A little while ago, Nalates Urriah pulled together a set of statistics from diverse sources (all of which are duly credited) which help to paint a decent picture of where SL stands away from all the hype over falling region numbers, etc.
When taken together, the stats – which cover daily sign-ups, concurrency (daily / monthly), region numbers and even forum usage, all for periods of at least a year – present an interesting picture of Second Life which Nalates interprets in her own inimitable way. While they show that Second Life has in many respects been on a steady downward slide (particularly in terms of overall usage), the situation is far from unrecoverable. Indeed, some of the figures are, at least for a moment, trending upwards again – although without more detailed data and a wider breakdown, it is impossible to draw any conclusions as to what this might signify in the short-term and thus how it might be projected in the medium-, or long-term.
There are significant gaps in the data (through no fault of those who gathered / present it – the information simply isn’t available). For example, while sign-ups can be shown to have been at least constant (or have increased slightly) through the 2-year period, there is no practical context to the figures in terms of users actually being retained. A further problem with the figures is that there is no indicator as to the percent / proportion of these sign-ups actually being alternate accounts, rather than actual new users (although LL does apparently have a mechanism in place for distinguishing between the two).
Certainly, Rod Humble did state at in his first (and last) SLCC address in August 2011 the rise of user sign-ups did coincide with an upswing in identifiable uniques (i.e. genuine new users, rather than alternate accounts), which he clearly defined as people signing-up, downloading the viewer and logging in to SL.
The user concurrency chart is somewhat more meaningful, in that it charts concurrency for a more extended period from December 2009 through to the present day. As such, any trend shown is liable to be somewhat more reliable, although there are still problems in interpreting the data as a whole. For example, it does show a consistent downward trend in concurrency since the late “boom” period when SL was at the height of its own Hype Cycle “peak of over-inflated expectations”; but precisely what this means is still somewhat open to interpretation.
Less Doesn’t Automatically Mean Fewer
A decline in concurrency doesn’t automatically mean a substantial drop in overall user numbers (although it is hard to completely divorce the two). There have been a number of factors which have contributed to some aspects of the decline outside of falling user numbers. Linden Lab caused something of a decline when they clamped-down on the use of bots. More recently, factors such as changing demographics and changing user habits appear to have also contributed to falls in concurrency.
These latter points were indicated again indicated by Rod Humble in his SLCC 2011 address, when he drew attention to the fact that the overall demographic of SL users was shifting, age-wise, more toward people in the mid-to-late 20s, and that they were collectively logging-on to SL for shorter periods. He also indicated that LL had charted a noticeable increase in the way SL users were interacting without actually going in-world – through the mechanism of profile feeds, for example.
Inara Pey: Living in a Modemworld 