When the A-List Gets It Wrong: Scoble/Rubel

One of the things I have found very interesting over the past year is how quick Internet viewers are to believe whatever an "A-lister" says. It could be somewhat misleading through absolutely incorrect, but slap a popular name on it and it’s gospel. I have two good examples, one from Steve Rubel last week discussing economics and from yesterday we have Robert Scoble’s attempt at defining analytics. Neither is absolutely incorrect but both seem to be going in the wrong direction with their analyses, so let’s see if we can get them back on the right track.

Steve Rubel – Web 2.0 Economics 101

Steve Rubel, VP at Edelman, attemps to teach us all about how supply and demand works. I have very much enjoyed reading Steve’s blog for a long time but lately more posts are negative which I frankly don’t get. Steve makes two economic points:

1. As the supply of content rises, attention decreases and demand lowers – e.g. traffic thins 
2. As the supply of ad-supported media rises, inventories swell – e.g. this equals less ad revenues

One of the commenters noted, "Just stick to using frameworks you understand. Don’t write about economics ever again, Steve." I agree. Had Steve presented these thoughts in the context of the Web without the economics, it might have worked better. Let’s look at each of his points and why they aren’t correct:

1. When the supply of content increases, the attention per site might decrease, but attention on the whole does not and demand overall does not lower. Steve should have also included the fact that while we are adding more content supply, we are also adding more demand overall as more people come online which basically wipes out the his economic theory. Using a real-world example, when CenterNetworks began, did RWW/TC/GO/MASH lose traffic? No, CN just added to the mix. For attention to decrease, it would assume that we are already at our max attention level and any new entry point would decrease attention to the other points of attention to which we are already paying.
2. As we add more supply of ad-supported media (I assume he means sites/blogs that are use advertising as their means of revenue), Steve believes that available inventory increases and thereby lowers the price for the available inventory. In the basic sense, Steve is semi-right here. If supply increases, price typically lowers to increase demand to meet the supply. However again, we are leaving out the apple in the pie. This assumes that there is less demand for the advertising slots. We just learned that Intel is shifting all ‘08 TV advertising to online. So now we have additional demand. Additional demand has to be filled at that same rate as currently available (or higher). The bottom line is that until demand goes down (it’s going up now), then price will not decrease because until that happens there won’t be as much fighting for the same ad dollar/euro/yen.

I think Steve had the right ideas in mind but as he noted, he didn’t take any economic theory in class.

Robert Scoble – The "New Metric"

This week we get another chance to check out the ScobleBoard and it’s about metrics (one of my passions). Robert has determined that there are four characteristics to his metric: Audience, Engagement, Loyalty and Influence. You can check out his video (5 minutes) for his explanation – his blog has it defined in words.

Let’s look at why he is off-course with his definitions. It’s too bad too because I think he could be on the right track if he defined the terms better instead of making generalizations that don’t fit. You can’t stick a size 12 foot in a size 5 shoe, won’t work. He uses Engadget (ET), Techcrunch (TC) and Valleywag (VW) as his example sites, so I will as well.

Audience – Robert uses Valleywag to define audience by saying that they have a sizeable audience based on how many people have loaded the player for Kyte – he calls this number 300. Then he explains that when TechCrunch embedded the Kyte player, they had 1,000 people viewing it. So based on that analysis, Robert believes that the TC audience is "at least" 3-4x larger than the Valleywag audience.

Why is this wrong? First, he is using one story as a basis for an entire theory. Second, how do we know that the day he was listed on TC wasn’t a major story day which brought more traffic than normal to the page which would default and load the player? Or the opposite where TC would normally have 5,000 visitors but it was a slow news day or a weekend?

Third, how do we know that VW doesn’t have a larger percentage of people who read their content within the RSS vs. TC (I have no idea). (and yes, I know TC is larger than VW, but you get the idea, these comparisons aren’t logical)

Engagement – "clicking or commenting behavior" – He notes that the average post on TC has hundreds or dozens of comments while the VW posts have generally no comments. He finishes off by saying his blog is smaller but has higher engagement than VW. Then he notes that few people "do anything" on a VW story and he is correct when he says that VW is entertainment and no one wants to interact with that.

What he misses here is that half of the stories on TC have great commenters and half have not so great commenters. And if he was to plot this out showing engagement higher on TC/RS than VW, he would be incorrect for exactly the reason he stated a minute earlier, VW doesn’t expect engagement.

I wonder what he thinks of blogs like Calacanis/Godin who turn off their comments. I know Jason believes that the engagement comes from those who provide their views on his stories on their own blogs and link back to his. How does Robert account for that?

Loyalty – discusses Engadget here – uses Google Reader to explain that there is more loyalty for ET than Gizmodo because ET has hundreds of thousands of readers and Gizmodo doesn’t. So we are using one of the 8 million RSS readers to determine that Engadget has a higher loyalty than Gizmodo – that’s like reading the balance sheet and determining that the company was extremely profitable on it’s own. And let’s not even get into the fact that RSS subscriber numbers are meaningless. Look at Shoemoney and John Chow bait their visitors into picking up their feeds in a contest. More subscribers absolutely does not equal more loyalty. More action on a feed could indicate more loyalty.

Influence – "Influence is separate from the other ones – it’s the numbers of posts on StumbleUpon/TechMeme/Digg/Reddit/Slashdot" – he calls these "news aggregation sites" – he then goes on to say that these sites "look for posts that they want to have in their aggregation". I call a LOLCAT here. I didn’t realize that Digg/Reddit were news aggregation sites. Out of the sites Robert lists, the only one that is a news aggregator is TechMeme. And getting on these sites is easy.

The defense rests.