Where Should Mobile Developers Focus?
I sent out this tweet yesterday:
I got a bunch of replies suggesting that Apple will remain the market share leader when measured in dollars and I really wasn't focused on that. I don't own Apple stock or Google stock. I'm not really focused on who makes more money in smartphones. But I care a lot about where our portfolio companies should be focusing their precious mobile development resources.
And this chart from comScore tells a very interesting story:
Apple is stuck at about 25% of the smartphone market. RIM is losing share and Google is gaining share. If we have two more quarters like this past quarter, Google will have 37% market share, RIM will be at 29%, and Apple will be at 26%.
Of course there is no certainty that the next two quarters will play out the same way the past quarter went. Many people replied that getting the iPhone on Verizon will be a boost to Apple's numbers. I suspect that will help. Apple could get into the mid 30s with the help of Verizon.
But the most interesting number in that table is the 6.5% increase in share by Android. I believe that is coming from three factors. First, I think many Blackberry users who still want a keyboard are moving to Android. Just look at this picture of the Droid Pro and you'll get my point.
I don't think you'll ever see an iPhone that looks like that. And Blackberry users want a phone that looks like that but also has a great browser and great mobile apps. That's a big factor in Android's surge and it is coming at the expense of RIM.
The second thing that is working in Android's favor is low cost. Only one in four mobile subscribers has a smartphone. But that ratio is changing fast. The number of smartphones rose 14% last quarter according to comScore. It's not that the other 75% all don't want smartphones. Many do want them. But they can't afford them. But when T-Mobile sells an Android powered smartphone for $30, that's a game changer. Android smartphones are cheap and getting cheaper. So I think a big part of Android's growth is coming from people who are getting their first smartphone.
The third thing that is working in Android's favor is the declining share of the "also rans" which include Microsoft and Palm. Microsoft's Windows 7 mobile may halt their slide, but I am not convinced. Palm is down to less than 4% market share. It's hard to see how they can remain viable in this market. People who have Windows and Palm powered phones could have bought iPhones. But they did not. Maybe they were deterred by AT&T and will move to Apple when the iPhone is offered on Verizon. But I think this market may also end up using an Android powered phone.
I am pretty convinced we are going to see the mobile OS market split between Apple and Google, with Apple having the better business in terms of revenues and profits and Android having the bigger market share.
I think RIM is going to struggle more and more every day. Moves like they are making against Kik, which provides cross platform BBM, are likely to come back to haunt them. They should be making it easier for their users to chat with iPhone and Android users, not harder. Open platforms win and closed platforms die. And RIM still does not get what being an open platform means.
So, when thinking about where to invest your precious mobile development resources, I'd say Android first and iPhone second. And think hard about HTML5. You may want to hedge your bets by having a kick ass HTML5 experience. I learned in the comments to this HTML5 post last week that there is an awesome open source library called Phone Gap that lets you port HTML5 apps to Android, iOS, Blackberry, Palm, and Symbian. Seems like developing in HTML5 and then porting to the mobile OS platforms is an interesting option as well.
One thing I am sure of is that developing solely for iOS, which is a very common thing I see out there, is not the right strategy unless you only want to serve 25% of the market.
Comments (Archived):
First off mobile developers should focus on every platform, just as video game makers make games across video games consoles.But I do think you’re right that Google will eventually dominate the market over Apple’s OS.Simple case and point. I have an iPhone and my little brother has a Google Smartphone. His hardware cost merely $100; he gets the same # of apps (e.g. Kik), and his cell bill is cheaper on a monthly basis.I think this may come down to simple supply and demand.Google’s phones are a replica of the iPhone and they are cheaper.
I tend to agree with you in the long run, but in the short run ie: 2-3 years – I think Apple will remain share leader. It’s easier to develop apps for the iPhone given the single hardware set and one OS. I think that’s an important factor and while it’s a bit of an achilles heal at times, the control Apple exercises over the combination of their hardware and software platforms does help on that axis.
Apple isn’t the share leader nowRIM is
I think Apple is the share leader in # of apps, and $$ from apps. So, for developers this remains the best platform. It’s easier to develop for this platform right now as the variation on the Android side causes development/test and deployment challenges.As others have pointed out, while the iOS share is 25% based on handsets, it’s much larger in terms of applications and users willing to pay for applications. (it’s over 30% of the mobile games market, for example, while Android is 10% according to at least one source.)Regardless, I think you are right: RIM is in a bad place, and it’s Android and iOS for startups. If the company has to pick — I think it may depend on the type of app and model, but you need to plan to be on both as soon as possible, if not at initial launch. But, work hard to launch on both simultaneously. That”s what the curent data shows – developers are frustrated with Android fragmentation — but almost 50% are already planning to launch on both platforms.
as i’ve said many times now in this thread, paid apps are a market aberration and won’t lastmobile will trend toward web economics as it reaches scaleand there aren’t many paid websites
I very much agree that paid apps are an interim economic ecosystem that will fracture. I still think there will be room for paid apps. But as you point out – I agree that web economic models will infiltrate.
This comment beats the blog post itself. For me.
Can’t agree with this one.What’s the monetization model for free apps? Subscription or credit based services/games might work for some apps, but what are the rest going to rely on? Ads? Don’t see that working for 95% of apps out there.
what is the monetization model for web apps?mobile apps and web apps are more or less under the same set of economic constraints
and most good web apps are premium, just as most good iOS apps are.Even so, iAd revenue seems to be picking up. Android offers nothing even close.
Whether mobile or web apps, premium content will demand payment of some sort (consumer subscription, advertising, etc). To me, mobile vs. web app debate is meaningless. They are just means of access to content. I suppose providing simplicity and convenience (through mobile or web app UI) can demand a certain premium, but the real value is in content.
You’ve repeated that paid apps won’t last several times — unfortunately repeating it doesn’t make it true. Show me a web site that acts as an universal remote control for my home theater system, and maybe I’ll believe you. Yet paying say $50 for an iPhone based universal remote is still a savings of hundreds of dollars from dedicated/single purpose, nightmare UI universal remotes (Harmony, Pronto, etc.).I can easily think of a 100 other similar examples — iOS apps do far more than websites could possibly be capable of. In most cases where sophisticated apps are concerned thinking that they can be supported by some sort of advertising model is equally absurd.If we are talking about apps that take half an hour to develop, by all means, go with the free ad supported model. But sophisticated, difficult to develop apps are easier to monetize being sold, and the end users would likely prefer paying for them than being confronted with ads. To go back to my above example — think of the absurdity of using your remote to skip the ads on your DVR recording only to see ads on the remote itself. No thanks.
but what happens when someone develops that same universal remote app and makes it free?watch what happened on the weblearn from it
They’ll run out of resources to support it. Free doesn’t always scale…at least not well.
Actually the software part of many of the universal remotes is free –they will charge you $40~$80 for the IR (infrared transmitter/receiver) dongle without which it it is useless.Nobody is going to start giving away the hardware unless they want to go bankrupt fast (see: :CueCat).
You’re correct, they’re not the share leader. I didn’t communicate clearly. I was trying to communicate that I believe Apple will remain the share leader over Android in the short term.
Off the bat, from that chart, Palm is sure to drop off. But it’s not inconceivable that this sector remains in a 4-horse race for the foreseable future, because I’m not discounting Microsoft’s inertia, nor RIM’s ability to at least maintain their heads above water.The conundrum remains- and we beat this to death already on this blog: Why develop on 4 platforms? Which one to start with? The cross-platform choice (phonegap, titanium, rhodes) is not necessarily the answer, because they compromise on the native look, feel and features of devices.HTML5 has one year to prove its mettle. I’m tired of hearing about its future. Its future is now. Let’s see some more apps that use it with CSS3 & Javascript and show kick ass apps. We’re embracing this paradigm fully in Q1 2011.It’s not entirely correct to say that developing on one platform gets you only 25% of the market. What if 80% of your target is in that 25%? Just saying…that the market share numbers don’t translate directly into customer segmentation numbers.Finally, I thought you would show everybody this cartoon on how iPhone/Android/BlackBerry users perceive each other: http://thelifestreaming.blo…Everybody…Enjoy your (mobile) Sunday!
That, and the fact that RIMM just bought a design company to deal with the OS, and I keep hearing positive things from the users end about WinMo7
“The cross-platform choice (phonegap, titanium, rhodes) is not necessarily the answer, because they compromise on the native look, feel and features of devices.”Rhodes supports all the native features and feels of Android, iPhone, Blackberry, Windows, and Symbian! Rhodes lets you take advantage of device capabilities such as GPS, PIM contacts and calendar, camera, native mapping, push, barcode, signature capture, and Bluetooth.
