Smartphone Wars: Q3 Scorecard - All market shares, Top 10 brands, OS platforms, Installed base

Its time to do the Quarterly update for the Smartphone Wars. First off, the overall market size. I told you last quarter that I felt the big industry analyst houses had it maybe too low, but based on their consensus view of Q2 unit shipments, I downgraded my annual shipment forecast. Now we have Q3 numbers (the average now works out to 354.7 million units worldwide) which is 5% growth compared to Q2. I do think the Q2 numbers were a bit too conservative and I can come back to my original forecast for 2015 total shipments of 1.55 Billlion units and yes, can be a bit below that so lets say its now between 1.53B and 1.55B. (I know these numbers are important to some of our readers here, who have their own models back home..)

So lets do the numbers. The Top 10 largest smartphone brands for Q3 ie July-September Quarter of 2015 are as follows:

BIGGEST SMARTPHONE MANUFACTURERS BY UNIT SALES IN Q3 2015

Rank . . . Manufacturer . Units . . . Market Share . Was Q2 2015 . . OS systems supported (coming)
1 (1) . . . Samsung . . . .  84.2 M . . 23.7% . . . . . . . ( 21.5% ) . . . . . . Android, Tizen
2 (2) . . . Apple  . . . . . . . 48.0 M . . 13.5% . . . . . . . ( 14.0% ) . . . . . . iOS
3 (3) . . . Huawei  . . . . . . 27.4 M . . . 7.7% . . . . . . . (   7.6% ) . . . . . . Android (Tizen)
5 (4) . . . Lenovo .  . . . . . 18.8 M . . . 5.3% . . . . . . . (   4.8% ) . . . . . . Android (Tizen)
4 (5) . . . Xiaomi . . . . . . .18.5 M . . . 5.2% . . . . . . .  (   5.6% ) . . . . . . Android
6 (6) . . . LG  . . . . . . . .  . 14.9 M . . . 4.2% . . . . . . . (   4.2% ) . . . . . . Android
7 (7) . . . ZTE  . . . . . . . . . 14.2 M . . . 4.0% . . . . . . . (   4.1% ) . . . . . . Android, Firefox
8 ( -) . . . Oppo . . . . . . . . . 12.5 M . . . 3.5% . . . . . . . (   2.0% ) . . . . .  Android
9 (10) . . Vivo . . . . . . . . . . 12.1 M . . . 3.4% . . . . . . . (   2.8% ) . . . . . . Android
10 (9) . . TCL/Alcatel . . . .  10.3 M . . . 2.9% . . . . . . . (  2.3% ) . . . . . . Android
Others . . . . . . . . . . . . ..  93.8 M
TOTAL . . . . . . . . . . ..   354.7 M

Source: TomiAhonen Consulting Analysis 30 Oct 2015, based on manufacturer and industry data
This table may be freely shared

So Samsung takes a little step up in sales and market share for a change. Apple continues its annual pattern where Christmas sees the 'step jump' and the rest of the nine months sees an annual erosion in market share, this is nothing unusual. For Huawei I adjusted the Q1 and Q2 numbers because we learned the first half sales were 50M and now we know the immediate period before and after, so I have the four consecutive quarters as: 24.4M, 24.4M, 25.6M and 27.4M (this again for those who are following the math back home). This pattern makes more sense than the see-saw we had before for Q1 and Q2.

There is a bit of jostling for fourth place with Lenovo taking that slot from Xiaomi. Then the fight is on the bottom of who is a global Top 10 manufacturer and who isn't. Coolpad is dropped out and as I warned before, China's Oppo was on the bubble and now they are in at 8th ranking. Who is next coming from China? Meizu. Where is the first non-Chinese brand? India's Micromax. And a little side-note on the previous giants, for this Quarter Q3, Japan's Sony (which includes remnants of Sweden's Ericsson) has overtaken Microsoft's Lumia (which is the rump of Nokia's old smartphone unit). Sony sold 6.7 million and Microsoft Lumia 5.8 million smartphones, but both are far out of the Top 10. Both former giants are still shrinking in size of quarterly shipments and have trivial market shares; and both units are unprofitable so its a battle of loser and "loser's loser". Next lets do the OS battle. No 'battle here, so just lets get it over with:

Smartphone OS Shipments Q3 of 2015
Android . . . . 84.2%
iOS . . . . . . . 13.5%
Windows . . . . 1.7%
Blackberry . . . 0.2%
Tizen . . . . . . . 0.2%
Others . . . . . . 0.1%
Total . . . . . . 354.7M
Source: TomiAhonen Consulting Analysis 30 Oct 2015, based on manufacturer and industry data
This table may be freely shared

