I wonder how much of the industry analysts consensus numbers for 2006 are affected by that “loss” of 9.6 M Chinese 3G users?

The problem with forecasts is that almost all analysts track what the others also say about the same numbers and then often adjust their estimates if they are becoming too much out of whack with the consensus opinion. (ie its better to be wrong where everybody agreed on that error, than to be the only one wrong where the consensus seemed to be right).

This kind of massive mistakes in the basic assumptions will then ripple through other forecasts, where other forecasters may downgrade their overall Asian forecasts for 3G because suddenly one has dramatically dropped theirs…

Russell – ha-ha good point. Some of us professional forecasters in mobile are more diligent about returning to former forecasts, and show not only what all has changed (and why) but also revisit old forecasts over time and stand to be counted for our many errors and occasional near correct guesses.

Tomi Ahonen

In other words, when some analyst firm says that there will be x billion MMSs sent in 2007 or that mobile advertising will be worth $Y billion by 2010, is there anyone who tracks these for accuracy? It would be very interesting to see if any of these forecasts are based on any kind of real judgement, or if someone could do better with a random number generator.

One thing is sure – you don’t get many analysts crowing about how accurate their forecasts have been. Which leads me to supect that there’s really nothing to crow about.

So how come companies continue to fork out thousands of dollars on reports that are really pretty useless in the main?

Russell