Thanks for clarifying this. We should talk. Do you work for Rhodes?That link you just shared with the table comparison is excellent: http://bit.ly/dtbkPn
10/10 link of the week William.Compresssed cross platform mobile utilities all in one post at a level non-techies can understand.
Thanks Mark! How about the Android/iPhone/BlackBerry cartoon in same post, in earlier comment? I thought it was priceless.
Absolutely, great information in a table form.
Hi William,Yes I do work for Rhomobile which is the company behind Rhodes… I’m here because I feel that mobile app development needs to be standardized and eliminate the entire question of “which smartphone OS will win” and let developers build once and deploy on all OS’s. I will get in touch with you to talk!-Alex
“HTML5 has one year to prove its mettle. I’m tired of hearing about its future.”There’s a big piece left here that isn’t in place. Most platforms don’t have a lot of the phone’s functionality available in HTML5 yet.That’s not a HTML5 problem. It’s a mobile OS problem.So expect to continue being tired – HTML5 is going to be great in 2011 for a variety of things, but it won’t “prove” itself per se that quickly.
2011 should be interesting, but I wouldn’t under-estimate the fact that Apple is a big booster to HTML5. And Gmail as a perfect example- it’s got HTML5 all over it and works great on the iPhone & iPad, almost better than on some web browsers.
2011 should be interesting, but I wouldn’t under-estimate the fact that Apple is a big booster to HTML5. And Gmail as a perfect example- it’s got HTML5 all over it and works great on the iPhone & iPad, almost better than on some web browsers.
“Apple is a big booster to HTML5″I really hope you’re right but we shall see.I watched intently when Steve Jobs said “we support two platforms” and described the differences between them.It didn’t sound like he wanted to completely open up the phone’s functionality to HTML5.After all, that would commoditize his app store and the competitive advantage he talks about on every investor call.
“It didn’t sound like he wanted to completely open up the phone’s functionality to HTML5.”Go back to the iPhone intro, when HTML was all they had. Greatest Thing Since Sliced Bread. !!!(C) And the audience sat stonily on their hands. Don’t tell me that they’re deprecating it just cuz they don’t push it now.“… that would commoditize his app store…”WAAY down Apple’s list of concerns: if an app works well in HTML5, and it’s easy to develop there, then it will be. But of course, there are lots of apps that will require a couple iterations of Moore’s Law and HTML12 before they will be half-decent in a browser. HTML is a fine way to enrich user interactions without Apple trying to talk developers into a proprietary library that few will support: remember the Rhapsody fiasco, cuz Jobs sure does. (Got him his job back at Apple.)Meanwhile, Apple just wants to sell hardware. The store, with its smooth desktop-laptop-pad-phone-TV ecosystem (at least in customers’ mindset) is just there to sell the hardware.This should be a pretty sweet time for smartphone developers, what with a huge user base on phones & tablets, increasing similarity across platforms and to the desktop to leverage skills, even a whole new set of toys (albeit, “complex”) in Android’s NDK.
Things have changed since the iPhone intro. The app store is now Jobs’ justification for why iPhone will never be beat by another platform.And he literally said a keynote or two ago when people were complaining about app store rejections…”we have two platforms, one can tap into all of these things on the phone, the other is HTML5 and has no curation at all.”So we will see if HTML5 is actually embraced on the iPhone.
I’m firmly in the camp that says that you provide the optimal user experience for the medium you are distributing your product or service on. Which means developing on 3 or 4 native platforms in all likelihood. Expensive yes. Time consuming, yes. Pain in the neck, yes. But if the value of a subscriber to your ecosystem is greater than the SAC – then it makes sense.Interestingly, if everyone follows that thinking – and we get a ton of native apps – and god browsers, then it won’t really matter what phone you have – what will really matter is just the data network you are on (and whether you need a keyboard or not!)
I really think that it depends of the features of your app. If it’s plain vanilla, why bother?
Fred – good post, I agree based on the anecdotal evidence I see in daily life.Here’s some: my kids went Android for their smartphones. They love Apple (we’re a Mac house thru and thru) but dismissed the iPhone based on the cool factor of “open,” and control over how they wanted to personalize the phones. They immediately founder helper apps to use with iTunes – zero barrier for creative millennials there.And I’ll be moving off Blackberry myself for the next phone. Had been waiting for Verizon iphone but will will undoubtedly go droid instead.I think handset/model choice is a huge factor in all this. My daughter, a college freshman, gravitated toward the HTC sets while my son really went for Motorola, and he wanted the keyboard slide for texts. They did their research.
nothing like college aged kids for market research!i watch what my kids do with intense interest
I did a quick search, but couldn’t find the data on this. My gut instinct is that iOS users are more likely to pay for apps than Android users, and I bet this is the case by a wide margin, despite the growing market share that Android holds. That’s not to say that one shouldn’t develop apps for Android, but if selling apps is your goal, and my instinct is correct, then iOS should be your primary focus. Additionally if the gap in paid apps remains large enough between iOS and Android, Android can still beat iOS in market share, while iOS beats Android in app sales.Long story short largest smart phone market share != most paid apps.
You are probably right. iOS buy more apps. But very few businesses have reached El Dorado just selling apps. I think that the amazing thing about mobile is being able to be reached by your users whenever they want. And then you do what you do to make some profit. That’s why so many apps are free. For those businesses the number of users is quite relevant.
i think mobile ends up adopting web economics and paid apps don’t have a bright future because i don’t see a lot of paid web sites
I would add a caveat for games, which are making a money on virtual goods.
rightvirtual goods is an awesome way to monetize free apps
any difference from revenues received from virtual goods, subscription renewal, paid apps? best strategy from hundreds of years of publishing is to diversify your sources of income.
@jamesI agree, 25% market share is underestimating Apple’s footprint in total app expenditure. I’d be willing to bet that Apple is the winner by a very, very large margin in total revenues from iOS app sales.Not to mention, iOS is bar-none the highest quality development platform. The best developers (well, most of them) and certainly the best designers are all focused on iOS. Why? Because they all develop on Macs.You need to look a bit above smartphones with the iPhone and consider the entire development infrastructure as an “Apple Experience”. We develop for the iPhone, because we own iPhones, and we own macbook airs. I’d love to by a droid, but I can’t justify it plugged into iTunes on my macbook air. It’s like wearing sneakers with a tuxedo. Cool, but doesn’t quite fit no matter which way you cut it.When considering the entire Apple “ecosystem” as a whole, we can’t forget about iTunes. iTunes makes it easier than anything to purchase apps– and the user experience is the best of any of the application stores.If you were a shoe manufacturer, wouldn’t you prefer to sell in the store with the bright lights, clean floors and clean presentation instead of somewhere in a back alley?
Mobile OSs post and you show your favour to Android again… Apple fanboys will come and demonize you again… so just a few quick thoughts before they take control to prove you are soooo wrong!-I think that people keep using Blackberry for a few more things than just the keyboard. Corporate policies being the most important factor there. As soon as IT departments start aproving other smartphones people make different choices (although some will stay with Blackberry).-The main cost for a smartphone owner is the monthly bill, not the hardware purchase. Buying the phone is relevant, but $200 upfront is nothing compared to a $50 recurring increase in the bill. And this favours Android because you have more carriers and options. It also favors Blackberry with BBM and reduced price plans.-Some Apple users love to think they are and elite and that they know better. How big can be an elite before it becomes the mass?-Will Google be able to reduce the problems from fragmentation in OS versions and devices specifications? For what I read in many places it’s becoming a huge problem for some developers.
We develop mobile apps and I get a lot of idea from what our clients ask us to do. Until now, the IPhone has been the first choice. However, in the last 6 months almost every client has asked us to develop the app on Android too. As the chart indicates more and more clients seem to now ignore the Blackberry & Nokia.I still think if revenue through sales is the intent of developing the app then Iphone should be the first bet and will continue to be so for some time. It is so much easier to find a good app and then pay for it via Itunes. However, if reaching their consumers and gaining spread is the intent then the Android should be the preferred platform.
If you are charging for the app, I would still recommend to do iPhone first. From everything I have seen both directly and indirectly Apple still crushes everyone else when it comes to paid apps. While it is only 25% of the phone market I bet it is much higher when it comes to the paid app market.However long term I do agree with your overall position, Google’s open platform will end up winning for sure.
i am not bullish on paid appshow many websites are paid?
I think you are 100% right on that. But for at least the short term Ithink that paid apps can help to fund companies for futuredevelopment.But I think it is a whole additional post about how people canleverage paid into free as mobile will have to folllow the web. Sortof like what angry birds did with android.
exactly
So how does one profit from developing a mobile app? By embedding ads in the app? Or are mobile apps useful mainly as extensions of brand for web businesses?
