The Google Trends chart for Flappy Bird is a wonder to behold.
If you were a PC OEM from, say, 1990 to 2010, you operated in a very clear ecosystem. You outsourced much of the innovation to Microsoft and Intel. You knew exactly what WinTel were doing, both because their roadmaps for the next few years were actually public and because Microsoft and Intel had very clear and widely understood strategies. Moreover, the core, fundamental strategies of OEMs, Intel and Microsoft were pretty much aligned. Everyone wanted more PCs to be bought, and preferably a good number that were high-end and high-margin. Intel, Microsoft and CloneCo all lived for the same things. CloneCo didn't necessarily make great margins (and eventually got killed by Dell, perhaps), but it knew what the game was.
The Android ecosystem today is superficially similar to the PC ecosystem, but I'd suggest that the clarity and alignment of interests of the PC ecosystem isn't present in anything like the same way. As an Android OEM you have very little idea what Android will be in 3 years - partly because Google itself may not have a fully-formed idea. There certainly aren't public roadmaps stretching out years in advance.
It's also questionable how much alignment of interest there is. Google certainly wants Webkit everywhere, and arguable Android everywhere (or, more precisely, Google Plus everywhere). But that doesn't translate to a burning hunger for an aggressive phone replacement rate at high prices. Indeed, Google Play Services reduces Google's interest in device replacement as a way to drive service penetration. As an Android OEM, your ecosystem creator doesn't benefit directly from the health of your industry. A healthy PC market was Microsoft's driving objective - 'A computer on every desk and in every home'. That's not quite the case for Google and the sales of Android smartphones - they're reach, and a means to an end, but not the reason why Google exists.
Next, it's not clear what a sustainable position for an Android OEM looks like. All the brands except Samsung are sub-scale and failing, and while Samsung looks dominant it is clearly feeling paranoid: the growth of the Chinese Android OEMs outside China is a huge question. Lenovo has made its first move by buying Motorola but the real story is whether 2, 10 or 100 others follow it, and if so how.
Finally, Google's control of Android itself is a question. Amazon forked it, but with limited broader effects. Almost all Android in China lacks Google services but then Google is largely absent from China anyway. The ways that forks of Android might become relevant outside China (and Google's tools for preventing this) are complex and a topic for another post, but we can't rule this out. Indeed, a lot of the most interesting ecosystem innovation is being done on top of Android rather than as a would-be competitor to it.
On one level, then, the smartphone platforms wars are over - iOS and Android both won. But actually, nothing is finished - we just move onto new questions.
Suppose that in 5 years or so I send you a Yelp review of a restaurant, from my phone to yours. What will that mean?
- First, I might well use something like Airdrop, or touch my phone to yours to pass it across, or tell Now to give it to you, or indeed Now might decide to give it to you without my even explicitly asking. Or the review might be invoked by a Bluetooth LE beacon as you hold your phone next to the menu on display by the door
- But for the sake of simplicity, suppose I send it to you using an internet messaging app - either one built into the OS or a third-party one - Facebook, Whatsapp or more probably one that doesn't exist yet but by then has 15 engineers and 1bn MAUs.
- It seems pretty unlikely that you'll see a dumb URL string on your screen. Rather, you'll get something rich and interactive, within the message.
- And you'll be able to go into that experience and tap the number to call the restaurant, or make a live booking, or swipe through photos.
- And if you tap 'book', it'll pass them a $10 booking fee in bitcoin, authorised with a fingerprint swipe.
- Now suppose you decide to save this item, as an icon on your home screen, or some other yet-to-be created place.
Now, what were you using? An app? a widget? Native code? What programming language? Did you install an app or surf the web? I'd suggest that none of those questions would really mean anything, at least not as we think of them right now. The programming language matters much less than the user flow. And some of this example sounds 'webby', but Google is the first to advance interaction models that are not remotely webby (such as Now).
