So Tomi uses Kantar again. I have no problem with it, since so far Kantar just proves him wrong.
He misused the Kantar data for Nokia’s Q1 2011 results, eventually just proving that Q1 was ruined without “Elop Effect”. (read here)
He misused the Kantar data for Nokia’s Symbian – Lumia transition, creating a pointless post that proves nothing like the things he tried to prove. (read here)
And this time Tomi says:
“Kantar Worldpanel surveys new smartphone sales in many countries, and releases their data on selected countries and at irregular intervals. It is always insightful to dig through their numbers when we get some.” 
Now this we agree. It is always insightful to see that Tomi is wrong. Especially if we can couple his own figures to Kantar and just prove him wrong with those. Like this time.
“This time Kantar includes the EU5 (Germany, France, UK, Italy and Spain the biggest 5 EU nations by population, GDP and smartphone sales) plus the USA, Australia, Brazil and Mexico. They give for comparison also the September numbers also for 2011, but that is not really the numbers we want to compare to.” 
Right… Makes kind of sense, the year-on-year comparison Kantar uses shows market share rise in GB, France, Italy and Brazil. So what did you plan to use, Tomi?
“Because we know how Nokia Lumia and Symbian did in Q2, we really want to compare Kantar numbers to the end of Q2. September is the last month of Q3, so we want the last month of Q2, June, which is also when Microsoft Osborned the Nokia Lumia series, crashing their sales. For that, we do have numbers from a previous Kantar release, for the EU5 and USA and Australia, for June 2012. I’d have loved to also have Brazil and Mexico for that period, but alas..” 
Right. Comparison kind of makes sense there. I mean – I REALLY want to point to you that Lumia sales did not collapse. So how were the figures? We know the latest  and the one that ends in June  so we can easily pull in the numbers, right?
Now one obvious trick to know: Kantar used to list Windows Phone and Windows Mobile as separate sales on their readings. Now it lists just “Windows”, which is Windows Mobile and Windows Phone combined. But the older figure is also sum-up of the old data so in a sense this tells us how Microsoft is doing, but has very little to do with Nokia sales (or Windows Phone 7 sales). Tomi uses the comparison nevertheless.
Let’s still stick along since Windows Mobile was selling very low numbers already in June this year, so it does not matter that much. Just remember that increase in market share is “extra” increase in unit sales since:
- Industry has been growing
- Windows Phone has eaten market share from Windows Mobile.
Now Windows market shares from Kantar:
Country Jun'12 Sep'12 GB 4.1% 4.4% Germany 6.8% 4.0% France 2.7% 5.7% Italy 6.2% 10.4% Spain 1.9% 1.6% US 3.6% 2.7% Australia 5.8% 3.8% EU5 ---- 5.0%
Now you may have already noted, Kantar did not publish EU5 figure in their report for June. So we do not know the figure. Tomi counted it for us and it’s reasonable (after the fix):
“WINDOWS SMARTPHONE SALES JUNE 2012 VS SEPTEMBER 2012 BY KANTAR:
REGION . . . . . . . . JUNE SHARE . . . AUGUST SHARE . . . DECLINE IN CHANGE PERCENT
EU5 . . . . . . . . . . . . 4.6
6.9 . . . . . . . . . . . 5.0 . . . . . . . . . . . . . 8% incline
USA . . . . . . . . . . . 3.6 . . . . . . . . . . . 2.7 . . . . . . . . . . . . . 25% decline
Australia . . . . . . . . 5.8 . . . . . . . . . . . 3.8 . . . . . . . . . . . . . 34% decline
Weighted Average . . 4.2 . . . . . . . . . . . 3.8 . . . . . . . . . . . . . 9% decline” 
Tomi Ahonen updated his post after I had made mine (EU5 figure fixed to 4.6M). My post has now been updated too. Sorry for the trouble.
Thank you, Tomi for fixing the numbers. Now what does this mean to Nokia?
