FTC complaints are all the rage in the privacy vanguard, yet we hold our position with the same defense:

“Give us more personal data and we’ll give you more relevant ads.”

But there’s one caveat with that statement:

People value their privacy more than relevant advertising.

Although, Yahoo intends to make big advancements on how they manage personal data and who retains ownership:

“I could envision an icon that appears when you see an ad, and if you were to click through that icon, you would see the data we’re leveraging…If we’ve got it wrong, you can tell us or you could turn it off” – Jeff Weiner, EVP of Yahoo’s Network Division [full CNET article here]

I’m all for transparency and choice, only most people are ignorant to the fact that their behaviors are being tracked in the first place. Releasing such a system to the public is a delicate matter and could end up being counterproductive (we saw what happened when Facebook launched Beacon).

First people need a real motivation.

For example, Yahoo could encourage advertisers to offer exclusive deals in return for deeper data-sharing. Since advertising costs are built into product pricing, the pitch could be made that as we collect more data and advertising becomes more efficient, marketers have more room for valuable promotions.

…after all, personal data is privileged information and we should always respect that.

From a celebrity gossip website (click to enlarge):

The sad truth is, despite the poor nature of the content and the users disinterest in Lasik, they will still click the ad. This goes to show that content relevancy does not always equal audience relevancy.

Remember last year when Google started penalizing bloggers for selling sponsored reviews? The story was that Google disliked the idea of advertisers requiring text-links within their reviews (i.e. paid links). As a result Google punished bloggers by removing their PageRank altogether – specifically PayPerPost bloggers. Soon after, CEO Ted Murphy announced a new initiative called “RealRank,” aiming to eliminate PageRank, Alexa, and Technorati as blog measurement tools.

I haven’t heard anything new about RealRank until yesterday when I received a beta invite for SocialSpark. This is basically IZEA’s (parent company) attempt to rebrand PayPerPost; meanwhile updating the interface and adding in some nice social networking features. It seems they’ve been using RealRank as a way for advertisers to filter through blogs. This is helpful, considering the thousands of blogs in the network. Although, taking a deeper look into RealRank it doesn’t seem much time or thought was put into the system. Here’s a look at how ranks are calculated:

RealRank Algorithm

70% weight toward visitors per day
20% weight toward amount of active inbound links per day
10% weight toward pageviews per day

From IZEA’s blog:

When creating an opportunity, advertisers are asked to select a rank from 1 through 9. This really equates to a percentage of the marketplace to exclude from an opportunity. If an advertiser chooses a 9, they are effectively excluding 90% of the marketplace and only allowing people in the top 10% of the ranks into the opportunity. [full post here]

The Problem

What we have above is a way for bloggers to measure their popularity, but I think the idea behind RealRank is to add value for advertisers as well. Putting so much weight on reach is a mistake. Relevance, influence, and audience composition are far more important when planning media buys. Those are the things that drive performance. Reach is the added bonus.

The Solution

Ted, the solution is pretty straightforward. Test more variables and use advertiser conversions as the benchmark. Look for common characteristics in the data and adjust the weights respectively. Also, keep a look out for high performance blogs that contradict the algorithm. They’ll serve as good learning examples and may stand out as candidates for “manual” weight biases (I know that sounds bad, but it’s practical for business). If you need to have two separate algorithms, one for the public (IZEARanks.com), and one for SocialSpark, that is acceptable as well.

Some starting variables to consider:

Reach

• Pageviews
• Unique visitors

Reputation & Influence

• Audience growth
• Content quality
• Conversion rates
• Link profile (based on a weighting system – yes, similar to PageRank)
• Visitor Loyalty (new vs. return visits)

Engagement

• Time spend on site
• Visitor to comment ratio
• Visitor to RSS subscriber ratio

When evaluating these variables you’ll need to think about how they change from category to category. The fact is every category represents a different audience, each of which is different in size and has different behaviors. Therefore a blog classified as an overall “5” may be a “9” relative to its category. How you use that is up to you.