This is a pretty simple illustration (an expansion of the super-hot card metaphor) of a broader point I've made before: on the desktop internet, the web was by far the dominant model and that didn't actually change very much for well over a decade (before that, the interfaces of Windows and Mac were also very stable for a long time). But on mobile, not only are other models just as important as the web, but they're not remotely stable, settled or mature. The platform war may be over but that doesn't mean things are settling down.
So I have very little idea what precisely I would mean if, in 5 years, I were to say 'I installed an app on my smartphone'. Further, I'm pretty sure that if it's an Apple smartphone it will run an iteration of iOS but I'm rather less sure what Google will have done with Android and Chrome by then. And of course I might be running a fork of Android from Amazon or, perhaps, Microsoft.
This is the key reason why the new social messaging apps are so interesting - not because they have users and inventory now, but because they can be vectors for some of this sort of behaviour - a third acquisition, discovery and distribution channel besides Facebook and Google.
This may also have implications for any discussion about what it means for Apple that its ecosystem will have a minority of mobile users. We need to think about what it means to call a ecosystem that might have 800m-900m live devices 'minority', but we also need to think about what 'ecosystem' might mean. What, if any, 'winner takes all' dynamics operate in this environment? One reason the Mac didn't die was because the web changed what it mean to be a computer ecosystem: the mobile ecosystem has lots of changes to come too.
This isn't really a post about the tech industry so much as one about maths. Or, perhaps, three charts.
The first shows Apple's share of global smartphone unit sales. Flat, stagnant and, to be honest, slipping. Right?
Hmm. Now look at the same numbers, this time shown as a percentage of global mobile phone unit sales.
Hmm. Now, actual unit sales.
Now, let's add Android to this.
There was a point in time where talking about share of smartphone sales was a meaningful and important metric. That time has passed. It's rather like talking about Toyota's share of sales of Japanese cars in the USA: it tells you something, and was very useful in the past, but not any more.
There are lots of issues and questions about Apple's future, and lots of different things going on in those charts, including a clear decline in the growth of sales. But 'smartphone share' is not a helpful way to think about those questions.
I'm pondering Apple's results, sitting in a cafe in-between meetings. It's been apparent for years that Apple was camped out at the high end of both tablets and phones, and that Android would take almost all of the rest. But it's worth working out on the back of a (rather large) envelope exactly how that would play out, assuming that Apple's pricing strategy doesn't change and of course that nothing else changes (which of course are rather unsafe assumptions). Hence:
- Right now, on the basis of a 24m replacement cycle, there are perhaps 290m iPhones in use on earth. Depending on the second hand market, this might be larger.
- Apple has sold 195m iPads - perhaps 180m are still in use.
- There are also a fairly small number of iPod Touches in use - perhaps 20m
That adds up to a rough estimate of 490m live devices. For comparison, around 900m Google Android phones were sold in the last 24m, and probably another 110-120m Google Android tablets ('Google Android' = 'not China').
Where might this go? Apple now has about 10% of global mobile phone sales, rising steadily. It's important to note that Apple is not losing share of phone sales to Android - it's just not taking as much share as Android. There are between 3.5bn and 4bn people with mobile phones on earth (there are far more connections but many people with multiple connections). This number is also rising slowly, but all the growth is from the very low end.
Over the next few years the great majority of that 3.5-4bn will convert to smart (and indeed the more important variable is affordable data plan penetration rather than smart penetration). If Apple continues the current strategy and share growth, it will end up with (say) 15% unit share. 15% of 3.5bn is 525m. (I told you this was a BOTE calculation.)
The great majority of the rest will go to Android (though quite what 'Android' will mean is an open question). That means perhaps 2.5-3bn Android phones in use. There might be some Windows Phone as well (assuming it doesn't become an Android fork) but we can ignore it for these purposes.
No-one has much idea what the total addressable market for tablets really is, let alone premium ones such as the iPad, and the recent sales trajectory is somewhat lumpy. If we assume a four year lifespan for iPads as the tech stabilises and look at the recent run-rate, that suggests a stable base of, say, 300-350m. This gives us a base of perhaps 850m iOS devices, with a lot of ownership overlap.
There will also be an ungodly number of Android tablets, of course, but we know neither quite how many nor what they'll be being used for (right now, mostly TV, it seems).
What does this mean? What does it mean for Apple to have a platform with a minority of users, indefinitely, in mobile? An minority ecosystem with only 850m devices? Or even 490m?
Certainly, this isn't 'Windows versus Mac all over again'. There are now 490m iOS devices in use, but PCs only hit that number in around 2000, long after Apple lost the last ecosystem battle. Apple sold 51m iPhones last quarter - total PC sales in 1995 were 59.5m. That is, the iOS ecosystem now is much bigger than the winning ecosystem back then.
Even beyond that, all the other dynamics are different - the smartphone market is not driven by corporate buyers who demand commodity product based on bullet points and don't care about design or user interfaces, for example. The relative market share of an ecosystem is relevant, but it's not the only thing that's relevant. We need also to think about value share, engagement share, and all the other dynamics that drive the viability of an ecosystem. The assertion that an ecosystem with close to 500m users now and over 800m in a few years will not be viable because there's another that's bigger seems pretty simplistic. It can't be taken for granted that any 'winner takes all' dynamics will work like that.
More importantly, though, these questions will probably have become irrelevant by the time we find out the answer. That happens quite often in tech. To me, the platform wars are now much less interesting than what happens on top of the platform. On the desktop internet, we had close to 15 years of stability with almost all online activity going through the web browser, but on mobile it is far more complex already and also far less stable. Nothing is settled on mobile. I have no idea what it will mean in 5 years time to say that I 'installed' an 'app' on an 'Android' 'smartphone'. All of those terms could change completely, and with it what it means to say 'ecosystem' or 'market share'.
Note: for supporting charts, see this post.
"However vast any person’s basic reading may be, there still remain an enormous number of fundamental works that he has not read." - Italo Calvino
It hasn't been possible to have read 'everything' for a couple of hundred years (since Erasmus, perhaps). It hasn't been possible to understand all of engineering or technology for perhaps a hundred or a hundred and fifty years. And today, it isn't really possible to understand all even of the internet or mobile - not all at once. There are simply too many things going on for any one person even to know the key basics in every relevant field, never mind become an expert or have some insight. I'm pretty sure it's not possible to have a really deep understanding of all of, say:
- The mobile semiconductor industry and the competitive positioning of Mediatek, Rockchip and Qualcomm
- The mobile handset market and the interaction of Apple, Samsung and the Chinese OEMs
- The online music business
- The latest developments in ad-tech
- SEO, social sharing and user acquisition across desktop web, mobile web and apps
- Power and processing budgets for very cheap/low-power hardware (Android, drones, thermostats, wearables, HDMI dongles...)
No-one can be an expert on all of those, yet all of them are relevant to someone trying to understand mobile ecosystems. And as mobile and software eat the world, more and more previously separate things overlap and become relevant to each other. For 20 years we've been talking about the 'convergence' of the three industries of 'TMT', Tech, Media and Telecoms, but now it's finally happening and they're not so much converging as colliding, like galaxies. (And a few new galaxies are smashing into the mix as well, like retail). That means there are even more things to keep track of and understand, because the opportunity set for new and disruptive businesses is expanding massively.
Hence, consider a few more things someone like me might now need to understand:
- The US pay TV industry and the evolution of affiliate fees
- Regional film rights windowing
- The outlook for operator handset subsidies
- The evolution of mobile data costs for low-income users in emerging markets
- The structure and economics of the global payments industry. Oh, and monetary economic theory
Part of the fun is that people in each of TM&T tend to think that they do understand the other two (and that the other two are run by idiots). But they see each other in terms of their own preoccupations, rather than looking at the real drivers in a quite different industry:
- Media companies look at tech and telecoms and say "well, they're just a channel for our stuff"
- Telcos look at tech and media and say "they're using our networks so we should control everything"
- Tech companies look at media and telecoms and say "a little disruptive technology would change everything"
So, rather like the blind men examining an elephant, who each feel one part and conclude the elephant is a snake or a tree, when TMT companies look at another industry they tend to see the bit that matters to them and assume that that's the important part. Tech people have this problem particularly badly: a repeated failing of tech companies is to look at media and telecoms, see some tech, and think it's the key point of leverage in those markets. Mobile networks and TV do look like tech should be a crucial lever, but that isn't necessarily so.
