I’m not much one to link to other’s research, but there was a very nice piece of work done by a British SEO firm. Their concept is basically that inbound and outbound links are a karmic boomerang. I really like this piece of work, and plan to do some follow on research in the next couple of months to verify the concept and add more depth about the type of audience that follows this return path. Kudos guys, nice work.

Tags: search links, seo, virality

I had a very nice conversation last night with some clients from BlogHer.  The topic, like that of many conversations among those who work in social media here at Ad-Tech Chicago, was: Should we use the same metrics to measure social media as we do more standard online media types?  This dovetails off of my post a couple of weeks ago creating a taxonomy for media properties.

In response to these conversations, I’ve started a list of metrics we should be tracking for social media buying and selling.  The first is virality, the amount a blog is linked to.  Influence is another one (although we’re still working on the best way to measure it).

I’m open to suggestions.  What metrics are you using to sell your social media site?  Send them my way, I’d love to build this list and make these metrics available for custom runs of data.

Tags: influence, social media, virality

One of my primary goals since returning to Nielsen Online has been the fusing (or smooshing as I like to call it) of BuzzMetrics CGM data and NetRatings audience measurement data. Yesterday at the IAB Forum on Social Media, Jeffrey Graham (of the New York Times, among other places) and I presented on some new research we’ve just completed on this – I was really happy with the results.

The over arching goal was to try to better understand the relationship between inbound links (blogs linking to other information sources) and traffic data. We also hoped to understand what this relationship says about media brands and hopefully build planning/selling metrics that could be used. I think we achieved all three.

Here’s a quick summary of the findings. Within the news category, inbound links and unique audience are correlated at about 0.73. When you add other metrics such as time and visits you can get up to an r-square of about 0.4. Given the less than impressive modeling results, we decided to stick to a simple correlation for this work. We developed a metric we’re calling the “Brand Virality Score.” Although we might need to change it due to a very easy spelling mistake I’m almost certain to make in some incredibly important presentation, the metric is fairly simple:

Inbound Links/Unique Audience = BVS

Our reasoning is that if size and inbound links are so tightly related, if a specific site is overachieving on inbound links it says something about the site itself, more specifically about consumer’s perception of its brand. So we ran the data and here are the topline results:

The New York Times, clearly does very well here. They have both a high BVS and a high UA. Not an easy feat to achieve. Here’s one for financial news sites/sections:

It makes sense to me that niche financial pubs with strong brands do well here. The Economist, Bloomberg and FT all have a tremendous clout in the market; it’s nice to see that reflected here.

Now an excellent question one might have is: “Why the heck should I care about another metric?”

The answer is, the degree to which a mainstream media brand is being linked to relative to its size is a strong measure of its impact within word of mouth marketing. That this media brand is “viral” suggests that it has a very strong brand. Perhaps more importantly though, viral media brands take their advertisers along with them, when they are being pushed out into the blogosphere. For brands or agencies interested in developing viral campaigns (and who isn’t?) we argue that it is important to focus on those media partners that can help distribute your message outward. In many ways this metric, along with UA and time per person form three critical elements of the media planning process – size, engagement (yeah, I should know better) and virality. I’m working on a fourth leg to this stool that will get at clutter. But more on that in later posts.

If you’re interested in hearing more, or want to get a copy of the deck I presented – drop us a line.

A big thanks to John Brauer who was main analyst on this project. He did a heck of a job!

Tags: brand, links, virality