Questions? (877) 643-7244
FutureNow Post
Tuesday, Sep. 18, 2007 at 5:10 pm

Microsoft Tries to Measure “Online Commercial Intent”

By Robert Gorell
September 18th, 2007

oldad.jpgRecently, comScore and Federated Media announced they would measure blog readership and social networking features. The hope is to validate blogs in terms of influence, and to give advertisers a way to quantitatively and qualitatively compare these niche media outlets.

That’s great. But how?

After reading comScore’s methodology overview, a 2-pager with scarce detail — if you don’t mind wading through jargon and registering for the download, click here — I’m stumped. From what’s been announced so far, it’s too early to say whether they’re onto something. The real story will unfold once comScore coughs up the data.

Meanwhile, on his Micro Persuasion blog, Edelman PR’s Steve Rubel shares his views on this research:

Based on a [sic] informal analysis, my belief is that many online communities, bloggers, social networks will never attract a critical mass of advertisers because they are not set up properly to attract visitors who have a commercial intent to buy products and services. Online media is not sold this way now, but I bet it will be in the very near future.

Today, most advertisers size up community sites, blogs and social networks using traditional media buying models – namely, reach and frequency. Unfortunately, the reality is that many Web 2.0 sites, can’t deliver marketers the numbers they want because of the effect of Long Tail [define]. It’s simple supply and demand economics at work. This is why efforts like the one announced by comScore and Federated Media are fundamentally flawed.

Enter (Edelman client) Microsoft. In order for advertisers to assess a blog’s value, Rubel suggests they use a tool developed by Microsoft AdCenter Labs that claims to measure Online Commercial Intent by using “…terabytes of search data to calculate the likelihood of a web site to attract buyers.”

According to the Microsoft tool, Amazon visitors have a 52% intent to buy, or at least find information about buying something. Very cool. Too bad it’s not so reliable (yet, anyway).

I tried to repeat Rubel’s results for several of the sites he mentioned and, in most cases, my results differed from moment to moment. That’s not to say he’s lying. It just seems that, depending on when you search, results vary as much as 20%. Try it out for yourself if you’d like to compare your results to Rubel’s:

Consumerist – 49% of visitors have a commercial intent
Gizmodo – 47% – 45%
Treehugger – 41%
Techmeme – 41%
Engadget – 40%
Gridskipper – 38%
YouTube – 38% – 37% – 34% – 29% – 27%
Wikipedia – 14%
Flickr – 14%
Facebook – 10%
Twitter – 5%

grok_commercial_intent.jpgAlthough it’s nice to see GrokDotCom come in at 49%* — give or take 5% — I wonder what all those terabytes can really tell us about niche visitor intent and/or buying modality.

Since we blog about buying online, it’s not a huge surprise that Grok did so well. But at a certain point, volume matters. Just ask celebrity blogger Perez Hilton (a great example from Rubel). Do you think his advertisers care about e-commerce buying modalities? Perhaps, but they must have some sense of whose eyeballs they’re getting, how many, and where they shop without waiting for comScore’s or Microsoft’s data to salt their spreadsheets.

But that’s the point. An entertainment blog has more in common with TV than Wikipedia. Is niche media simply because it’s a blog? Not really.

Sometimes volume combines with relevance to create something real. For instance, a search engine’s job isn’t to help you search, it’s supposed to help you find things. In 2003, from data collected by WebSideStory (now Visual Sciences), we found that Google was the lowest-converting search engine for e-commerce. Ask Jeeves all ya want. More people still prefer Google.

*The half of you who intend, at least partially, to hire us should do it.

Add Your Comments

Comments (2)

  1. Hi Robert,
    I don’t think I could have said it better myself. Let me know if you agree. ;-)

    BTW, how come no email addresses for the writers here? I’d think you’d encourage the dialogue more.

    Keep up the good work.

  2. Tim,

    I have to disagree with you, only in the sense that I think you did say it better — or at least more concisely.

    Your points about Rubel’s methodology are spot-on, but the conclusion was my favorite:

    “The key point is it doesn’t matter if it’s Web 1.0, Web 2.0, or what-have-you. What matters is how effectively you understand your users and meet their needs. That’s the way you monetize your site.”

    Cheers to that!

    P.S. — Check your email. We may be a bit incognito, but we do write back. ;)

    (If anyone else cares to reach out about something blog or business-related, contact us anytime. Our door is always open.)

Add Your Comments


Print this Article

More articles from Robert Gorell

Marketing Optimization Blog
FREE Newsletter Sign-Up
send it once every: