This is (hopefully) my second last post about Elop Effect. I already pointed out it did not start the crash of Nokia, although I have admitted several times that Elop Effect caused lot of other things. (Heck, I called it “Communication mistake of the century”) I re-explained why market share is the proper tool to prove it. But I have been still keeping the best bit to last. This is it.


Let’s start with Tomi Ahonen telling us how Elop Effect killed Nokia Q1 2011 results (for those new to the subject):

While the industry grew smartphone sales from Q4 of 2010 to Q1 of 2011, Nokia lost unit sales in smartphones, saw declining average sales prices, lost revenues, declined profits. However, we knew this would happen. My blog was not the first one, nor the only one to explain why the February 11 announcement of a two-year shift from Symbian (and MeeGo) to Microsoft as the operating system for Nokia’s smartphones, would damage short-term Nokia smartphone sales.” [1]

He has been very certain about this. In fact, so certain that we can find him explaining it in the comments of his blog as late as April this year:

I have actually calculated, for the fun of it, the EXACT growth rate of Nokia from Q3 to Q4, divided by month, extended into January and February. Then I took the EXACT rate of decline from Q2 to Q3 – see it is the more steep decline. And split it by month, and used the same rate, and plotted it BACKWARDS into Q1 ie to March and April.

And those lines intersect almost exactly at mid-month of February. And then I counted the actual sales of Nokia smarthpones for China, and the rest of Nokia markets, and it comes out almost exactly square and correct. The growth of Nokia Symbian powered by S^3 and the N8 and E7, continued – GROWTH to mid-February, and from that point, the collapse is at that steep decline curve as we see from April on. If Nokia was reporting the data by month, it would be obvious. But you cannot provide ANY evidence of your hypothesis, that would conform to the math, or can you (I have tried out all the plausible scenarios, you know how nuts I go with my stats and math haha).

No, you make a good point TomiFan, but the truth is, that Nokia smartphone sales did continue its growth up until Feb 12 or so, and then collapsed. And that can be calculated and the math is consistent with all Nokia reported data including regional sales etc.” [2]

So what was this nick name TomiFan talking about? All so familiar graph from Tomi Ahonen:

Now I did go through this graph already once. However, that was hardly half of it. But for now, let’s all remember what Tomi just said: “Nokia smartphone sales did continue its growth up until Feb 12 or so, and then collapsed. And that can be calculated and the math is consistent with all Nokia reported data“.


Tomi’s Elop Effect needs the Q1 2011. If Q1 went down without Elop Effect, then his whole point is lost. What point, you ask? This point:

The Nokia Symbian based smartphone sales grew – not declined, grew – unit sales (the first measure of performance). The Nokia Symbian based smartphone sales grew – not declined, grew – revenues (the second measure of performance). The Nokia Symbian based smartphone sales grew – not declined, grew – average sales prices (the third measure of performance). And Nokia Symbian based smartphones grew – not declined, grew – profit (the fourth and last measure of performance). By every measure that the industry uses, Nokia Symbian smartphone sales grew from Q3 of 2010 to Q4 of 2010, literally the last full quarter before Elop released his Burning Platforms memo.” [3]

Something happened in February 2011, specific to Nokia and its smartphone market that caused strong continued dominating growth to instantly turn into the world-record-setting market collapse.” [4]

If collapse starts without Elop Effect, something ELSE happened before February that caused the collapse. That is unacceptable for Tomi – Elop must be the reason and needs to be fired (preferably now).

So first quarter of 2011 needs to be ruined by Elop Effect for it to be reason for collapse of Nokia. (Elop Effect will still be the cause of several issues, but it has not started collapse.) In reality, Q1 2011 is the last quarter Elop Effect could not ruin. You see, even though Tomi claims he has counted his curves and seen them intersect in mid-February, those curves should not intersect in mid-February.

I’ll explain why:

From text of Tomi Ahonen one would assume that when somebody bought a Nokia smartphone 11:59pm in last timezone of this planet March 31st, it would be counted in to Nokia reported sales of 1st quarter.