Two useful examples of this sort of cross-industry misunderstanding are Joost and mobile WiMax. These proposed radical technology that would disrupt the TV and mobile industries respectively, but in both cases that technology was actually applied to a part of the value chain that offered very limited leverage to disrupt anything, and both were total failures:
- Joost had a P2P platform that gave 'free distribution for TV'. But distribution is actually a tiny part of the cost of running a major TV channel - all the money goes to content. It took Netflix to work out how internet distribution could change things - as an enabler for a different operating model, with a new UX combined with data as a tool to rework content acquisition (and meanwhile YouTube solved the true long tail story)
- Mobile WiMax was supposedly 10-20-30% cheaper than then-current cellular network technology for equivalent capacity. But the equipment itself is typically only a quarter of the cost of an urban base station, and the number of urban base stations you need is a function of in-building coverage as well as capacity, and WiMax was actually only available for deployment on high-frequency spectrum which is bad for in-building coverage. And you were deploying a weird non-standard tech with only one vendor and no handsets. So when you ran the maths, the tech that was "30% cheaper!" could actually be much more expensive to deploy. (Yes, this is a huge simplification)
Naturally, one thing Joost and WiMax had in common was that TV and mobile people fell off their chairs laughing when they heard the pitch. And of course that showed that they didn't 'get it' - but actually, they did. They got the tech, but they also knew the broader industry context that meant the tech didn't matter.
Both of these are of course inversions of the way that tech often does disrupt industries - by affecting parts of the industry that no-one paid attention to but which were actually key leverage points. Not many magazine people thought of themselves as being in the trucking and light-manufacturing business, for example, but they were, and that was why the internet had such an impact on them. But the opposite can also be true - there are industries where tech doesn't look important but is actually crucial, but there are also industries where tech looks crucial but doesn't actually matter very much at all, outside narrow constraints.
The problem is, this sort of ignorance and misunderstanding is often how we get true disruption - people are so ignorant that they don't know something can't be done and won't work, so they go and do it, and it works. Dropbox and Paypal are particularly good examples of this, while Bessemer's 'anti-portfolio' is a fun look at the sensible reasons why some amazing companies would never work. The challenge of venture investing is that the model depends on investing in things that are laughable, because those are the only things that can make billions of dollars from zero in a few years. So you kind of want people to laugh at you and think you don’t understand the sector. You just have to be sure that you understand why they’re laughing.
If the number of smartphones on earth has not yet passed the number of PCs, it will do so any day now.
I've been looking at technology and mobile for 15 years, ever since I was a junior analyst working on tech IPOs in 1999. What's always fascinated me has been that point at which an idea or a technology becomes something tangible, and starts actually to work, and turns into something real. I've spent most of my time working out how, when and why great ideas and great technology can 'put a dent in the universe', and what that might mean for everyone else.
Late last year I spent some time with the Andreessen Horowitz ("a16z") team, and it's pretty clear that this is one of the best places there is both to see those dents being made but also, given the a16z model, to contribute a little to that process.
So I'm happy to announce that in February I'll be joining a16z in Menlo Park.
I'll continue to analyse the industry in public, here and elsewhere online, but in addition, I'll now be working with the rest of a16z to find, understand and support great ideas and great companies.
430m active users and 18bn messages sent per day, which is pretty close to global SMS volumes (20bn or thereabouts, and maybe lower). All with just 25 engineers.
"No ads, no games, no gimmicks". Interesting that by far the biggest mobile social app is the least complex, and of course the most focused. No canvas, no platform, no ecosystem, barely any API - and massive growth.