this blog but I would have no objection to ponying up the data if it were possible to make sure none of the data released publicly could be used to identify an individual visitor.

However, I would be even happier if advertisers stopped trying to measure random variables on the way to a sale and simply analysed whether campaigns worked or not using actual sales. That might stop advertising agencies chasing awards and focus their efforts on things that work, rather than measuring other aspects of a campaign because that is the easier way to analyse a campaign. It's like the habit of PR companies chasing up whether hacks have received each press release. Who cares? What matters to them really is whether any of that work translated in column inches. But, apparently, that is harder to measure.


Sorry Chris, this is still "risky media" for advertisers. Eyeballs are nice, but they'll only pay for clicks -- and only for a short time if those clicks do not lead to conversions.

A fair point about the clicks and conversions rather than the eyeballs as a proxy that advertisers 'want'. However, I still have my doubts as their value as a proxy for response given all the problems with click fraud on high-value search terms and that having active responses to ads only works for certain things.

At some point, the people making the bookings are going to realise that advertising for most things is not going to trigger a Pavlovian response in human beings. The only measures they are going to be able to use are the old-fashioned ones, such as surveys of campaign awareness and actual sales after a campaign.