Alex Burmaster, Communications Director, UK & EMEA, Nielsen’s online division
The Pareto principle, more commonly known as the 80/20 rule, is the idea that roughly 80 percent of activity will be accounted for by 20 percent of the participants. Vilfredo Pareto’s initial observation at the start of the 20th century, that 80 percent of Italy’s land was owned by 20 percent of the population, has become a common rule of thumb in business, but does it hold up when tracking activity for the U.K.'s most popular social networks?
To see if it does, we equated activity with ‘time spent’ to reflect who is posting or consuming content. The U.K. audience to each site in January 2010 was divided into groups: ‘light’ visitors – those spending no more than five minutes on the site during the month; ‘medium’ visitors – those spending between five minutes and 60 minutes: ‘heavy’ visitors – those spending more than 60 minutes. It’s worth noting that equating activity to ‘pages viewed’ instead of ‘time spent’ produced similar results to the graphs below.
Grouping Twitter’s U.K. audience in this way reveals ‘light’ users account for 67 percent of the site’s audience and average less than 2 minutes on the site during the month. The ‘medium’ users account for 26 percent of the audience, averaging over 22 minutes. The ‘heavy’ users’ account for 7 percent of the audience but average almost 6 hours and 17 minutes.
Pareto would suggest that 20 percent of the visitors to a site would account for 80 percent of the total time spent on that site. However, for Twitter in the U.K., almost 80 percent of the time spent on the site is accounted for by the 7 percent of ‘heavy’ visitors. Conversely, 93 percent of the people who visit Twitter account for just 20 percent of the content posted and consumed.
Twitter's activity metrics are not unusual when compared with other social networks. Three percent of MySpace’s visitors account for 63 percent of total time spent on the site; 14 percent of Bebo’s visitors account for 87 percent of total time. For LinkedIn, 5 percent of visitors account for half the total time. Half of Facebook’s U.K. audience (52 percent) account for almost all (98 percent) of the time spent on the site. This is because during the month of January, Facebook’s ‘heavy’ visitors averaged 14 hours 20 minutes, their ‘medium’ users (32 percent of the audience) averaged 30 minutes while their ‘light’ visitors (16 percent) averaged less than 2 minutes.
In Facebook’s case, where the average time per person is so high, defining a heavy user as someone who ‘only’ spends at least one hour on the site might downplay the significance of the remaining 48 percent of visitors. Whilst they account for just 2 percent of the time this 48 percent still represent over nine million minutes of activity on the site during the month. If we changed our parameters and defined a ‘heavy’ Facebook visitor as spending at least five hours on the site, it would show that 26 percent of Facebook’s visitors are ‘heavy’ and that they account for 88 percent of total time on the site.
In fact, when looking at a distribution graph of the U.K. Facebook audience segmented by how much time they spend, something very interesting happens. Normally, as the time segments increase, the percentage of the audience in each segment decreases (i.e. the majority of any site’s audience are people who don’t spend very long and very few people spend a large amount of time). However, for Facebook, we see the distribution graph follow the standard pattern until we get to the 9+ hours a month visitors, when the curve starts to go up relatively dramatically.
The 20+ hours per month visitors are the fourth biggest segment in Facebook (almost 10 percent of the site’s audience falls into this band.) Only the ‘less than 1 hour’ (48 percent) and ‘1-2 hours’ per month (12 percent) segments are larger.
The downside for publishers, not just social networks, in the “how popular I am” stakes, of course, is that a large part of its audience (the ‘light’ users) probably aren’t going to be of much value to advertisers. The flipside is they have an extremely hardcore group of engaged visitors who are extremely attractive to advertisers. This latter group are particularly ripe for a fundamentally new approach to the online display ad model in terms of both ad units and ad inventory that was proposed in our 2009 Global Faces and Networked Places Report. A successful ad model, particularly for smaller social networks, being one that is built on a deeper, richer two-way conversation between an advertiser and a consumer that adds value to the latter.