I promised in Twitter to post about N9 sales. Of course this is due to Tomi Ahonen. He started this on January 26th 2012:
“Luckily I didn’t have to do the math for this, the nice people at All About Symbian had tracked the numbers (read through the comments) and calculated the limits, finding N9 sales to be between the level of 1.5 million and 2.0 million units in Q4. Wow! Nokia specifically excluded all of its richest and biggest traditional markets where it tried to sell the Lumia, and these countries achieved – lets call it the average, 1.75 million unit sales of the N9 in Q4. So the one N9 outsold both Lumia handsets by almost exactly 3 to 1.” 
Now for Lumia sales he compares to his own calculation of 600 000 in Q4, which naturally is not verifiable, but I’ll look at that too. For starters we can just remember that official reading of Lumia sales for Q4 2011 is “well over 1 million Lumia devices to date” – where date is January 26th 2012. 
Tomi rounds this down to 1 million and justifies his number by saying that it’s 600 000 for Q4 2011 + 400 000 for January 2012. I did my own guesses in recent post (here) and I proposed 700 000 to 900 000. No big difference but read that post and you see why it matters. 😉
But now this was not about Lumia sales, this was about N9. And for N9 Tomi remembers to always mention how N9 was sold on very small markets like New Zealand. Guess what? The countries where N9 was sold include every “BRICS” country save India. Actually, Nokia uses term BRIICA – Brazil, Russia, India, Indonesia, China, Africa. Of those, N9 was sold in ALL but India. Yes, N9 was launched in China – world’s second largest smartphone market – where Lumias arrived way later! A phone that is sold in thirty seven countries including likes of China, Russia, Brazil and Italy is being sold in “such far-away nations as New Zealand (population 4 million) or Singapore (4)“? Give me a break.
Then to actual sales. Tomi says he didn’t do the math since “the nice people at All About Symbian had tracked the numbers“.
Wait, what?!? Tomi Ahonen, world-class market analyst, over ten times author of Mobile does not do the math himself? This has to be reliable data!
So do we get numbers from some marketing professional guest writer in All About Symbian?
Or is it just regular article using list of verifiable sources?
Or perhaps they do some more detailed article that is so good he chooses to use it without checking the numbers?
No – nothing like that – two commenters in All About Symbian did their own math! I’ll list the thing here:
“N9 outsold lumia, thank you very much.
Page 10 of the report.
‘On a sequential basis, the increase in our Smart Devices volumes in the fourth quarter 2011 was primarilydriven by the broader availability throughout the quarter of the Nokia N9 and the shipments during thequarter of the Nokia Lumia 800 and 710 in selected markets, as well as increased seasonal demand for ourdevices.’
Under the following conditions need you be reminded
-Heaviest promoting every (buy lumia get free xbox)
-Released in the biggest smartphone markets (US excluded) where nokia and Microsoft considered themselves to be strong as a brand.
-Christmas season most lucrative season in the world!
1.3 million sales and January was counted for lumia sales! (to date!)
It’s a joke
N9 was sold under these conditions:
-Little or no actual marketing
-Anounced in few select markets with very few people with disposable income
WILLING to spend 1000US Dollars on a phone
-Really high price
1.5 million sales, January NOT counted.
It’s a succes.” 
Further explained by P (someone disagreed suggesting 500 000 units of N9 instead):
“Did you read the report at all? Symbian sales in Q4 declined. Q3 was 16.8 smart devices. No N9 and Lumias were on sale back then, page 9.
Sales for symbian declined in Q4 over Q3 so even if you leave the symbian sales at 16.8m eventhough it declined.
2.8-1.3=1.5m Nokia N9.
I don’t see how you make up that 500k.
N9 sold at least 1M” 
Also covered by Shaun Murray:
“Lumia figure must be much less as the ‘over a million’ figure includes another month after Q4. Rafe’s guess of 600-700k is probably as good as any.
So that puts the N9 at 2 million roughly.”
Now this is indeed comments in a web forum. And after these few comments we get one more comment:
“P – you are reading your quoted paragraph completely wrong
Nowhere in that report (earning release pdf), in no single place, ever, Nokia says that they sold less Symbian devices in Q4 then they did in Q3 2011.
The only decline in Symbian device volumes in Q4 Nokia is talking about is year on year decline from Q4 2010. When they launched S^3, N8 and sold 28.6M Symbian devices. That’s it.
They never ever talk about sequential Symbian sales decline from Q3 to Q4 2011. Never. In fact, in that paragraph you quoted:
‘On a sequential basis, the increase in our Smart Devices volumes in the fourth quarter 2011 was primarily driven by the broader availability throughout the quarter of the Nokia N9 and the shipments during the quarter of the Nokia Lumia 800 and 710 in selected markets, as well as increased seasonal demand for our devices.’
Nokia says the opposite thing. They say that Symbian device sales INCREASED from Q3 to Q4.
Let me parse that paragraph with some comments:
On sequential basis – meaning from Q3 2011 to Q4, Smart Devices volumes (meaning Nokia smartphones including Lumia, N9 and Symbian) increased. Obvious. They increased because of 3 things:
—1. “Broader availability throughout the quarter of the Nokia N9” – So yes – they have sold more N9s in Q4 then they did in Q3. N9 – started shipping in Q3 – Sept. 27th – so some N9 sales were recorded in Q3 numbers. Still – I fully agree the majority of N9s sold in Q4.