Or a third choice…your business model is profitable without fees or advertising on the web. Mobile can be just another distribution mechanism for the same model.
how does one profit from building a web app?web and mobile are ultimately the same thinga way to access internet based services
“how does one profit from building a web app?”How does one profit from building one?________________________________
web and mobile are not the same thing. the former is primarily a vehicle for consumption of content while the latter enables rich functionality for productivity and entertainment. until HTML5 gains mass adoption web and mobile will be uniquely different.
Two thoughts for consideration:For iOS the market is bigger than iPhone, though the value in developing for iPad and iPod touch obviously varies depending on how ‘mobile’ you need your mobile app to be. It will be intestine to get a read post-holiday on how many iPads were sold with 3G & how many of those were activated, too, as that will influence the true size of this maker for developers.For Android, I think there are some obstacles to be worked through regarding distribution. John Gruber has written about this at daringfireball.net, touching on the lack of IP protection for developers as well as the difference in consumer experience that the Android marketplace(s?) provides versus the cozy, curated, trusted environment of the iTunes app store. Add to this the OS and hardware fragmentation issues that Jonathan Berkowitz alluded to in this thread (though Apple has experiencing this to some degree as both iPhone & iOS turn 4) and you are really only ever dealing with a subset of that subscriber number.(Come to think of it, I think Apple uses cool apps that leverage current OS and hardware configs to spur users to stay current, not just through the app store but through their TV ads as well. Clever.)None of this is to say that Android won’t be the dominant player in market share and it may even be the biggest opportunity after various fudge-factors are applied—the iOS opportunity is bigger than just it’s share of the phone market, though, and distributing to those individuals is more straightforward.
yupbut i think tablets are a slightly different opportunityi love the iPad. we have two in our homebut the computer you have with you all the time is the phone
That is why I think Jobs was showing some concern underneath as he did all of his lambasting of Android…particularly the size of the screen for the tablets. His claiming ten inches had to be the established norm just isn’t true…and the confessed surprize over how many iPads were purchased by businesses shows the speed in which we are moving toward a more customized size tablet, depending on the industry.That is why you MUST develop for both iOS and the Android Tabs.
ok, so what about the Samsung Galaxy? I completely agree with your post – I am a Blackberry user and my next phone will be a Droid – but which one? I have to admit I have a bit of iPhone-envy – I like the UI, but will not get an iPhone for many of the reasons you have indicated here – I won’t even get an iPad, I’m getting a tablet when they become mainstream.Does anybody here have experience with the Galaxy yet? It seems that it could possibly fill the use case of phone + iPad in one device.
I can’t figure out why I would want a phone that big.And I’m not going to buy two tablets (at this point) so I don’t want a tablet that small either.
I think that you are confusing the Samsung Galaxy Tablet with the Samsung Galaxy phones, which are only slightly bigger than an iPhone. The Nexus S presented today is a variation of one of these.
I was responding to “the use case of phone + iPad in one device” – I thought he was referring to the Galaxy Tab.I keep hearing people say that a 5 or 7 inch tablet is going to fill both use cases and I’m not sure I buy that.
Then I agree with you. If it doesn’t fit in my pocket it’s not a mobile phone! Skype didn’t turn my laptop into a mobile phone.
The reasons most companies focus on iOS includes:1. iOS users buy more apps and spend more on those apps.2. Development costs are lower for iOS.3. Testing is much simpler compared to Android.This article completely ignores those kinds of nuances.
i think the development cost issue will go away quickly as devs get comfortable with Androidand i think android users will buy plenty of apps once google fixes the payment/purchasing in androidat the end of the day, you want to be where the users are
The other huge advantage that android will have is that with iOS you’re limited to a single distribution channel.
Two more reasons I think iOS will maintain a lead on the apps front in the near future.1) DiscoveryIn terms of on-device discovery, both sides could still improve substantially. I feel like most of the apps I download on my phone I discover through word of mouth. I think as recommendation engines like Apple’s Genius make their way on to the device with a focus on apps, this experience will improve.For discovery off-device, Android’s biggest challenge is that their ecosystem has no “iTunes”. I’m dumbfounded that I can’t manage my Android phone “in the cloud” since it’s already tied to my Google account anyway. That said, you can imagine a time in the not-so-distant future where you can discover an Android app on any site and click a link to “add to my phone” and Google automatically tells your phone to download the app in the background when it has a connection.2) Ability to TransactPaid apps are still a really tough putt on Android. The only way to take a transaction is via Google Checkout (unless you are on a handful of carriers who support operator billing). Apple has the massive iTunes account database (every iTunes account has a credit card or PayPal account linked to it for payment). PayPal integration with Android Market would go a long way for Google, especially in international markets, where credit card penetration is much lower than the US and a lot of those less expensive smartphones will sell like hotcakes.Of course, you can always opt to go the free, ad-supported route, but I think only a small handful of iPhone apps can either draw huge sponsorship buys from big brands or aggregate enough audience to deliver meaningful impression numbers.Seamless In-app purchase still only exists on iOS. This is a huge drawback of development for Android. Tons of game publishers are adopting the free-to-download & monetize with in-app purchase approach (like ng:moco). No way to do this on Android yet.I think both of discovery of apps and the ability to make money in the Android Market are bound to get better in the long run, but if your business model relies on making money from paid downloads or in-app transactions, it might be better to prioritize iOS for now. If the business is solely ad supported or uses mobile as an extension/utility on the service, then now is a great time to get on the Android bandwagon and ride the growth.
i think discovery is a huge opportunity for entrepreneurs on android and not an opportunity at all on iOS because of itunesand itunes is leaderboard driven and horrible for discoveryso my bet is android will be much better for discovery once entrepreneurs start focusing on it
Fred, i disagreethere are tremendous startups building discovery services on IOS right now. Many of them monetize with much ease than on Android. the reason? because most of the discovery is not taking place in the app store any more. The download does. but not the discovery which occur on a set of opportunities like social networks, blogs, media posts, conversations with friends…and many more.The leaderboard indeed is a driver to discover based on hit. But not based on needs.There are also many discovery services for android. Some are really good (disclosure we created one with appsfire). But the problem is that the monetization is just not there and for 10 requests from an advertiser for apps on iOS you get 1 for Android. To date.The switch is not happening yet.
i stand correctedso discovery is a big opportunity across all platformsbut it is certainly not iOS’s advantage because of iTunes
i am sure you ll agree since you have a portfolio company in the discoveryspace 🙂
(1) True, but that’s arguably not a good thing for Apple. Right now you have people making money on fart apps (still..I know some.) That’s because iOS developers are conditioned to pay, and partially because it costs a lot more to get going with iOS development. You need to buy a Mac and you need to spend $99 to get into the developer program. Contrast that with free for Android, $25 to publish on their market but free elsewhere.That’s all a problem since the lower end of the market will get those apps for free on Android. So you can buy an iPhone and pay for your apps or buy an Android device and get them for free. Hmm.Net net, I expect that for-pay apps, outside of games, are likely to go away or transition to either a subscription model (remember the milk) or ad supported. Apple has got to drive down the end user cost for apps to what Android users pay, or be at a serious disadvantage.(2) Done both and I disagree. Think it’s easier to develop for Android. This may be my bias, you have a reasonably hard to approach language (Objective C) versus a simpler one (Java). The frameworks are similar enough; I think Android is better thought out, but YMMV.(3) True for games. Have not seen this at all in my apps, which are not games. On the whole, both iOS and Android are pretty simple testing environments, and fragmentation is vastly overrated (IMO.)Now, iOS had *better* fragment for the games guys, and fast. Right now, Apple hardware is very homogeneous, because Apple really hasn’t done much innovation on the hardware side. Android is about to be at least one generation ahead (dual core, next gen GPUs) on phones and already is for tablets (Tegra 2 tablets.) If Apple doesn’t match that, they have deep problems. If they do, they will cause fragmentation.So iOS will fragment. Or die.
So you are saying that developing for Android is $74 cheaper, so thus one should develop for Android instead of iOS? Seriously?Anyone who isn’t willing or able to invest $74 in their development work isn’t a professional computer programmer and never will be. Real programmers value their *time* for developing a full blown app much higher than $74 (or the $600 cost of a Mac Mini for that matter).
1. It isn’t just fart apps that get paid for :-). In fact it is mostly NOT fart apps that get paid, the majority of those are free.The fact that the iOS user base are prepared to pay for their quality apps, is only a good thing for Apple and for developers, and ultimately users because there are far more quality applications available on iOS than there are for Android. It becomes a virtuous circle.As for the costs, Ted_T already nailed that one – the setup costs are so low in either case, the investment is in TIME.2. It will depend on the background, a Java programmer might find Android easier, but all of the developer comparisons I have seen rate iOS as superior – the frameworks aren’t actually “similar enough”, the iOS frameworks are more evolved and capable. But probably Android will catch up on that front. A lot of people will consider Java a disadvantage, and Java isn’t known for being used to produce a nice UX. At the moment the development costs are lower because you can do more using better tools with iOS.3. Any apps using hardware-specific features will suffer from this, and a lot of apps with resolution dependence will also suffer. For any app that stretches current devices, testing will be a larger effort on Android.As for fragmentation, the benefit with iOS is that fragmentation will be limited. By definition, the Apple closed systems mean much less variety of hardware. Both a limitation, and an advantage, but the advantage is a good one.As for iOS dying because of not fragmenting… yeah right 🙂
Have you watched as popular games and apps try to move from iOS to Android? It’s horrible. Look at Angry Birds even, struggling to support Android users because of the fragmentation.It’s not pretty.
Indeed — also Rovio has ended up with a bunch of angry customers on Android. Go to their website and read the support forum — not only are people complaining about how poorly it works on Android handset X, Y or Z, they are bitterly complaining about the ads (as Angry Birds as per standard Android practice is free) and giving instructions on how to disable/block the ads. What sane person wouldn’t rather spend 99 cents on a game they play for hours instead of putting up with ugly ads?The free ad supported model that Android encourages is broken — for some of us, anyway.
I just bought Angry Birds Seasons for 99 cents last night. My kids love it, I love it, my wife loves it. The longer I have an iPhone, the more comfortable I get spending money on apps. My spending is increasing, not decreasing because of the convenience and utility of the app store.I can’t imagine getting a game for free and then complaining because it doesn’t work right, but I’d also prefer to own an ad-free version so that I can focus on the game experience.On the topic of support…I’m constantly amazed by the technical skill and knowledge of most iphone end users. Support has been wonderful.
just wait until angry birds develops a virtual goods model and makes its app free for everyone
I haven’t had any issues with Angry Birds on the Droid but that is besides the point (and half my home computers are Mac)…Ask yourself: Will Android and companies like Rovio eventually smooth out and fix the free ad support model you cited as concerns or simply throw in the towel and say forget it, everybody can pay up! The market in my opinion is still going to go hard in this direction over time as people like free stuff. The typical consumer is used to free web content, free email tools, etc. No matter how irate an Angry Birds support forum may look at the moment I foresee the kinks being worked out and the moderate smartphone user just fine with downloading freeware.Both Apple with the iPhone and Google with Android are already immensely profitable, but what was once an iPhone and everyone else market now begs the question of whether or not developers need to be strongly considering an OS which may be the market share leader in a brief amount of time. Good debate all around, and a debate that few would have imagined after the first couple of iPhones hit the market and changed everything.
that’s what i was trying to sayyou said it betterthanks John
You could add:4- Hype 🙂 (And I am an Apple fan boy)
Agreed :)But part of the reason for the hype is the quality, and the sheer wave of excitement that Apple creates with its products 🙂
I wonder what those numbers look like if you include non-phone mobile OS devices. (iPod touch, Galaxy tab, palm PDAs, etc)
From a Start-Up’s perspective, HTML5 just makes complete sense. When we were looking at developing an app for our school, we found that a high percentage of students had RIM phones. This meant that if we wanted to build a scalable native app, we’d have to build it for Iphone, Droid and RIMM…tripling our initial development costs.That in and of itself isn’t a deal breaker – what is a deal breaker, however, is the cost to iterating. We would have to iterate on three platforms every time we wanted to change something, and that simply wasn’t sustainable for a lean startup.HTML5 is here today, you can do almost all of the same things that a native app can, and development costs (in terms of time and money) are compressed both initially and over the long run.
I agree; my startup is planning on going 100% HTML5 to lower the cost of deploying on mobile devices until such a time as user demand sustains a move to a dedicated platform.Thanks for the tip on Phone Gap, Fred!
Only HTML5 is not consistent across platforms. Here’s a great article, about the hard realities of developing mobile websites:http://www.alistapart.com/a…Support for different tags is all over the place for different browsers, making it very costly to support a wide variety of devices. Better off looking into various tools that can create native applications for a variety of devices (the mentioned Phone Gap, Titanium, Adobe AIR, etc.).
Agreed. Starting out with mobile-optimized web gives startupers access to broadest base of users (most smart phones plus desktops), then customer research can determine which (if any) native apps are required.
Broadest base, but without any marketing machine.
Use HTML 5 to cover all the bases – but still develop a native app for your core market. We are a LONG way from HTML 5 (or a browser experience) offering comparable quality to a native application. If you want to really create great experiences and engagement with your audience, give a iPhone user and iPhone experience, an Android user an Android experience. Natively. That is how you will break through the clutter and win loyal fans and users.
Welcome to Chrome OS 🙂
To your second point Fred, one of India’s largest local manufacturers just launched their entry level Android device for approximately $160 – this is non-subsidized and has no contracts. I think they must have read your mind when they came up with the name of the website – http://www.myfirstandroid.com
tmo is selling the non subsidized Comet for $154. it runs froyo.
Now ask them to take it down to $60.
Does it have access to the Android marketplace? Is there a model for monetizing your free app via ads aimed at an Indian audience?There is no doubt that Android will end up with more units worldwide than the iPhone, and possibly more than iOS (iPhone + iPodTouch + iPad), with a good chunk of the installed base being in China, India, as well as numerous other countries in Africa and Asia. Most of those Android phones will not have access to the Android Marketplace and will run decidedly non-stock versions of Android (like the ones in China that have Baidu as it’s search engine instead of Google.)What will be the benefit of all those extra Android devices to app developers wanting to make money? That’s an issue: one iOS device is more valuable to developers than one Android device.
I think Android has an edge because the number of devices they make for it. They have so many different kinds of droid phones, which allows them to tier the prices on the devices, The X, Evo, Incredible, Backflip, Pro etc. They can market a certain device for a particular demographic, and add software to it. Let’s say droid comes out with a device for college students that has some sort of promotion around a university campus, then students will jump on it. My thought is droid will even grow to a greater market when Droid tablet devices become popular. The challenge iPhone is facing is that it only has one device, and not everyone wants to use this device despite all the apps. I think it also depends on market, if you’re looking at the European market Nokia and Symbian market is bigger, yet still there’s Android. I think Android did a better job at marketing in some ways, because they focused energy on traditional marketing, opposed to putting stores up as Nokia did, and then later had to take some stores down. There are things like the Maemo browser which are certainly good to look out for, yet it hasn’t seemed to gain as much publicity as the Droid.
Two other factors: Many people don’t have smartphones because of the cost of the plan, not the cost of the phone. There are many kids who want an iphone because they have ipods. Apple has seeded the market so that when kids can afford the smartphone plan, the graduate to an iphone. I’ve seen this a bunch of times.
Starting from this point:”I care a lot about where our portfolio companies should be focusing their precious mobile development resources.”I think you are looking at the wrong data (phone market share) to get to your conclusion. This analysis is based on the underlying assumption that a phone market should mimic the computer market. Network effects being the most important determinant of a winner. Maybe that happens, maybe it doesn’t. Nobody owned a computer before GUI operating systems.The important distinction is that phones were just phones before. These new OS’s are added functionality, but that functionality has to be learned over time. So the data you need is not phone sales, but propensity to use and download the apps on your phone.I would imagine this kind of data would show it is a two horse race between iOS and Android. The RIM users who actually download and use apps are those most likely to defect in their next upgrade cycle. Microsoft has something to keep an eye on and HP is focused on printers. Given this, I would imagine all of companies in your portfolio should not ignore ~50% of a market, but the choice on which platform they rank first will depend on the type of customer they have/target.Understand your customers, what devices they use and what distribution channels you have at your disposal to reach them. Execute on those parameters today and watch the data for the strategy of the next cycle.
It´s very easy to describe, it´s like Linux and Microsoft on the PC.Windows has always been a “product” for the PC. Linux and open plattform, with nearly no support.iOS is a product with support, which costs something (same as Windows). Android is Linux (somewhere it really is), an open plattform with no support.You are right. Android market share raises. But as a developer, you have to care about a lot of versions of the Android OS. Every hardware producer has his own driver for the smartphone display. If a new version of Android launches, the producer has to update, for example, his display driver. The point is, that the producer is not interessted in updating, because he already sold his hardware.With iOS the developer do not has to care about all these issues. It is just one hardware (or two), the iPhone and the iPad.Hope that Android will solve their problems with the versions. Actually it is nor really a product.
The growth of Android is even more impressive when you look at the last 18 months.We looked at the change in the market from 18 months ago to today and saw some impressive stats. iPhone had over 2/3 of all page views before Android launched and today Android trails iphone by just a couple points. iPhone has dropped by more than half.
i’m seeing similar data on AVCandroid still lags iPhone, but the growth is amazing
While I agree on the overall vision, we too often read data about market share, and subscribers, but I think it’s also important to look at USAGE. iOS is still leading the wave in terms of web/apps usage in rich countries, while Nokia (hello !) is by far leader in the emerging world: http://bit.ly/g7rKKIIn the PC era we could look at the US market and then extrapolate and make it an internet success. In the mobile era, emerging countries need to be part of the dashboard and strategy. Keeping an eye on Symbian is important, and thus maybe your open source HTML5 library is an interesting option.
i think you’ll see usage of apps on android catch up to iOS once you get high quality apps on android
I’m a developer and I read a great book recently about developing iphone apps with html5 + phonegap:http://oreilly.com/catalog/…even though I was really skeptical of this book at first, after reading it I think this approach makes sense for many, many apps out there and I plan to use it for a few of my personal projects. things like sencha touch are making cocoa touch/objective-c more of an option than a requirement, which is a good thing for the market and choices like this allow us to make (hopefully) better, more informed technical decisions.that said, I read recently that sony was adopting objective-c and I wonder if that will broaden the market for that skill set a bit.
In a few years we will look on accessing the internet via apps on a smart phone as we now look at accessing the internet via AOL on a computer.It was a phase that we got over and moved past.
As with most things – the answer will vary for every company. If they’re developing something for global use, then I’d ditch the U.S stats – they’re pretty misleading when it comes to Nokia’s overall market share around the world.I’ve managed app development on all the major platforms, and until now the two leaders have always been Symbian and iOS. Symbian has the market share, and Nokia are making improvements in terms of development tools and the Ovi Store – meanwhile iOS owners have been the most app-happy downloaders.Android’s increased share and potential is definitely coming from the huge increase in lower cost handsets – the problem is that mobile advertising revenues are consistently dropping in line with fixed internet prices, and if someone is parting with as little money for their phone as possible, they might not be the most lucrative customer.Windows Phone 7 is really interesting as it’s the first platform to really go in a different direction, it’s got a lot of marketing and advertising spend, and the OS is pretty good. It’s also a decent attempt at making something more intuitive to non geeks, and should open up the market even further – plus apps are pretty quick to develop, depending on the functionality you’re after.
I love these articles from you. They always start a firestorm. There are positives and negatives to every platform on the market right now and which ones you choose and don’t choose needs to be business based, not adopter based. If I am in gaming in the US market I am looking hard at iOS first and Android second. If I am corporate or government, RIM first, Android second. Social networking then Android first, iOS second. If I am primarily targeting emerging markets than Symbian first, Android or iOS second. It isn’t one size fits all.Given that, if you were to ask me to talk to your portfolio companies about mobile, I wouldn’t tell any of them which platform to pick. I would tell them that if they are a mobile-only company they will fail. The future is in a connected environment and access everywhere. The web is critically important, not just as an interface but as an intermediate point. The important part of this isn’t which mobile OS you pick in the long-term, it is about using mobile as a distribution platform rather than as a business model.
This was also my initial thinking here, since the in the study one is concentrating only in US markets forgetting other parts of the world which is around 6 bn people. Once Indian people start using smartphones then the game is changing raplidly!
I say choose EITHER.Because… I totally agree with eliafreedman here. The cloud is poised to ultimately be a hub/arbiter for an access-everywhere product/service. Thus a business is really reliant on *multiple* mobile platforms serving as UI endpoints, or sensors. So I would have to agree that the long-term focus is multi-platform, from a mobile standpoint. Said another way, to choose only one platform implies that there is some functional dependence on your product/service to that platform, inherently. This is almost never the case.However, Fred is focused on the zero-sum scenario where mobile R&D eats precious early-stage capital. Most agree this scarce-capital scenario is real for startups.Consider this: that both Android and iOS installed bases are large enough, with a diverse enough set of users, and similar enough capabilities, that if a startup had to absolutely chose ONLY one platform to launch their mobile strategy with, then EITHER would be both sufficient AND insufficient. Sufficient to establish your features and refined your home-run use cases via early adopter feedback and system performance. Insufficient to grow into a fully realized mobile modality for a “sustainable” connected product/service offering.In the EITHER school… the “Team” is probably the biggest factor about which platform to build for. Some technical teams simply know one of the mobile platforms much better than the other, or are better prepared to do the added testing required for Android’s more highly fragmented device-space.I also think it’s fair to say that many founders will subjectively want to pick one platform over another for marketing/evangelism purposes. There are social considerations for each mobile OS ecosystem, and some folks have the necessary feel or inroads to establish clear signal for their message more effectively in one of the two noisy landscapes. Some good-reasoned resulting bias is unavoidable.
I was just at the Sencha conference in San Fran where the CTO of ATT was the keynote. He made it very clear ATT was committed to HTML5 on all of its platforms. Sencha has hired several developers from the webkit team. (webkit is the browser in iOS, Android, Palm, the latest Blackberry, and Symbian) And you have the major platforms adding hardware acceleration and rich features for Javascript.Applications developed with Sencha Touch (a HTML5, CSS3, Javascript platform) are hard to tell from native apps. Wrap them up with Phone Gap and you have a very robust cross platform solution. We’re also seeing studies showing significant developer productivity gains on HTML5 over native code.All together, unless you have a real strong reason, its hard to justify not using HTML5.
I’m sorry, but Sencha Touch does not look like iOS or any other mobile OS. Sencha Touch is at best a very poor imitation of some features of iOS, kind of like Window 95 was to Mac System 7. The widget they have lack sophistication and subtly like iOS widgets, as well as functionality. Every time I look at their Kitchen Sink demo, I’m reminded of an interface created to resemble baubles for toddlers–it just looks cheap and toyish. They really need to hire some good designers and some people with better CSS3 skills. I would not recommend Sencha Touch for anyone trying to create a product for commercial user. Freebies, maybe, but don’t expect accolades for the default UI.
More reading: Measuring the Nomads by Frédéric Filloux at the Monday Notehttp://www.mondaynote.com/2…andhttp://www.netmarketshare.c…
One factor about the business incentive for app developers that is not always mentioned in these dicsussions is about their ability to monetize their apps via advertising. Apple with iAd showed the ability to enable HTML5 advertising for app developers, but via a closed system. We think that app developers will have more options in 2011. I’m from Crisp and together with Weather.com, TringApps, JumpTap we have put forward a standard for open-sourcing the in-app advertising SDK using HTML5 principles called ORMMA which had an initial, positive review from IAB’s mobile council http://blog.ormma.org/ last week. A standard could open up agency spend for mobile advertising – 2010 was mainly about Apple winning spend directly from brands which is less sustainable.Even if HTML5 does not change the nature of how apps are developed in 2011, open HTML5 advertising across iOS and Android can change the economics of mobile advertising in 2011. Consider the example of Angry Birds in the past month. By making their app free on Android – they are seeing tens if not hundreds of millions of impressions a day http://googlemobileads.blog… . An app with that size can sell not only performance advertising but premium rich media campaigns to brand advertisers. Not every app publisher can slingshot themselves into plush toy sales, but having Rovio catapulted into the premium publisher ranks by virtue of launching an Android app is very encouraging.If a publisher / app developer wants to offer a free app, we think it makes sense to spend some time to get to ORMMA compliance on iOS & Android. They can start by looking at the BSD licensed ORMMA reference SDK. Here is a link to the google code project: http://ormma.org
I really don’t like how so many folks make blanket statements based on the US market only. Add in the fact that Comscore and other metric ‘guessers’ are often off by at least a factor 2 (or .5, if you like), and you really have very little to go on. As always, it depends very much on the market one operates in *and* the app. If you are creating a paid app, iOS is still the best way to go by a long shot, especially if you add in the iPod Touch marketshare. If you are talking a free app, HTML5 seems like the best bet to me by a long shot (why target Android only, when you can actually target iOS at the same time too?).HTML5 really makes sense when you remember that big sites already have a captive audience when they are visiting them on their smartphones. Rather than hoping some directory managed by Google, Apple etc sticks you at the top, you are actually targetting existing users of your product already. But hey, I’ve been saying HTML5 web apps is the way to go for 2 years, and so far most aren’t taking that route 🙂
I think the HTML5/CSS/JS SDK/Framework tools we’re seeing now (Sencha Touch, Strobe, JQTouch, etc) are going to help developers get closer than ever to blurring the lines between the open mobile web and native apps. I’m way long on mobile web and HTML5!
My personal study sample for the adoption of new tech channels/dev trends is always this:1. Mom2. Sister3. GrandmotherIn the last 6 months, both my mom and sister have purchased Droid based handsets as their first smartphones.When I’m making decisions about where my startup should go with mobile dev, I’m building for me, and I’m building for them.Excellent post.
my parents are about 80. they just got a droid x and are able to use it right away…
1)what do the flurry people say about the statistics?2) even with HTML5 and openess, I still think the market for paid apps is much bigger than first described, because as that keyboarded droid pro shows – there is a business market with much less tapped needs. The web on the go is different than the web right here. And the ability to leverage on the go efficiently is something I don’t see fully tapped out at all.Take lawyers who have to have documents ready asap at odd times. Engineers sending back and forth photos and plans for approval…those sorts of activities don’t do well in free land.
Anecdotally, all the major investment banks seem to be piloting or launching projects to support iPhones and iPads in addition to Blackberries. One of them said they thought based on the pilot, after it was rolled out firmwide they expected 50% of users to take advantage of it. (some of them might keep multiple devices, ie a Blackberry and an iPad).
we’ll see how android does in corporatesbig questiongreat comment
“Just look at this picture of the Droid Pro”Exactly why i’m going with the pro rather than any of the other versions. I tried it yesterday and it was great to have that keyboard with the functionality of the OS and internet speed. Its a blackberry killer easily
I think that mobile developers should extend themselves to the winning platforms, but not concentrate on a single OS. A good example for developers working on multiple platforms is Viber, the currently iPhone only free call app. Viber on multiple platforms (i.e Android and BB) should probably make my dream of at least 80% of my contact list being approachable with free calls. If it sticks to iPhone alone, it won’t be as widespread. I think that if developers show the can work their apps seamlessly across different platforms, that you can really bring communication up a notch, and mobile users won’t have to think what sort of phone their friends have before launching social gaming apps, free call apps or any other social app.
Fred I think you might be missing a point here. The 25% iPhone users are demographically very different, and from a use-case pov are also much more likely to adopt smart phone software. Android could have large shares of handsets but low shares of app users. Instinctively that feels like the case to me. So Apple’s share of handsets significantly under-measures its software footprint.To put the same thing another way (and this is a sociological point) Android is the equivalent of Symbian. Its free, hand set makers will embrace it, but the users who get the phones will not be early stage software users. If you want to make a real impact IoS is the clear front runner and the first to focus on as a developer. Android is a nice-to-have. HTML5 is a good way to go for some apps but if you need to be native and porting doesn’t cut it for your app, do iPhone first.
i don’t see that behavior with my friends and family who have androidsthey add apps like crazyi think apple’s PR machine wants you to think their customer base is better in some waybut i don’t think that is true at all
I agree.So one of the few slices of empirical data shows significant handset growth for Android, and of course a well-entrenched iPhone market share. Why is the next presumption that Droid users do not add apps like the iPhone customer? The use-case pov is simply conjecture. When I take a quick scan of my cousins, nieces, aunts and brothers that use an iPhone I see few using it for much more than pictures, text, facebook and MAYBE twitter. Given this forum I think many of the respondents view the consumer through a super-user technical lens, rather than the standard cell phone consumer; the vast majority of whom to do not yet have a smartphone. Many of these future smartphone owners will be just as happy with an Android operating system where you can do “that cool swipe thing with your finger, just like an iPhone!” and will not see much difference between the two, aside from maybe a higher cost for the iPhone. If you are an app developer and forecasting for the growing smartphone consumer, you need to look at the market with the typical consumer in mind, and I think that leads to the web and HTML5. Those that do not have a smart phone yet are not early-adopters and as a result, aside from downloading a facebook widget, see the app stores as overwhelmingly. Where they will likely be most comfortable is where they are spending all their time online, on a web site.
In the US perhaps the customer base isn’t very different, as Android phones have a similar cost of ownership to the iPhone (upfront price + carrier plan). However once the iPhone goes to all major US carriers, Android will have to start competing on price, at which point the US based demographics will start to change.Worldwide there is already a huge demographic difference between iOS and Android users.
Could you point us to some numbers backing this up? Like the amount of $$ spent buying Android apps versus iOS apps? Or are you using numbers only when they go your way?
Buying apps vs. adding apps is not the same thing.The only developers who think a customer base that “buys” apps is inherently better than one that “adds them like crazy” are ones with very limited vision.
jeez,can you chill on the hostility?this blog is a free service where i tell people what i think honestly and completelyit is my opinionif you don’t like it that is perfectly finebut i would appreciate a little less hostility
There’s an interesting chart here, comparing useage of the various app stores, unfortunately the data is a bit of a mess and not very up to date.. might be worth bookmarking all the same.http://en.wikipedia.org/wik…
I see microsoft throwing huge resources at wp7. Putting a massive effort behind it. Listen closely to Ballmer when he talks about it. I think that over time that platform is going to look better. Good enough to make it to #3 in the us and #4 worldwide. I see palm losing out. I see rim taking a big big fall. they have the most to lose. Management there seems to have their own reality distortion field in place. So I see ios android and wp7 all taking share from rim, which increasingly is the odd man out. Nokia is a wildcard. But they have to get through a massive product transition first. and that usually makes things look worse before they look better.
i can def see this scenario playing out
Do not confuse market share with app share (for free ads) and app revenues (for paid apps). I suspect, but do not know, that iPhone users buy more apps. Please prove me wrong in this assumption, but I find that teams TODAY will develop iPhone first and then Android, and then iPad and I think that will continue for some time.
but how much longer will people buy apps john?people don’t pay to use most websites
Quite some time I suspect. Music has gone from paid to free and back to paid and apps can do the same. What beats free? Convenience and simplicity.
Fred i am not sure i agree. If you are a small startup and have to allocate ressources on the healthiest app ecoystem right now iOS is clearly the winner but also the best way to iterate on your service before expanding to other platforms. The fact Android is going to get bigger is not enough as an indicator that this should guide your choice to develop there first.The second reason you want to start with iOS is because it is a much tougher and competitive ecosystem and that the rules learnt there to become successful will accelerate the learning curve on other platforms.Finally HTML5 is not a good bet right now, simply because there is no real agreed homogeneous standards for all browsers (the impression is that there is) and because there is not right monetization and distribution/discovery layer built for those yetI d say go first iOS, then Android, then wherever you see an opportunity,
Hi Fred,This is the first time I am commenting here. Couple of things I wanted to share:- Follow @TomiAhonen to know what is happening with manufacturers (on a global scale)- Read this blog: http://www.visionmobile.com…- Android and iPhone are equal in importance. – Symbian for volume in developing countries- RIM if you are focussed on youth (YES youth)- MeeGo and WM7 only if you get paid to port
Limited resources to me means focusing on a polished mobile and web interface, app later. Unless the business is dictated by interface (Flipboard for the iPad).Beyond smart phone apps I’d suggest text message integration first based on the next gen’s usage.
Great post, to which I agree wholeheartedly! BTW, actually it’s even worse for an iOS only strategy: it’s 25% of 25%, which is to say a tad more than 6% of the current smartphones market!
This is a tremendous post that I feel the need to blog about. Your enthusiasm for the mobile web smells like money. This post is futuristic. Saying all the right things about Android and HTML5.
A couple of things Fred.First of all many applications now are basically branded content or function given away as a brand. For this kind of application or service you want to balance the cost of development and testing versus the potential reach of the device. For this sector the iPhone is likely to win out for the time being due to the device specification fragmentation we are seeing on the Android platform.For applications that are either paid for or freemium services (like Flickr) it is less clear cut. Since these are personal devices is not only about the device specification but about the device context; for instance I am less likely to have a virtual beer glass on a work phone. Even if the device is for home use, if it looks and feel like a business device it may affect the way I relate with it. There is a whole lot of good socio-antropological stuff in there that we don’t fully understand that could reveal new markets. Its also the stuff that bootstrapped start-ups tend to be less good at understanding.
Fred Wilson On Android And HTML5 http://goo.gl/fb/bqscs
If you are looking out for your portfolio companies and their investments in the mobile arena then it would be better to invest in mobile applications services that can be accessed by the larger pool of un-smartphones.The growth for services that are based on SMS technology that provide services to the wider audience would be dollars better spent.But then again maybe a lot of the portfolio companies are not targeting that sector of society/economy either.Would you suggest to your portfolio companies to pick a platform to support and for you that means Android or would you say they should pick one and then build out for the rest as well.I still think the growth in mobile is not in the U.S. smartphone or otherwise.I still think there is a case to be made for the ability to do more with less and so far all new tech startups seem to still be able to do more with more and that is not a good growth model in parts of Asia, Africa and South America where the growth is.
