Thursday 16 September 2010

Lies, Damned Lies & Mobile Statistics

The mobile industry is one of many which seems utterly obsessed with statistics. It doesn't matter what they are about, someone will horribly abuse them to make a point. There are plenty of statistics which show that Apple has an absolutely unassailable market position, just as there are plenty which show that Apple has a relative low market penetration. You can use these however you want to justify any position you wish to take. The problem is that it can be very difficult to see through ill-conceived headlines and soundbites.

In the past week, I've been confronted with statistics which just didn't seem to add up. Firstly on Twitter, Ewan MacLeod retweeted (original Tweet by Ken Shimada) this message:
http://twitter.com/kei_shimada/status/24344277211

on iPhone: BCN reports that 22.6% of July mobile handset sales were Smartphones. 80% of them were iOS based, and the remaining were Android.
I was surprised by this. We all know that iPhone and Android sell well, but they're not the only players, and frankly I don't believe that RIM and Nokia each only sell such a small amount as to be negligible in the stats. I replied with:
http://twitter.com/rival/status/24361145816
@kei_shimada @ew4n There were no BlackBerry sales? My Mobile magazine keeps telling me that BB Bold and Curve top contract lists
This is true. I get Mobile Magazine through the post every fortnight. Each issue contains a breakdown of UK retails sales for the previous week, both prepay and postpay. Postpay top ten is always dominated by RIM, and they often feature in Prepay. Prepay is dominated by cheap Nokias. I've never seen Apple on this list, not once. Actually, I don't necessarily read every issue, but I have still never seen Apple in the UK top ten retailing handsets, and it isn't even as if their sales are divided across a wide range of phones. I know that this is UK only data, but what I've heard anecdotally is that Apple sells lots of phones, but they're still not on the list. There is some back and forth between Ewan and myself, and then I get a reply from Ken.

@rival @ew4n these are Japanese stats.
OK, so now we're getting somewhere. There are two problems becoming clearer to me. Firstly, that Twitter encourages us to just write down the bare bones of what we want to say. We don't have the space to write a lot of qualification for statistics. Secondly that people will easily bend stats without explanation. Ken followed this up with:


@rival @ew4n read carefully, nobody said market share.
No, nobody did say market share, but the did start talking about shipping numbers. Now that I know I'm not comparing worldwide sales and UK sales, it starts to make more sense.

Ken then sent this absolute classic:

@rival @ew4n not all stats are bullshit if you understand what they mean :-)
That is my point. Stats are obviously not bullshit if we understand what they mean, the problem is that stats are being banded about without any qualification. Without even mentioning which market the stats we're for. Without any visibility of the underlying data allowing independent analysis. Stats like this are bullshit. They don't aid anyone in understanding a market, they are distortions of the truth.

Anyway, yesterday I found this blog on TechCrunch (yeah, I know, its my own fault).


Apart from the immediate nonsense of it, it has this to say.

Nokia’s stock has been hammered by its failure to gain real traction in the high-end segment. Its US listed shares have tumbled to just under $10 a share as of Wednesday morning trading— from an all-time high of more than $60 back in June 2000. Furthermore, although Nokia proudly waves the number “260,000,” the number of Nokia smartphones sold per day according to the company, the figure distorts reality.

Nokia is scrambling to defend its market share. On the high-end it’s dwarfed by RIM, Google and Apple in the United States— Symbian only captured 2% of the smartphone market in the first quarter, according to Nielsen.

The figure only distorts reality if your reality only exists in the United States. What Evelyn Rusli is saying here is that it doesn't matter that Nokia shifts 260,000 smartphones (not feature phones) each day, but that they only have 2% market share in the United States. Yeah Evelyn, the rest of the world are just wrong and will come around to the American way of thinking.

Gartner has a more complete view of the worldwide smartphone market. See Table 2 here http://www.gartner.com/it/page.jsp?id=1372013

Table 2
Worldwide Smartphone Sales to End Users by Operating System in 1Q10 (Thousands of Units)

Company

1Q10

Units

1Q10 Market Share (%)

1Q09

Units

1Q09 Market Share (%)

Symbian

24,069.8

44.3

17,825.3

48.8

Research In Motion

10,552.6

19.4

7,533.6

20.6

iPhone OS

8,359.7

15.4

3,848.1

10.5

Android

5,214.7

9.6

575.3

1.6

Microsoft Windows Mobile

3,706.0

6.8

3,738.7

10.2

Linux

1,993.9

3.7

2,540.5

7.0

Other OSs

404.8

0.7

445.9

1.2

Total

54,301.4

100.0

36,507.4

100.0


So, now we can see a much better picture. We have actual data, not stupid headlines, not misleading quotes. Worldwide, Nokia is in decline, but it still is by far the largest supplier of smartphones, selling just about as many handsets as iOS, Android and RIM put together.

Now, compare and contrast these three statements:

on iPhone: BCN reports that 22.6% of July mobile handset sales were Smartphones. 80% of them were iOS based, and the remaining were Android


Furthermore, although Nokia proudly waves the number “260,000,” the number of Nokia smartphones sold per day according to the company, the figure distorts reality.

Nokia is scrambling to defend its market share. On the high-end it’s dwarfed by RIM, Google and Apple in the United States— Symbian only captured 2% of the smartphone market in the first quarter, according to Nielsen.
Nokia is with world's leading manufacturer, selling 260,000 smartphones worldwide each day.

Lies, damned lies and statistics. I once had an argument with someone who responded with "Facts? Don't give me facts. You can prove anything with facts." That stopped me in my tracks, but I now think it may be a quote from something. At the time I thought that the person was an irrational fool, but every day I'm coming round more and more to that point of view.

No comments: