Scale and polarisation in mobile

There are lots of ways to look at the global handset industry, but the polarisation evident in this chart is pretty compelling. Not shown - the hundreds of 'other' manufacturers, mostly in China.

The CFO of Qualcomm recently described the industry as a barbell - Apple and Samsung at one end, then the smaller and mostly sub-scale players (though some, such as Sony, are showing signs of increased health, albeit from a low base), and then at the other end, invisible, the Chinese. 

As an aside, this also illustrates the way that Apple has become so cyclical that it's really only the December quarter that gives a good directional steer.  

Lumia and BB10

Interesting to compare the replacement of the legacy platforms with the 'future saviour' platforms at RIM and Nokia. Neither is going terribly well.

Screen Shot 2013-06-28 at 15.50.16.png
Screen Shot 2013-06-28 at 15.50.49.png

Of course, the underlying problem is that though both platforms are perfectly OK (though with their flaws), they're radically sub-scale. iOS now has about 400m active devices (iPhone, iPad and iPod Touch, though obviously with overlapping users) and Android over 800m, plus China. Lumia and BB10 combined have sold a little over 23m units in the last 18 months.  As a developer, why would you target these?

It's also interesting to ponder what would have happened if both companies had swallowed their pride and gone with Android, or even forked Android. I don't actually think Blackberry would be in a better position, but Nokia might have been. 

BBM and the rest

Blackberry gave a few operating metrics for BBM today as part of their developer event, at which they stated BBM will go onto iOS 6 and Android ICS+. There are: 

  • 60m active users
  • 51m using it for 90 minutes a day (definition unclear)
  • 10bn messages a day (sent and received)

Blackberry claims that this is 'nearly twice' the message/user/day volume of any other platform, but this isn't quite true: on a sent-and-recieved basis BBM does 167 but Whatsapp does around 100. On a sent basis (assuming a 50/50 split), Kakao does 56 and BBM does 83. 

The two charts below give a somewhat impressionistic sense of the landscape. These are just the biggest players, and the definitions are variable. WeChat has stated 190m active users as opposed to 300m registered: Whatsapp says 'over 200m' actives. In total, there are several billion accounts on all of the services out there (with massive duplication, of course): I've lost count of how many such services there are in total. 

Screen Shot 2013-05-14 at 3.58.32 PM.png
Screen Shot 2013-05-14 at 5.02.39 PM.png

BBM is now one of many such services, and not the biggest any more. Cross-platform may not be enough. 

(Note - I updated the second chart after Blackberry said it is quoting 10bn messages as both sent and received Link)

Device prices, self-selection and a Catch 22

One of the many challenges faced by BB10 is that the launch hardware is priced to compete head-on with the iPhone 5 and high-end Android phones, yet the platform lacks the range and depth of apps that those devices can offer. Nokia's Lumia launch faced exactly the same problem 18m ago.

The obvious argument is that these new platforms would do rather better at a lower price point, competing against mid-range Android and the residual featurephone base, where the target customer cares less about apps and the user experience is more differentiated (i.e. the mid-range Android experience is less good than the high-end one).

However, if you only target users that don't care about apps, then your install base will never be an attractive market for developers, and you never will get apps.

This seems to be what is happening with low-end Android tablets. Google (arguably) tried to short-circuit the 'no users therefore no apps therefore no users' issue by selling the Nexus 7 at cost, carrying an effective subsidy in its working capital and making the device cheap enough that people would overlook the lack of apps. It remains to be seen just how many actually sold as a result: the launch quarter sales were a little over 2m, though with limited distribution, and we don't yet know how many were sold over Christmas (though some people have had a guess). Meanwhile the Nexus 7 is not even the cheapest Android tablet on the market - devices at $100 or even lower are to be had, though with very poor performance. (The Kindle Fire is available as well, but that's a rather different proposition, with more going for it than price.)

The challenge with all of these tablets, though, lies in the choice presented to consumers. In effect, they are asked: "would you rather buy this Apple tablet with lots of apps, or save $100 or so and get this black plastic thing with far fewer apps?"

That's a perfectly legitimate question to ask, and Christmas was one big A-B test as to what tablet proposition people actually want. However, what does it tell you if someone says 'I want to save $100 and get the cheap-looking one with no apps'? Are they a good target for any publisher or developer? This is at the root of the staggeringly low engagement on Android tablets that all publishers report - under 5% of what they see on the iPad: self-selection by the users. People who buy cheap tablets are effectively declaring that they value the saving over the apps.

In other words, you can sell to a high-end user and hope that they'll forgive the lack of apps. That's hard, and you won't sell many. Or, you can make a cheap device (setting aside the technical challenges RIM and Nokia faced in actually doing that), which is rather easier to sell with no apps - but then the user base you do get is even less likely to buy apps. Catch 22.

RIM: a numbers game

RIM, to me, has numbers working against it. 

Start with the product: on first impressions, BB10 promises to be good enough to keep corporate users, but it is not really competitive (in my opinion) with the iPhone or Android for consumers both on features and price, and it is a non-starter (and this is not a matter of opinion) in developing markets or for the 'teenaged girl' market purely on price. 

So, if we assume for the sake of argument that it can now retain most renewing corporate (and, in general, high-end) users and also convert some back, how many people is that?

35% of revenue in the last quarter came from Canada, the USA, and the UK: the rest was in emerging markets. That probably works out at less than 35% of handset sales, since it will mainly be at higher than the company's average ASP, though a significant portion of even that 35% was also to low-end users - in the UK, teenaged girls and C2DE consumers. 

Screen Shot 2013-01-30 at 7.34.26 PM.png

Hence, if we look back over the last 24 months, it looks as though RIM sold maybe 20-25m phones in these markets (out of 85m total, or 25-30%), weighted towards the beginning of that period. Maybe 5-6m were in the quarter 2 years ago - i.e. the people now coming up for renewal. 

What share was high-end? Say 2/3, for the sale of argument. That gives 4m potential BB10 sales from renewals in the next quarter - PLUS of course whatever they can tempt back from people who deserted BB for iPhone two years ago - presuming they haven't just upgraded to an iPhone 5. Will people come back? We'll see. The ASP will be an obvious give-away. 

Where does all this get us? These are VERY rough numbers, and all could be wrong by 50% or so. But putting the most optimistic spin on things, RIM's 'loyal corporate base' isn't going to deliver sales of much more than, say, 20m BB10 units in 2013 (Nokia sold 13.5m Lumias in the last 12m)

Now, suppose that's totally wrong. Say RIM sells 30m units in 2013. 

That would add up to an install base of 30m (obviously), compared to an iPhone base today of over 200m and an Android base of well over 500m. 

If you're a developer, where do you put your efforts? Assume for the moment that these lovely corporate users are actually allowed to install apps (many currently aren't). Do you write a native app for BB10? Do you update it every few weeks with new stuff, the way you do for iOS or Android? Do you take RIM's money and put it out to see what happens? Do you stick your Android app in store to see how it performs (in both senses of the word)? Or do you just wait and see? 

For consideration: how much does it matter to RIM if no-one codes native apps? How far can it get slip streaming Android? And, of course, how much does RIM really need? Is this just the last pump before an acquisition? RIM is certainly in play (as are Nokia, HTC, LG Mobile, Sony Mobile and arguably in some sense Motorola) and this certainly doesn't worsen the story to a buyer. 

RIM in the USA

Yesterday I posted a Google Trends chart showing relative search volume for "BBM for iPhone" and "BBM for Android", the hypothesis being that RIM user who switched platforms would look for BBM on their new phone (though it doesn't exist).

Narrowing this into the USA is equally interesting: it looks like rather more Blackberry users went to iPhone than Android, and also that they've now mostly already left (borne out by most other market stats). "BBM for iPhone" is red, "BBM for Android" is blue. 

Screen Shot 2013-01-18 at 11.14.42 AM.png

This speaks to the question of whether the RIM strategy of offering Android apps on BB10 (sort of) will work: the overlap between iPhone and Android app availability is far from perfect. 

What platform do Blackberry users go to?

Google Trends data is fascinating. It's very partial, and in particular doesn't capture app store searches, but it can still give a very good indication of where things are heading. 

This chart shows search volumes for "BBM for iPhone" (red) and "BBM for Android" (blue). The hypothesis is that when a BBM user dumps their Blackberry and gets a new phone on a new platform, they Google to see how to get BBM on that new platform (they can't). 

Screen Shot 2013-01-17 at 11.52.16 AM.png

Hence, this chart illustrates the way that RIM lost high-end customers first, in developed markets, who went to the iPhone. Then, it started losing lower-end customers (in emerging markets), who went to Android. The high end departure looks like it's mostly over (though this obviously doesn't tell us how many loyalists have stayed) - the departure to Android  is still going on. 

Google Trends also gives regional interest data, which supports this interpretation: "BBM for Android" (blue) clusters in Indonesia (not a big surprise), "BBM for iPhone" (red) in Canada and the UK, and more recently South Africa. 

Screen Shot 2013-01-17 at 11.46.47 AM.png
Screen Shot 2013-01-17 at 11.46.38 AM.png

US iPhone revenue is now over 8 times US RIM revenue

A painful observation, this.


RIM reports US revenue, which was $700m in the May 2012 quarter, down from $2.1bn in May 2010, a 66% decline. 

Apple doesn’t split out iPhone revenue by region, but the US operators that sell iPhones report their quarterly unit sales. If you apply Apple’s overall ASP (around $650) for the iPhone to those numbers (a rough but reasonable assumption), you get about $1.8bn in the March 2012 quarter, and $5.9bn in the March 2012 quarter, a 3.3x increase. 

In other words, in the last two years: 

  • US iPhone revenue has tripled
  • US RIM revenue has fallen by two thirds
  • US iPhone revenue is now over 8 times US RIM revenue. 

Despite both starting from roughly the same point. 

(Caveats: this doesn’t include any iPhone revenue share from operators,nor other factors like Google payments for Apple’s use of Google Search on the iPhone. Equally, a third of RIM revenue is service fees rather than direct hardware sales).

WhatsApp versus BBM


Another interesting Google Trends chart (and again, this isn’t really reliable data): WhatsApp is about to pass BBM in global search volume. 

Hardly surprising, given that Blackberry sales are collapsing and iPhone and Android are exploding. And WhatsApp has been camped out at the top of the ‘top downloads’ charts on both platforms for months. Great execution, and just proves you can succeed with a silly name…

How many Playbooks? 400k sold, 1m+ unsold?

A few datapoints for the over-excited out there: 

  • First, Facebook’s app for the Playbook has 170k monthly active users. So there must be at least that many Playbooks in consumer hands, and more probably something over double that. iPhone, Android and Blackberry handsets all have about 50% penetration, and the Playbook both has a decent browser and is aimed (partly) at corporate users, so one would expect lower Facebook install rates.  
  • Second, in RIM’s statement today the company said: RIM sold into the channels approximately 150,000 BlackBerry PlayBook tablets in the third quarter and sell-through to end customers, based on RIM’s internal data, was higher than this amount. (My emphasis). 

So, RIM’s systems tell it that end-customers bought something more than 150k Playbooks in the November quarter - the implication is that some of these are Playbooks that shipped but didn’t sell in the two previous quarters in which it was on sale. That includes the effect of some new and dramatic price-cuts, so one would expect lower sales in the August quarter, and perhaps a higher figure in the May (launch) quarter. 

Finger in the air? Real customers have probably bought 4-500 thousand so far - maybe 150K + 100k + 175k, or 3x the Facebook app install base. 

So how many Playbooks are sitting in warehouses? Well, in total RIM has shipped 850k Playbooks - 500k in the launch quarter, 200k in the next and 150k in the quarter just ended. It’s sold 4-500k so that leaves maybe 3-450k, right? Wrong. ‘Shipped’ is just the stuff RIM persuaded distributors to take. There may be more - much more - still in RIM (or Quanta) warehouses.  Afterall, the distributors wouldn’t keep taking stock they couldn’t sell. 

This is why the size of the write down is important. If the Playbook was on the books at a cost of $350 each and they were written down to zero then, simplistically, that write down would amount to 1.38m unsold Playbooks ($485m/$350). But they probably haven’t written them to zero - they still have some value. If the write down per device is smaller, then simple maths suggests there could be well over 2m million unsold Playbooks out there. 


If nothing else, this shows just how delusional RIM was about the appeal of this product. I looked at it at MWC/3GSM in Barcelona in February and knew it was a dog within about 30 seconds. I have the suspicion that RIMs CEOs still don’t quite get it. 

Also, what happens to the Android tablet market if RIM dumps 1-2m deeply discounted Playbooks onto the market? At $200 each?