BBM is the only reason i still use the blackberry. if KIK gains traction i will switch to android. RIMM is right from a business perspective to try to stop KIK
If you want a cross-platform IM, try Whatsapp. It has more features than Kik and works on BB, Android & iPhone. Also, Tango is amazing for free video-call & slightly less cumbersome than Skype.
its not what cross platform IM i want its what everyone else will use. 🙂 kik started to get real traction.
Yes and No. I still prefer Whatsapp over Kik. Kik is missing so many additional features.
Anyone know what happened to the ability to subscribe to a poster’s comments on Disqus? Seems like it disappeared after the revamp of Disqus – I am still subscribed to JLM, Fred and Suster but can’t add any new ones.
so helpful. thx
While I agree that Android devices will take the number one spot in the mobile space in terms of market share, I don’t agree that your portfolio companies should focus on them for development. Why?Fragmentation.Rovio, the creators of Angry Birds, are having trouble with the Android platform because of fragmentation. Their game doesn’t work across the board, and they have to spend extra time and effort to make sure that it does. They even went as far as creating a “Lite” version for not-so-powerful Android devices. For the iOS? No problem, develop it one time, and it should work across all the devices – because there is really only one make and model, not counting older versions.I’m not saying don’t focus on the Android, but you have to be careful about developing for that platform because you have to develop for so many different form factors with their phones. Expending extra effort on a platform just to make sure it works for ALL devices is such a large headache, and I’m sure your portfolio companies have enough to worry about.
For our apps, we just jumped on Android. As a startup, we all had iOS devices, but it is pretty clear that Android is going to win out in the end, so from a business perspective, the Android market was the logical choice.We still are porting, but have an “Android First” policy. We can work out the fragmentation kinks pretty easily, and from a dev standpoint, it is basically a non-issue. You just have to be a little more extensive in app testing.It’s really not about picking a side from a personal perspective, I love my iPad, but Google is just doing a very good job of getting market share.
That’s an interesting way to look at it, and I respect that you are doing so because there are a greater number of users.However, I will have to ask a few questions:1. Are your apps solely reliant on sheer numbers? If so, what is your revenue model like?2. How resource intensive is your app, and does your design change from platform to platform?3. Are you seeing a significant difference in the Android market and the iOS market?I ask these questions because you seem to be relying on getting a much larger market through Android, and you appear to be making the correct bet as they are rising to the top rapidly. However, will the return on your investment in the Android platform be a significant gain over what you had with iOS?I am genuinely curious.Also, while RIM is losing ground, they do still hold a significant market. Are you making any plans for their platform?
I think it would be substantially different if we were making big games or resource heavy. We really just have a suite for testing purposes and for name recognition later, with a full on server-based model much later. Right now the games we make are simple and the apps are more utility and consumer based.What is most notable about our experience is that we get so much more money from ad revenue than we do selling the apps. Google’s model is so much more valuable than what Apple is giving. The free versions of our apps bring in much more than our paid, which is not true on the Apple front. The two balance each other out, but when Android grows to the point we all expect it to, then the constant stream of revenue for the ad based free apps is going to be the sleeper force of this mobile app movement. I actually read about this somewhere else-http://arronla.com/2010/08/…Ironically, there it is a task manager, which is hilarious since such functionality is useless in Android now. This one person gets all of that revenue, and is on a much smaller scale than most other operations.As for RIM, well I used to have a Blackberry (before I had an iPhone and ultimately an Android device) and we just don’t develop there. I like them a lot too, but its just not the same thing right now. We’ll see how that new OS they are building pans out.
Thanks for answering my questions, really cool to find out these things. For your Apple apps, do you have free versions that run ads as well? If so, I am assuming the numbers you see with those are much lower?
If you intend to charge money for your application, the iPhone is the better platform because you can provide a better user experience without worrying about the particulars of the user’s device as much. Plus, iPhone users are more likely to pay for apps. If you are planning on anything other than app sales driving your revenue, Android is the better choice because it will have the numbers.Thoughts?
I agree – I know there are numbers that support the fact that iPhone users are willing to pay for apps, and the unified experience is definitely much easier for UI designers, so I am definitely with you.However, I am willing to bet that Android can support driving your revenue, simply because of sheer numbers, but the revenue you receive may not be worth the extra effort you will have to expend to make sure it works the same on all form factors.Boiled down: there is money in both platforms. I am simply arguing that one is more fragmented than the other, and may affect your business decisions because of the extra effort required.
as i’ve said now many times in this thread, i believe mobile economics will trend toward web economics. there aren’t many paid web sites. freemium works, but often you are in a free model.i believe the paid app model is a market aberration that will go away soon
I would disagree about your view on the paid app model. I put mobile apps on the same level as computer software – apps are little bundles of software or games that can serve many different purposes, whereas websites are limited in their capabilities. (For now.)Sure, people can pirate computer software as they can with apps, but digital distribution and fair prices have really helped create a strong app (and software) economy.I think mobile web sites will be picking up more and more steam and capabilities, which will eliminate some of the frivolous apps in the app store (As you mentioned with HTML5 experiences) but there will be many apps for sale in the app store that will do much more than any web site could.Besides, Apple is creating the Mac App Store and Windows is really pushing hard for Games For Windows, which tells me the model of digital distribution through a centralized location probably isn’t going anywhere anytime soon.
“Sure, people can pirate computer software as they can with apps,”A very significant difference comparing the Apple App Store to Google’s:You cannot pirate an app on the iPhone. Just not possible. I think this is huge and will make all the difference in the world for developers:A world with no pirates, a gift from Jobs and Company.
I’m not sure if you are joking or not, but you are aware app piracy is very much real right? Jailbreaking phones isn’t just a way to add additional features through apps that cannot be found in the app store, but plenty of apps can be obtained through non-legitimate ways.Hopefully I’m not just missing some sort of hint of sarcasm…
Here’s the thing Fred: you’re using numbers to cite why people should move to Android for development, but when provided with a proper rebuttal, you cite gut feeling.Though I believe the numbers tell a different story, it’s very possible that you could be wrong and the numbers will shift towards free, but right now, iOS still has the edge for developers when it comes to profitability.Every time I see a Droid user, I ask them if they’ve paid for any apps. Their answer is always no. The ask me when I will port my apps to Droid and my answer: as soon as you’re ready to pay for them.Droid works if your app is an extension of another product or service, but if we’re talking stand alone, it’s not even close.
everything on this blog is my opinioni thought that was obviousif it is not, let me restate iteverything on this blog is my opinionlet’s just see what the situation is in a few years and then we can say who was right
I know, I’m just giving you a hard time :)It wouldn’t be any fun if you played the middle.
I think the $0.99 is a real breakthrough that has permanently changed consumer behavior around app purchases. But I also agree with you that something else is making developers a LOT more money — in app purchases. 75% of the money entertainment apps are now making comes from in app purchases according to Flurry’s studies and here again iOS is way ahead…Android (or I should say Google) will get its act together but right now it is a MESS
NO QUESTION – here is the latest results from our survey of top developers – http://scr.bi/eCtqaq every month we talk to 100 top selling or downloading app developers about their experiences with either Android Marketplace, Apple’s App Store, or Nokia’s Ovi. The data speaks for itself – all the money is being made on iOS
The problem is more with the hardware than the software. Software wise Android is not that fragmented anymore, with 83% of phones being on Android 2.1+. The biggest gap in compatibility was from 1.6 to 2.1.And it doesn’t seem like they’re going to complain too much now that they’ve almost caught up to iOS revenues (1.2 mil/month on iOS):http://www.intomobile.com/2…I expect the Android revenue to surpass the iOS revenue soon because on iOS users only pay once, while on Android, the game has high replayability so even old users will keep on pressing those ads.
For the iOS? No problem, develop it one time, and it should work across all the devices – because there is really only one make and model, not counting older versions.You can’t discount the older versions of the iPhone, I’m sure they make up a significant portion of the market. They are just downplayed because iOS is largely unchanged in look and feel even though the hardware and associated capabilities of each handset is vastly different. For example, you still can’t send MMS messages on the original iPhone. The answer to that, from most fanboys anyway, is buy a 4G. My answer to Android fragmentation is buy a new device. Now we’re all on equal footing.
You’re right that you can’t discount the older versions of the iPhone, butI would like to argue that the much older iPhone devices (iPhone 2G and 3G) can barely run the new OS, it’s Apple’s way of pruning the devices that can’t handle the newer generation apps.Can you still develop an app that works for them? Definitely! In fact, the UI will probably remain constant regardless of what model the app is on, it’s whether or not the phone itself can support the app.That’s all I meant when I said “not counting the older phones.”