So yeah, we have Tizen in the charts... Note its statistically tied with Blackberry, but as Tizen didn't announce any number and Blackberry did, well before Samsung results, I expect Tizen did not ship more than BB, so the Tizen number is likely below 800,000 units for Q3. Its by selling one smartphone, the Tizen Z1 by Samsung, but now for Christmas, Sammy has launched their second Tizen phone, the Z3 and they did promise it would be sold in more than the two countries where the original Z1 was available.. We will have to see. On the top, this race was long since decided, Android won. Apple does a nice niche market for the premium price end with iOS. Windows continues to die, now they are down to 1.7% global market share in new sales and as I predicted, now yes, their market share has fallen to below 2% for the first time. The developers have not just stopped making or supporting Windows Phone apps, now many are pulling their old apps away from the Windows app store. It is becoming a Zombie extinction (wow, that is a phrase I never expected to write. And for casual readers, yes 'zombie' is what they call an app that has been abandoned by its developer but still lingers, without support or upgrade, in the app stores. More than 70% of all apps even at iPhone App Store and Google Play, are zombies. But there is an extinction of zombies happening now at Windows Phone app store. That is the very end indeed). Then lets go to the installed base.

INSTALLED BASE OF SMARTPHONES BY OPERATING SYSTEM AS OF 30 SEPT 2015

Rank . OS Platform . . . . Units . . . Market share  Was Q2  . Main Manufacturers of current base
1 . . . . Android . . . . . . . 1,808 M . . . 76 % . . . . . . (  76%) . . . . . . Samsung, Huawei, Sony, ZTE, LG, Lenovo/Motorola, Xiaomi, TCL-Alcatel, Vivo, Coolpad 
2 . . . . iOS  . . . . . . . . . . . 463 M . . . 20 % . . . . . . ( 20 %) . . . . . . Apple
3 . . . . Windows Phone . . 45 M . . . . 2 % . . . . . . (   2 %) . . . . . . Microsoft(Nokia)
4 . . . . Blackberry . . . . . .  19 M . . . . 1 % . . . . . . (   1 %) . . . . . . Blackberry
Others . . . . . . . . . . . . . .  31 M . . . . 1 %
TOTAL Installed Base . 2,367 M smartphones (2.4 Billion) in use at end of Q3, 2015
Source: TomiAhonen Consulting Analysis 30 October 2015, based on manufacturer and industry data
This table may be freely shared

Nothing dramatic here. The total installed base of smartphones in use is already 2.4 Billion. Android is at 76% and still growing (but growth now so slow it was in fractions of one percent). Meanwhile Apple iOS, for the first time I see the peak of iPhone installed base having passed in Q2. It is yes, also only fractions of one percent but so you know, last four quarters were 19.5%, 19.7%, 19.7% and now 19.6%. As we already know from Apple own guidance for the Christmas season, there is no dramatic growth in unit sales compared to last year Christmas, but the industry has grown a lot, it means iPhone iOS installed base will now repeat the same market share decline we have seen in the annual sales market share for a couple of years now. This will be far more muted, of course, as the installed base reflects all devices in use, and Apple iPhones have a long life of use in the after-market, as hand-me-down phones and in the second-hand used phone market (also globally, used iPhones are big sellers in Emerging World second hand phone markets). For Apple fans, don't despair, the installed base will still be around 19.5% by year-end ad the rounding-off will probably still show 20% to the end of this year in the installed base. But just please note, the peak has passed and now will be in the same downward path as iPhone annual sales market share of new smartphones has been. That is inevitable math... But now we know it has happened and the gradual decline has started. By about 2020, the iPhone installed base will be down to about 12% when iPhone new unit sales are down to about 10% of all smartphones sold worldwide. (My readers care about these things..)

We also get a convenient rule of thumb now. iOS to Android installed base is now 1:4. So for every iPhone in use worldwide there now are 4 Android smartphones (this excluded tablets obviously where also Android leads but I don't study such tiny markets as tablets or PCs, haha, mobile consumes all of my time in tech). Oh, if you want the ratio to include Windows? Then its 1:10:40 for every 1 Windows smartphone there are ten iPhones and for every 1 Windows smartphone owner you might find, there are 40 Android owners. Nobody makes Windows apps anymore...

Other than Android won these wars, iOS is a healthy niche market for wealthy customers, Tizen is hoping to pass Blackberry and Blackberry is switching to Android, there is one more obvious refrain for us all .Say it with me, readers: Windows Phone continues to remain dead (and Lumia unit shut-down watch is now to 18 months of life left) That the Q3 report. Tune in, in three months, for the Q4 and full year 2015 report.

Comments

[Winter]

Tomi, you write:
"(the average now works out to 354.2 million units worldwide)"

But in the table you write that to total is 354.7 million units. Which one is correct?