That, of course, is far from truth. Nokia has no means to track the unit sales from shop counters. That smartphone could have arrived to shop already in December last year, counting towards Q4 2010 results, just waiting for a buyer for 3 months. And Nokia does not count their unit sales towards successfully delivered shipments either. Nokia counts “units sold” from “units shipped“. The last smartphone that made it to Q1 2011 results is basically the one that left factory in a shipment sent before midnight, 31st of March. It may take weeks of ship cargo in the oceans before it hits shops.

Now I know Nokia has highly optimized logistics and factories near the sales areas. We can trust on them delivering with minimum transfer distance, so let’s assume Tomi took this into account in his “almost exactly at mid-month of February”. So I admit that all three – Tomi’s graph, Tomi’s “instant turn” and Tomi’s math that he claims to be “consistent with all Nokia reported data” do not get broken by shipment times.

As said, Nokia does not sell smart phones from factories to consumers. Consumers buy their phones from the retail/operator stores. Customer rejection can not crash unit sales of Q1 2011 – retail can. And Tomi has been very specific that retail started boycott against Nokia. So let’s look at retail next.

Now retail chains/operators make contracts with Nokia and later they get a shipment. They sell out the shipment and order some more. But devices do not appear to retail instantly either. They do not even come out from the factory the very same day. If retail makes a contract with Nokia today, Nokia will ship the devices weeks later. In fact, the delay is four to six weeks.

In case you didn’t catch it, I say it out loud: All smartphones counting towards first quarter of 2011 unit sales were ordered during February 2011 the latest. Even if all retail shops around the planet would not have ordered any Nokia smartphones after February 11th, Nokia would have shipped close to same unit sales in Q1 2011. The first shipment date to be impacted is March 11th, 4 weeks after the February 11th. Most likely first affected date is March 25th, 6 weeks later. That leaves 6 days of the quarter without any shipments – at worst.

Sounds unbelievable? Have you ever wondered how Nokia knew their sales for Q2 2012 will be down already at April, when Q2 had ran only 11 days? [5] Now you know. And you would kind of expect that “former Nokia executive who lectures at short courses at Oxford University and is regularly quoted in the press in over 400 articles published in over two dozen languages on all six inhabited continents” would know that too.
Somehow Tomi seems to forget these things from time to time. At least when there is a quarter he has to count in.

Not to mention him saying that “the industry grew smartphone sales from Q4 of 2010 to Q1 of 2011” [1]. We saw from my post about market shares that according to official Nokia figures and numbers given by Tomi Ahonen, industry grew practically zero during Q1 2011.


EDIT: One of the frequent commenters Xizzhu was kind enough to remind me that I did not take into account the retail cancellations/returns. In fact, I did look for that option too, but dropped it from blog. Since he brought it up, it’s a good idea to cover that issue too. This chapter was added for that purpose.

So how about if Tomi is right about the intersecting lines? Let’s first see how much there should be incline:

If you asked me, what is the fair base case to use, I would take the last full quarter (Q4 of 2010) before the Burning Platforms memo, and take the trends of that period, and project from there. So if Nokia smartphones grew 7%, the reasonable assumption is that the 7% rate would continue into 2011. If all things stayed the same, then Nokia would continue about 7% growth rate each quarter in 2011.” [3]

Tomi says 7%, so we go by 7%. Problem is: when to start? I doubt retail can “cancel their order” if ordered devices are already coming out from production line. Honestly, I don’t know what kind of time window there is to do such a thing. So I’m now assuming that retail actually CAN cancel orders on the very day of Feb 11th 2011.

