Stats
Well if they lie, oops sorry, I mean, have a rounding problem in their stats like this then it's hard to take any of their other claims very seriously, isn't it?
Biz customers with broken Acer kit might want to sling a call into the technical centre: then put on the kettle and make a cuppa, because they could be on hold for a lot longer than they were expecting. Certainly longer than the half a minute average response time that Jakob Jersild Olsen, veep at Acer's commercial division …
It seems unlikely that they lied and there is nothing in the article which is inconsistent about the reported call stats and the author's experience. It simply underlines that reducing a complex, highly subjective experience isn't very useful. This shouldn't come as a surprise since I assume a greater proportion of the the reg audience have some idea that human brains aren't really wired to easily process probability or any but the simplest of statistical analyses.
I depends what you class as hold / ASA.
I consumer often starts this when the get the 1st message, a call centre will class it as when you go through to the queue (hence On Hold).
Also it's ASA, sl calling up at 9am Monday morning, as opposed to 3pm on a Thursday, will get very different results.
Two minutes is much much better than the following experience with Dell (after 2 minutes of being transferred everywhere)
Them: Welcome to Dell support, how can I assist.
Me: I have a 6 month old monitor and the lamp has died.
Them: Have you updated the bios to the latest version?
Me:It's a MONITOR.
Them: I am sorry, unless you have the latest bios we cannot continue troubleshooting your issue.
Cue 15 mins of arguing with a script before hanging up and sending a lengthy tirade to our account manager, who knew less than the script guy.
From your small sample size an average wait time could be badly skewed by a one off event where their system went down for 30min. You can't get the opposite otherwise you'd be going back in time, so giving us the median wait time will be a more representative figure.
I imagine it probably doesn't make much difference to your results, but it will help us decide if your statistics mean anything. Alternatively if you let us know what the maximum and minimum wait times were along with your already stated average then we can make a guess about how badly skewed your data is from the paltry 12 data points.
Of course you could alternatively try and up your number of data points to something more significant (in the 1000s).