—2. “The shipments during the quarter of the Nokia Lumia 800 and 710 in selected markets” – yes, all those Lumias, however little there were of them, added to the overall Q4 number
—3. “Increased seasonal demand for our devices.” There are no other Nokia smart devices, except for Symbian, to have a seasonal demand yet. “Increased seasonal demand” means higher demand during Q4 (Christmas shopping), as compared to demand in Q3, and which happens every year. Nokia only had Symbian in Q3. Except for a few N9s, but they are already accounted in #1. And, Nokia states very clearly here, that that increased seasonal demand for Symbian phones, which means more Symbian device sales, was 1 of 3 reasons why overall smartphone sales increased from 16.8 to 19.6 in Q4.
How can you interpret that as an evidence that Symbian sales decreased in Q4, and not that they have actually INCREASED in Q4, I have no idea.” 
…which eventually cools the discussion down.
Now I have no clue if Tomi read the comments before or after the time Staska posted his reply. But the fact is he ripped sales figure of 1.75 million from two Internet commenters and never looked back.
It is rather convenient for Tomi that the counter argument is from Staska. Those two have disagreed in the past too. It may be reason for Tomi’s ignorance.
However, Staska is the one who is right. P and Shaun were completely lost.
WHERE ARE THE REAL NUMBERS?
So that much for credibility of the figures of Mr. Ahonen. Now people at this point usually go and say:
Oh yeah? Well, I don’t see YOU giving any better numbers!
…as if I need to? The numbers were so badly goofed up I could throw 6 random digits and be more accurate. Frankly, I said it some time ago:
But let’s play along this time. I need something a bit more credible than anonymous blogger (me). Something more objective than known Microsoft-hater (Tomi). What could we try?
Tomi usually pulls in Canalys or Gartner.
Canalys says 1.2 million Lumia and 600 000 units of N9 for Q4.  “Other OSes” worldwide shipped amount is 1.8 million (includes also others than MeeGo, but not Windows Phone).
Gartner does not go to such details but for “other OSes”, it uses 1.2 million.  So even less than Canalys. (It may be Canalys included Windows Mobile OS to their “other OSes”, table naming is unclear.)
Numbers are not allowing 1.75 million N9, far less 2 million.
Also note: Canalys apparently uses the Nokia official “well over million up to date” statement at end of January for Lumia. Real number is probably below 1M, since Nokia did not say “1 million in Q4”. Don’t know if that is the case but 1.2 million sounds too high and this would explain it.
That, however, does not fix the fact that Gartner number cannot house N9 sales of 1.5 or 2.0 million and Gartner did not do Lumia estimate.
And of course we can try Nokia CEO.
He answered the question in shareholder meeting and we can find coverage from MyNokiaBlog (Tomi used quote from the same article but he left this part out):
“The CDB article on the Nokia AGM does not mention Nokia N9, but in an early post (http://communities-dominate.blogs.com/brands/2012/05/what-do-we-now-know-after-nokia-shareholder-meeting-that-the-future-is-far-worse-than-we-thought.html) there it was estimated that the N9 sold 1.75M in Q4 and 2.2M in Q1, give or take a margin of error. More so, it was argued there that the N9 has outsold Lumia (which we know sold more than 1M by late January and more than 2M during Q1).
This was actually asked about from Stephen Elop at the meeting and he put the question to rest in no uncertain terms: while agreeing that it is not a fair comparison because N9 launched in smaller markets and with less marketing, he did say Lumia outsold the totality of N9s sold very quickly after its launch – in a matter of weeks I think he said.
This would seem to concur with the likes of the Canalys analysis which estimated around 600K N9 sold in Q4 and then probably less than that in Q1. So, obviously the N9 did not, according to Elop, sell 2.2M or anywhere near that in Q1. The N9 probably did outsell Lumia in Q4, though, since it launced much earlier.” 
Now I got good comment on pointing out the possibility that Elop is lying.
Sure, he could be. But considering how well Nokia is leaking info (Burning Platforms Memo, Lumia 800 presentation,…) would he take a risk to lie about something that could leak the very next week?
I’d say we can put the question to rest here too.
BUT THE LIE CONTINUES
Now MyNokiaBlog caption already mentioned Tomi’s numbers for other quarters.
He indeed has listed those too, based on natural sales increase/decrease of a newly launched product, assuming the Q4 2011 sales were…
…1.75 million units. *sigh*
Q4/2011 1.75 million  Q1/2012 2.2 million  Q2/2012 1.2 million 
Source for last two? Gartner? IDC? Canalys?
No, none of them gave MeeGo numbers.
- Gartner says 1.2 million in its “other OSes” number for Q1 2012 
Quite different compared to 2.2 million N9’s by Tomi Ahonen.
- Gartner gives 860 thousand units for “other OSes” what comes to Q2 2012. 
Canalys says 1.2 million units (once again, could include Windows Mobile, unclear). 
Gartner is quite far from from 1.2 million MeeGo sales. For Canalys it would mean MeeGo was the only non-mainstream OS sold.
So where is Tomi pulling the numbers? He gives us source this time. Same for both, in fact:
“Source: TomiAhonen Consulting Estimates” “from vendor data and other sources” 
Wow. That’s a source to refer to if you are Tomi Ahonen. The “other sources” apparently count out the strategy houses he usually pulls in whenever he wants to prove someone wrong.
So I’m putting this issue to rest. Assuming we don’t get official numbers from Nokia, we can at least say Tomi Ahonen is pulling these numbers from his hat.
(I don’t know if the rabbit living in the forementioned hat produces these numbers. Could be.)