[E.Casais]

Tomi,

From the statistics you publish, it appears clearly that the mobile phone market has now fully matured: two big incumbents, two platforms, disappearance of small OS and players, only one production region (Eastern Asia), uniform design factors (touchscreen slabs -- everything else has practically disappeared), growth slowdown. To be uncharitable: boring. I do not see any major innovation in the terminals themselves taking place today.

What about usages? Your reports about services and applications in marketing, emerging markets, Africa indicated many intriguing usages of mobile phones. It would be nice to revisit that aspect of the market -- it seems to me that genuine innovation is coming from there, not so much from handset manufacturers.

[Gonzo]

@Tomi

We also get a convenient rule of thumb now. iOS to Android installed base is now 1:4. So for every iPhone ...
iOS is a healthy niche market for wealthy customers...

Wow, amazing ! which company in the world can consider itself a healthy "niche" worldwide and command such 1:4 ratio

Nike would be delighted to sell 1 out of every 4 pair os shoes sold worldwide

Now more respect for Apple ..

iPhone vs all manufactures in the world - 1:4

Unthinkable 7 years ago.

[Per "wertigon" Ekström]

@Gonzo:

Yes, at the height of Apples market share it was 1:4.

Going forward that advantage will shrink to maybe 1:6 or 1:7.

Apple will continue to be profitable of course, for atleast a decade more.

But it's influence on smartphones from here on out will be minimal.

[abdul muis]

@Tomi

A typo,

5 (5) . . . Lenovo . . . . . . 18.8 M . . . 5.3% . . . . . . . ( 4.8% ) . . . . . . Android (Tizen)
4 (4) . . . Xiaomi . . . . . . .18.5 M . . . 5.2% . . . . . . . ( 5.6% ) . . . . . . Android

should be...
4 (5) Lenovo
5 (4) Xiaomi

[abdul muis]

@E.Casais

" (touchscreen slabs -- everything else has practically disappeared), growth slowdown. To be uncharitable: boring. I do not see any major innovation in the terminals themselves taking place today."

Fear not my friend, Project Valey device will make phone become interesting again.
https://www.youtube.com/watch?v=TmQ7SlvYMzk

[abdul muis]

@Gonzo

"iPhone vs all manufactures in the world - 1:4"

Samsung vs all manufactures in the world - 1:3

[abdul muis]

@Tomi

"also globally, used iPhones are big sellers in Emerging World second hand phone markets"

Tomi, I lived in Emerging World, and witness the "cheap" (used/refurbished, but MARKETED as BRAND NEW by 3rd party well known store) US$100 iPhone as we speak now. I'm curious, how big is this market?? Because every year Apple launch a new iOS and the old iPhone will become slower and slower with newer OS. Not to mention that apps become more resource demanding by each release.

[Tester]

@abdul muis:

Right now my impression as a developer is as follows:

iPhone 4: too slow, already irrelevant for new apps and droppoed off Apple's support list for OS upgrades
iPhone 4s: still ok for apps that are not very resource demanding, will most likely be dead next year
iPhone 5: on the way to where 4s is right now
anything better is still fully supported.

So my estimate would be 4-5 years of max. lifetime for a phone before it falls off the support roster. Of course we now enter the phase where these cycles get extended by slower developing hardware advanvements.
Of course, things can be accelerated when Apple prematurely decides to leave a device out for the latest OS upgrade and that upgrade becomes an important one. Right now, for example, the prime development target is mostly iOS 7 with 8 and 9 features made optional, if possible, so any phone that can't install this version is dead, for all intents and purposes.

[Paul]

@Gonzo

> which company in the world can consider itself
> a healthy "niche" worldwide and command such 1:4 ratio
> Nike would be delighted to sell 1 out of every 4 pair of
> shoes sold worldwide!

Sorry, this is the wrong way of seeing it!

Tommi stated very clearly that "iOS to Android installed base is now 1:4. So for every iPhone in use worldwide there now are 4 Android smartphones".

In order to compare apples to apples in the shoes comparison it should be something like:
- "shoes made out of plastic" versus "shoes made out of leather", or
- Nike versus its next largest/biggest competitor.

Also, shoes market is much more fragmented and has several orders of magnitudes more players that phone market and your comparison does not take this into account at all.

[Tester]

@Paul:

Well, that's what you get from a true iFool.
There's math and there's iMath. Sane people use math and get reasonable results.
iFools use iMath which always contains a heavy Apple bias and often even some magic to present Apple in the best light possible. It also tends to ignore any factors that might not be in Apple's favor completely.

[Winter]

In this discussion, I thought it would be nice to know what fraction of the phones is decommissioned every quarter.

