Consumer Segmentation is now being measured by comScore Media Metrix by dividing audience into light, medium and heavy users.
It's fair to say that if you can segment your users by audience and get your analytics to track that segment's behavior; and start measuring Visitor Engagement by audience. The way ComScore determines a Heavy User vs. a Medium User vs. a Light User is:
"..The Heavy segment group will be defined as the top 20 percent of consumers, based on time spent online at the category of sites; the Medium group will be defined as the middle 30 percent; and the Light users are the lightest 50 percent. H/M/L user segments will be available in all 32 countries where comScore syndicated data are available and across all 110 comScore defined categories and sub-categories of sites.
It is all about time spent - but that would fall short of Eric Peterson's Engagement Metrics because we don't know what they're doing when they're talking more time. We need to know what they do AFTER the come to the site and spend time (or what they're doing at the site) before we can say if the extra time spent is a good thing or not. Omniture, Coremetrics and Visual Sciences can segment audiences by time spent - it might be possible to create a custom segment in line with ComScore Segment Metrix and link up site analytics with the rest of what ComScore provides in the Segment Metrix.
According to ComScore Press Release:
"..Consumer segmentation is not new - it is an important part of marketing, and over the years comScore has conducted a great deal of custom segmentation work,” said Bob Ivins, comScore executive vice president. “comScore Segment Metrix enables segmentation on a broad scale, offering marketers the ability to analyze online trends by key consumer segments and from several different perspectives for themselves and their competition.”
"..The first segmentation to be offered is one defined by online behavior. This approach, called comScore Segment Metrix H/M/L, will allow marketers to analyze online activity by heavy, medium and light users of the Internet and of any category of sites reported by comScore. The Heavy segment group will be defined as the top 20 percent of consumers, based on time spent online at the category of sites; the Medium group will be defined as the middle 30 percent; and the Light users are the lightest 50 percent. H/M/L user segments will be available in all 32 countries where comScore syndicated data are available and across all 110 comScore defined categories and sub-categories of sites.
“Medium and Light Internet users can be just as important to advertisers as Heavy Internet users. Knowing where to find these less active consumers will help our advertising customers build incremental reach more cost-effectively – which is a good thing. Having this behavioral understanding is a huge advantage,” said Lynn Bolger, VP, Advertising & Sales Insights, Yahoo!."
"..comScore Segment Metrix offers H/M/L segmentation data for 110 categories and sub-categories including Business/Finance (with 7 sub-categories), Career Development (4), Community (7), Directories/Resources (7), Education (2), Entertainment (9), Gambling (3), Games (3), Government, Health (3), Hobbies (4), ISP, News/Information (4), Portals, Real Estate, Regional/Local, Retail (21), Search/Navigation, Services (11), Sports, Technology (2), Telecommunications, and Travel (8)."
So now that we have Consumer Audience Segmentation by Heavy, Medium or Light User - we need to define what kinds of activities each did and then we'd have something even more meaningful.
Also, WebAnalyticsBook wrote about ComScore now segments Heavy, Medium and Light Users earlier today, where he said:
"..Definitely a great idea. Now advertisers will be able to figure out where to find the very attractive “heavy” users and target these specific websites. The Comscore Segment Metrix should make media buying a little bit easier".
I think ComScore's Consumer Segmentation is the right direction; as Sebastian points out, you need to know where to find the "Heavy" users that would come to your site (bearing in mind that a heavy user for one subject might be a light user for something else).
It's also another example of Quantum Analytics, I first heard this term earlier this year at AlwaysON NYC in How Does the Internet Change Media Metrics?
"..Dick Costolo, CEO, FeedBurner - As more people view content away from the original site it's up to the publishers to track it; the more things you try to measure the more questions you get (more like quantum physics).
Your hearing more and more the desire to measure "engagement" - and if you ask 5 people how to measure engagement you'll get 5 opinions. You can get more answers but those answers will lead to more questions.
People are going to prefer to spend where there is measurable ROI but every answer you get generates more questions (IE: traffic from Digg - spends a couple seconds vs a referral - will advertisers begin to not want pay for traffic from Digg but will for a referral - that's the kind of issues we come up with in the new metrics measurements)."
So, it's fair to say that in providing more information qualify how many Heavy, Medium and Light Audience Users your site gets - your going to end up having more questions about why that audience is a Heavy User, where they came from, what they're doing on the site and what they do after they leave the site."
