Originally Posted by pdq2
Cross-posted from the comments at CNET by someone who claims he is quoting the conference call transcript:
...and then the maths as above.
I think the CNET writer probably misinterpreted "record Android smartphone sales, both in total numbers and in the highest percentage of total sales" as Android phones having the highest percentage of AT+T sales overall, but he doesn't actually say that, and it appears to be incorrect
That commenter at CNET was me. It's even more interesting than this ... now that Sprint just released their iPhone numbers.
Take a look at this article posted at All Things D a few weeks back. When I read it, I was take aback that this research firm (CIRP) had estimates for Q2-2013 since, neither Apple or any of the US carriers had released any information. So I took it with a grain of salt, but since it was the only piece of information that was available at the time, I used it in my Excel model that I created for following AAPL stock.
So when VZ announced earnings and indicated that they activated 3.8 million iPhones, I used the CIRP data and came up with an estimate of how many iPhones were sold in the US ... which I use as a proxy to estimate the number of iPhones sold world wide.
The assumption that I made was that the 4 carriers only sold their version of the iPhone, meaning that you could not buy a Verizon iPhone at an AT&T store. I think this is a pretty safe assumption for estimating purposes. Next, I assumed that the percentage breakdown of the different carrier models at the retail stores mirrored the percentage breakdown between the 4 carriers. I'm less confident in this assumption, but I had no other basis.
OK, going back to VZ's announced 3.8 million iPhones activated ... this indicated that Apple would had sold 11.9 million iPhones in the US for the quarter. With a breakdown as follows.
- AT&T - 44.7% share - 5.3 million
- VZ - 31.8% share - 3.8 million (actual)
- Sprint - 11.4% share - 1.4 million
- TMobile - 12.1% share - 1.4 million
So when AT&T announced their earnings and chose to not disclose their iPhone activations, I had to scramble to dig deeper. The CNET article lead me to read the Conference Call transcripts as I couldn't believe what CNET was indicating. That led me to realize the author misinterpreted the CEOs statements. Furthermore, without any other data, it made it hard to determine the exact number of iPhones that AT&T activated in the quarter. We know it's more than 3.7 million and less than 6.8 million. That's a big gap, but without any more data, that the only safe assumption to go on.
So my table now looks like this.
- AT&T - 35.8% share - 3.7 million (quasi-actual)
- VZ - 36.7% share - 3.8 million (actual)
- Sprint - 13.5% share - 1.4 million
- TMobile - 14.0% share - 1.4 million
So here's where it gets interesting. Sprint announced this morning that they activated 1.4 million iPhones. Hmmm, I was fully expecting to have to update my model, but Sprint came in right where the CIRP data had indicated they would. So 2 out of the 3 carriers that announced matched the CIRP data and AT&T doesn't match, because they didn't give actual figures.
So I'm left to wonder if AT&T did actually activate closer to 5.3 million iPhones in the quarter. I haven't done my homework to see what "a record percentage of Android phone" means. Nor did I research what "a record number of Android phone" means either. If no other types of phones were sold, other than iPhones and Androids, a 5.3 million iPhone number means 1.5 million left for Android (it also means 22% share for Android). I'm pretty sure that Android phones on AT&T have seen higher than that.
Unfortunately, Apple doesn't breakdown iPhone sales by region ... and even if they did, I think they would mix in Canada. If TMobile comes in at 1.4 million as well, then it would lead one to believe the accuracy of the CIRP data and one could argue that AT&T activated closer to 4-5 million iPhones than 3.7 million.