Social Media Now: Measuring the Market

In the media world we call them “metrics.” In English we call them “measurements.” Known by any name, these comprise the lifeblood of the business. We buy and sell advertising and make biz dev decisions based on what measurements tell us.

The Internet promised to be the most measurable medium ever. And it is. But on-going disputes about the accuracy of social media metrics continue to point out just how primitive and uncertain those measurements remain. And perhaps these uncertainties are holding back the growth of social media.

This morning Heather Green of Businessweek points out a gulf between comScore and Hitwise when it comes to measuring YouTube. According to comScore, US traffic to YouTube (as measured by unique visitors) climbed 2.6% from December to January. According to Hitwise that growth was a whopping 33%. How do you explain a 30% difference?

Writes Green:

Hitwise says the difference is in the methodology and sample size. It partners with ISPs and has a sample size of 10 million folks. comScore says its panel of folks it recruits and tracks is at 2 million.

Certainly a sample size of 10 million and traffic recorded directly ought to deliver more accurate numbers than a recruited panel 5 times smaller. But who knows? There are so many variables that could be involved–who exactly are the folks sampled? How disparate are they demographically? Psychographically? How does Hitwise’s ISP head end technology work? Is comScore’s panel self-reporting or automated? Does either company track users across platform–from home PC to laptop to smartphone to company desktop? Is traffic from inside corporate gateways measured at all? How about colleges?

We may never have agreement among metrics companies. And frankly it may be valuable to have difference companies sampling the market with different methodologies. What we need in order to have faith in these measurements is extreme transparency. Forget the secret sauce, this is the Internet, the more we know about how you do what you do, the more we’ll believe the results.

Meanwhile, an entire universe of micro-metrics has sprung up enabled by blog widgets and RSS feed providers. Every blogger can track RSS subscribers, unique visitors and more with the addition of one button or another.

Yesterday Techcrunch wrote about the data being collected by AddThis–a meta-tagging widget–information that includes relative use of various social bookmarking services, or even data about which items are tagged most. But even micro-metrics create numbers that are open to dispute. Data Mining points out the conundrum using Nielsen’s BlogPulse ratings which monitor the addition of buttons like the one offered by AddThis.

While AddThis appears to have had a peak at 0.009% of all blogosphere posts, del.icio.us was recently tracked at 0.25% a huge difference. Interestingly, AddThis claims 2MM clicks per month – do we conclude that del.icio.us gets 100s of millions of clicks per month?

Great question. I’d be curious what if any measurements are most trusted and used by SMC members and SMN readers? What other measurements would you like to see? What are the problems with the metrics out there today?