So I took the total installed base for every quarter and subtracted the installed base for the previous quarter. That is the growth of the number of smartphones in use. This I subtracted from the number of Smartphones sold each quarter to get the number of Smartphones that were decommissioned every quarter. This I divided by the total installed base.

When I run this on Tomi's numbers from Q1 of 2012 (Q4 2011 is the first with installed base numbers) I get interesting numbers. Up to Q1 of 2013, around 5% of smartphones were decommissioned each quarter. For the remainder of 2013, the numbers vary widely, from 9-20%. From Q1 2014 on, a very consistent 11% of smartphones are decommissioned each quarter.

11% per quarter means a turnover of roughly 45% per year. It also means that, globally, the average life time of a Smartphone is 9 quarters (27 months). Which is quite long.

[chithanh]

@Winter
For an accurate picture on that you would have to account for the difference between sell-in (to retailers) vs. sell-through (to customers) too. That can sometimes be considerable, remember Tomi writing about the Lumias which were never activated.

Anyway, new smartphones have a replacement cycle of around 18 months or so I've read. An additional 9 months of second-hand use appears reasonable.

[abdul muis]

@Winter

"When I run this on Tomi's numbers from Q1 of 2012 (Q4 2011 is the first with installed base numbers) I get interesting numbers. Up to Q1 of 2013, around 5% of smartphones were decommissioned each quarter. For the remainder of 2013, the numbers vary widely, from 9-20%. From Q1 2014 on, a very consistent 11% of smartphones are decommissioned each quarter."

That's because your calculation is not the same as Tomi's.

It should be something like this
in first year, x% broke down
in second year, x% broke down, y% retired (not used anymore), z% have second life (hand-me-down / sold)
in third year, x% broke down, y% retired, z% have second life, n% second life broke down, m% second life retired
.... and so on, and so on....

PS: I don't know the exact percentage that Tomi's secret formula...

[abdul muis]

@chithanh

"Anyway, new smartphones have a replacement cycle of around 18 months or so I've read. An additional 9 months of second-hand use appears reasonable."

In the past, PC replacement cycle is also short, but lately has change to 7 years (if I'm not wrong).

Any android phone that have 2GB ram+, even with slower CPU (such as SD S4Pro) from 2012 (such as Nexus 4), still as smooth as 2015 mid-offering with SnapDragon 615 series. This 2012 SD S4Pro phone even better (better GPU) at gamming than SD615 from 2015. I think the how-long-can-a-phone-survive of flagship phone is mostly down to how-long-before-I-crack-the-screen.

[Winter]

@Chithanh&All
Sorry, the procedure I used to estimate the time of life of Smartphone handsets was wrong. Things are more complex for exponential growths.

A better procedure is to sum the quarterly sales to S(Q) (Total Sales until quarter Q, inclusive). The "average" time of life of Smartphones at time Q is the number of quarters q such that the current installed base B = S(Q) - S(Q-q)

For instance, at the end of 3Q2015 the total number of Smartphones sold since 4Q2011 was 4139.6M (using Tomi's estimates). The installed base at that time was 2367M. 8 quarters earlier, at the end of 3Q2013, the total number of handsets sold was 1541M. The number of handsets sold since 3Q2013 is 2598M. The installed base in 3Q2015 was approximately the size of all smartphone handsets sold since 3Q2013. So, on average, phones are used for 8 quarters, the difference between 3Q2015 and 3Q2013.

If you do this for all data you get a consistent life time of 8 quarters all the way back to 4Q2013, where the series stop as data only go back to 4Q2011.

So, the average life time of a smartphones is around two years, at least for phones sold since 4Q2011. That would be 6 months more than the average time to sell-through. All according to the estimates given by Tomi.

[Winter]

@abdul
"PS: I don't know the exact percentage that Tomi's secret formula..."

My calculations assume that the installed base is a reasonable estimate of the real number of smartphones in use at a certain time. If the installed base statistics were estimated using the reverse procedure, assuming a life-time of 2 years and then calculate the summed sales for the last 24 months, then my results are trivial and uninteresting.

Which might very well be the case.

[Catriona]

It's time to start separating Google Play Android from Open-Sourced Android, particularly since the focus is shifting to China, where it makes a real difference. Also since Google just announced that they are merging Android and Chrome, expect to see them reassert control over Google Play Android in a big way.

[Tester]

And the point being - aside from making Apple look better? Stop FUDing around.

[abdul muis]

@Catriona

Google did not merge Android and Chrome. It was a mistake by some WSJ writer. Perhaps he doesn't really know Google android & chromeOS product, and when google say ChromeOS will be able to run Android apps, he thinks it will be merged.

http://www.androidauthority.com/google-chrome-os-committed-652794/