By 10th of February we have spent 41 days out of 90 days of Q1. If smartphone unit sales were 28.3M in Q4 2010 and we expected them to be 30.3M in Q1 2011, it’ll be average of 336 thousand devices per day, but naturally in increasing line. Since average for Q4 2010 would be 308 thousand devices per day, we’ll get rough line to start at 322 thousand devices per day, ending to 350 thousand devices per day, leading to average of 336 thousand devices/day. For first 41 days Nokia shipped therefore 13.5M devices.
Now remember, we start from assumpion that all 30.3M devices had been ordered and drop of shipped devices is cancellations/returns only. At February 11th when cancellations start to pour in, Nokia has 16.8 million devices to go for the Q1 production. Now we know that actual shipped amount of devices for Q1 was 24.2M devices, not the wanted 30.3M. That means Nokia ends up producing only 10.7M devices during rest of the quarter. 6.1 million orders get cancelled. That’s 36% of the planned production Nokia would have had left to go. More than every third order needs to be called off.

Now cancelling orders at that stage most likely include some sort of penalty fee. That kind of cancellations “en masse” would be visible in Nokia’s quarterly report of Q1 2011, right? Unfortunately I cannot find anything like that in the Interim report. Sure enough, we have this: “On a sequential basis, the 12% decrease in our global mobile device volumes was primarily due to lower seasonal demand for our devices and an intense competitive environment, offset to some extent by improved component availability.” [8] That’s the reason why industry practically did not grow at all during that quarter – lower seasonal demand. Intense competitive environment is the reason of market share collapse – i.e. reason why Nokia’s unit sales fell during that quarter.

No, we don’t have any Nokia announcements, news headlines, nothing suggesting that there would have been 36% cancellation of Nokia orders. Nothing of that sort. Far less massive return rates of millions of devices, which at least would have leaked to news in form or another.
(Which in the end is the reason I originally did not cover this issue at all.)


But let’s get back to graph. What Tomi keeps on saying us all the time is that Nokia was larger than its two closest rivals combined. He has Samsung and Apple in that graph well put. And he said already in January 2011 following (talking about Q4 2010 results):

Nokia still today is bigger than RIM and Apple, the two nearest rivals making smartphones – combined.” [6]

Wait, what? RIM and Apple? RIM mentioned first? Why is he showing us Apple and Samsung now, if he talked about RIM and Apple in January? Where’s RIM?
(Trick question, Nokianonymous already told us where RIM is.)

But let’s try different graph. Here’s RIM for you:

(Image copyright

These are unit sales (so those who demand unit sales are happy). Feel free to compare to Tomi’s graph, it’s consistent.
Nokia grew unit sales Q3 and Q4 2010. So did Motorola, RIM, HTC and SonyEricsson. But Elop Effect was so strong it crashed Nokia, RIM and Motorola!
Note: RIM reports its Q1 2011 results from time period of December, January and February, not Jan-Mar as others. Therefore Mid-february “Elop Effect” naturally affects Q2 2011 results, not Q1 results. (Unless this all was caused by something from outside those three companies.)
None of the other companies informed they will switch platform. (Osborne effect)
None them had CEO saying their products were crap. (Ratner Effect)
None of them had Elop on board. (Elop Effect)
All of them lost unit sales, while still increasing sales in Q4 2010.

Actually, Only smartphone manufacturers increasing their unit sales after Q3 2011 are Samsung, Apple, ZTE and Huawei. And remember when Nokia was larger than Samsung and Apple combined? Well, today Samsung and apple combined are larger than rest of the industry! This space was not “released” by Nokia. Look at the growth of those two monsters! Did Elop Effect make those two grow with such a speed?

So Tomi chooses to show us Apple and Samsung, skipping RIM and HTC that are fighting for 4th position of the industry, RIM being larger than Samsung to begin with! Anyone smells fouled-up data?
Yep, FUD it is.


Elop Effect did not start crash of Nokia’s market share.
Elop Effect did not start unit sales drop of Nokia.
Elop Effect did not cause crash of Nokia.

Still, I already said it is “cause of several issues”. I explained it to some degree in my post about Qt, MeeGo and Meltemi. I will cover effects of it much further in my (hopefully) last post of the Elop Effect. It’s under work.










Guideline for commenting: I hate the way Tomi Ahonen deletes criticizing comments from his blog. However, I plan to follow three of his principles: I’ll delete comments that are

  1. Personal insults to someone
  2. Duplicates
  3. Spam

In addition, if you wish to challenge my previous posts, please comment to those.