“However, as Europe forms nearly half of Nokia’s total smartphone sales, and over half of Lumia sales in Q2, as per my analysis of the Nokia Q2 results;” 
Sorry – interrupting you. Let’s see your Q2 analysis:
“Nokia Lumia Regional Market Split
Region . . . Lumia Smartphones
Europe . . . 2.4 M . . (1.6 M)
MEA . . . . 0.0 M . . (0.0 M)
China . . . . 0.1 M . . (0.0 M)
APAC . . . 0.1 M . . (0.1 M)
N Am . . . . 0.5 M . . (0.3 M)
LatAm . . . 0.9 M . . (0.0 M)” 
Do you mind if I apply your growth/decrease percentages? You don’t?
Now we don’t know China or APAC from Kantar, but we do have LatAm related figures (the ones you left out):
Kantar September figures – Windows Market share:
Country Sep'11 Sep'12 Brazil 3.8% 12.2% Mexico 2.9% 2.7%
Now Brazil was included also in August figures of Kantar, where Windows was having over 14% market share. Therefore I would assume Windows Phone sales have not taken a significant dive in LatAm during Q3. Let’s keep them “as is” since we don’t have solid data.
How about China? I told in my previous post that we have installed base info of China, namely saying installed base increased by 0.4M units in July-August.  Let’s assume flat sales and we have 0.6M for Q3. Now Nokia holds 76% of Windows Phone market in China. That’s 0.3M for Jul-Aug and 0.45M for Q3. I’ll round down to 0.4M.
Kantar-based Q3 Lumia sales prediction, using percentage changes from Tomi Ahonen:
Region . . . Lumia Smartphones Europe . . . 2.6 M . . (2.4 M) MEA . . . . ? M . . (0.0 M) China . . . 0.4 M . . (0.1 M) APAC . . . ? M . . (0.1 M) N Am . . . . 0.4 M . . (0.5 M) LatAm . . . 0.9 M . . (0.9 M)
That’s 4.3M units, about 8% global growth. Not bad.
Of course the figures are way too low, since I used the growth/decline of market share and applied it to unit sales. You see, since market has been growing, the unit sales that produced 4.6% market share 3 months ago won’t give as much market share today. Equally, unit sales that produce 5.0% market share are way more units than those that would’ve given you 5.0% counted 3 months ago. So my unit sales numbers from now on will be too low.
Please Tomi, continue your analysis.
“If that ratio holds true for all of Nokia Lumia sales, as we saw in those 7 countries, then for Q3, Nokia Lumia sales will fall from 4.0 million to 3.7 million units.” 
“If Lumia still accounts for about 87% of all Windows Phone smartphones sold globally as they did in Q2 – and there is little to lead us to believe any other handset makers have suddenly released obsolete Windows Phone 7.5 handsets that would have become big hits – so if we say these 3.7 million Lumia sales are 87% of all Windows Phone sales in Q3, then Microsoft can celebrate its total sales in Q3 projecting from these Kantar sales numbers, as a whopping.. 4.3 million smartphones in total.” 
It’s amazing how far I got by adding B and C in “BRIC market” to the mix. Too bad we don’t see R and I here. 😉
So big difference here is: I see sales increasing (with numbers that are too low) for Lumia line that Tomi said was “Osborned”. The one for which Tomi predicted sales of 3.1M Lumias for Q3 2012.  The line that was not going to sell anywhere at all. I guess we indeed read the figures with different glasses.
This was fun but let’s stop here. We will soon get Q3 results and THAT will be interesting since Tomi already waved sales channel stuffing card for 4M sales of Q2. Had Nokia indeed shipped over 4M phones in Q3, that would eat the whole base from his “Lumia lime has been Osborned” mantra.
Also – even if unit sales would have been flat – it would mean Nokia has been able to STOP the downward spiral.
But that’s me – the optimist – speaking. I’ll let Tomi do the closing so we get some contrast to this:
“I would suggest getting some tissues to wipe out the tears, when Nokia reports its Q3 results, and if you were somehow invested in Nokia’s smartphone strategy with Stephen Elop’s mad Microsoft strategy and Windows Phone and Lumia, may I recommend a good bottle of Finnish vodka perhaps to drown out the misery. I can recommend a good Finlandia vodka if you are prone to the traditional, or if you are more adventurous, perhaps the ‘Black vodka’ ‘Salmiakki Kossu’ licorice flavored black Koskenkorva vodka.. That should help you forget the misery for at least one night..”