“butI would like to argue that the much older iPhone devices (iPhone 2G and 3G) can barely run the new OS, it’s Apple’s way of pruning the devices that can’t handle the newer generation apps.” -Understandable, I think it’s easy enough to think of the Android schism the same way. If you’re running 1.6 or some other old version, sorry not supported. Sure there are still some variations with the later versions but with 83% of Android devices on 2.1 or better, I think the schism is largely overstated and probably closer to Apple’s ecosystem than most would care to believe or publicize.
You’re absolutely right on the software front – the older devices can’t really run the newer OSes and aren’t given much support.However, software updates are also determined by carrier for Android devices (If I am not mistaken) which slows down the update process. In addition, I also mentioned that it’s also a hardware issue – many different devices offering different form factors, which means extra effort on the UI front.Not necessarily a bad thing, especially with numbers, but still something to considering when developing on a tight budget and deadline.
True, life within the confines of the walled garden will always be easier, your choices having been prescribed to you before you begin the journey of life. You know the lay of the land before you begin the journey and plan accordingly. In the wild west where things are a bit uncertain, you have to keep the edge on your knife sharp at all times, your gun always loaded and your wits about you wherever you turn. I think the latter breeds more innovation than the former, though the former is probably easier to use to a profitable advantage.Forgive the metaphors, I’ve recently watched The Matrix. LOL
Should have taken the blue pill, my friend.However, I’m not going to say one is more innovative than the other, I simply wanted to rebut Fred’s point about how Android should be the main target for apps moving into the mobile space simply because of numbers.Lots of variables at hand, and I would choose on a situational basis.
I recently saw an interview with one of the Rovio founders commenting on the fact that they now make almost $1mil per month in ad revenue from their apps on the Android platform, where all of their apps on iOS are all ad-free but cost money. I’m too tired to find the video right now, but I believe it was on TechCrunch in the last two weeks, I would check it out before making claims about other developer’s experiences with the Android platform. (and I’m 90% sure they said Angry Birds now works fine on any Android 2.0+ device)
Fragmentation is something on the minds of only of Apple fanbois. 80% of all Android phones are currently running 2.x. I repeat… 80%… and guess what? that’s more than iOS 4.x is running on iPhones. Running iOS 4.x on iPhone, iPhone 3G and iPhone 3GS is a complete disaster. In fact on the first 2 it can’t even be done. Apple has more of a fragmentation problem than Android will ever have.
You mean like the fragmentation which forced id Software to make Rage HD for iPhone 3GS and iPhone 4 and Rage for the first and second gen iPhones? Do you expect to run DVD authoring apps on netbooks with Atom processors running Windows 7 just because you can run it on on a core i7 machine? It’s not a OS issue. It’s a hardware issue. I can get a $200 subsidized Android phone, just like a $200 subsidized iPhone and run all the apps I want. Or I can get a $30 lower end Android phone instead of a dumb feature phone for the same price and run most Android apps. People who can’t afford or don’t want to spend $2500+ over two years on a phone will happily use a cheaper phone with a lower TCO over two years and enjoy an outstanding OS and apps instead of a dumb feature phone. Android developers get access to users who would otherwise not have had a smart phone at all. As a developer, I have been able to go from Android 1.1 to Android 2.2 spending no more than 2 days total across 5 versions. And my one app runs on 1.5, 1.6, 2.0, 2.1 and 2.2.The real fragmentation on Android is not a developer problem, but an end user problem. I can’t upgrade to the latest version of Android unless the manufacturer and carrier makes it happen. Of course all my phones are rooted, but this is not possible for the masses and this is what Google needs to address.
Fred I think it’s important not to overlook the potential iPhone marketshare once Apple looks beyond AT&T. You put a ‘Maybe’ in the post but imagine what these marketshare numbers would look like if the iPhone adoption percentage of AT&T customers was applied to Verizon and other carriers.I like Android and eventually it may overtake iOS on both tablets and phones, but by the end of next year? That’s a stretch… Especially if iPhone goes to verizon and they also release a new iPad.
You are assuming all users app using/spending habits are equal between the different OS. Are they? From the surveys i’ve seen, this doesn’t hold true.
One important caveat, though: with Apple, you need to create and maintain one (OK, two if you want to optimize for older phones) versions of your software. With Android, it’s many more and growing. You need to optimize for different screen resolutions, different OS versions (and often decide not to support 1.X, losing about 40% of audience) and often even for individual devices. It’s becoming more and more like J2ME once again. After that, you need to put distribution in place through different App Stores, OEM deals etc, while with Apple you only need to care about one channelI agree that supporting Android is fundamental, and that there are great opportunities, some of which aren’t available on Apple (e.g. widespread adoption in emerging markets), but it’s a completely different effort in terms of development and marketing. There’s no way a 2 or 3 person team can support Android properly across all devices and markets, while that’s possible with iOS
One reason I don’t develop for iPhone is that I would need a multi-thousand dollar development system. I can develop for pretty much everything else on a $300 PC with 8GB of RAM and lots of virtual machines.There’s definitely money to be made from apple users, but it’s always going to be a boutique platform. That’s their position. I’ll develop for HTML5 and give it a spin on an iTouch. That’s about as integrated with Apple as I plan to ever be.
Fred, there’s one important thing about the iOS platform that (as a developer on that platform) I think is missed when you view Android and iOS as just two different platforms.Developing for iOS is actually a more more fun, even though it’s a bit quirky. The iOS environment is simply much better thought-out and polished, and the Objective-C language is just Smalltalk with C control structures. Android is Java-based, and thus a lot more verbose and less dynamic; coupled with a less-polished environment, that adds up to a real negative for programmers who want to actually enjoy what they’re doing.
Author once -> deploy many is the ideal. (Hey! Wasn’t that what Java was going to accomplish for us?)HTML5 looks to fit that bill better and better as we chug along. Is that PhoneGap? Lord forbid Apple, Adobe or MS acquire those guys. 😉 Actually, I think you can count on each of them to come up with their own HTML5 authoring solutions once the standard is locked and loaded. It’s nice I can go from PhoneGap to native with just a little more effort.However, the whole channel strategy for marketing and reach seems to beg for native “app stores”. It seems that’s where all the eyeballs go. Think Angry Birds would be the phenom it is without app stores? Doubt it. And, it’s darned nice to not have to build a bunch of notification, subscription, in app purchase, analytics, version control, ecommerce platform stuff just to sell my apps. All that kit is provided (in Apple’s case) for their ~30% cut.I’m hard pressed to think of a WEB app I’ve paid for be it “in app” or as a subscription. So, the biz model there almost has to be ad supported. Native apps are another story. And, I’m not so sure I’d have found half of the apps I really thoroughly enjoy without some sort of portal or app store to make DISCOVERY easy.So, HTML5 has a lot of appeal but the native app stores hold a LOT of distribution power (for now).
I just want to share a real-world case study of an established player developing its mobile strategy using PhoneGap as a cross-platform development platform. I think it’s an important lesson in these early days of mobile development.Safari Books, which is (I believe) owned by O’Reilly Publishing, lets users subscribe for a monthly fee and read any of the hundreds of technical books in their library. It is a godsend for software developers, and you probably won’t meet a single one who hasn’t read at least one O’Reilly book.They announced their iPad development soon after the iPad’s release in April, with a target release date of July. July and August passed, at which point they admitted the app needed more polish before they released. The app was finally released on November 3rd, and it was immediately an absolute disaster.Developed in PhoneGap, the app was reviled and flamed by users within the first few hours. It was buggy, slow, and seemingly worst of all, adhered to none of the native interface guidelines used by all the other apps that the users had expected from an iPad app. The response was so bad that the app that had been in development for over 6 months was finally pulled from the Apple store entirely after 3 weeks.The question is: is it worth it to save your development dollars trying to build a one-size-fits-all mobile app using cross-platform tools? On the phone side, with Android, iOS and other contenders it is possibly a tougher call. On the tablet side, I see very few contenders to the iPad at the moment, and it’s clear Safari Books miscalculated their development gamble. Either way, I would personally prefer to deliver the best experience for a fragment of the market that be known for delivering a sub-par experience by the entire market.You can read Safari Books’ blog and follow the story yourself here:http://safaribooksonline.wo…
The interesting thing about Safari Books Online is that they don’t need a mobile app. I read my books every day on my HTC HD2 that runs Opera browser in Windows Mobile 6.5. I usually read several books at once in separate tabs.I use it through my library card account, which means I can’t even use the m.safaribooksonline.com site and have to use their primary “desktop” site instead.Thus, their entire desktop-facing site runs perfectly fine on my phone once I activate “HTML Reader” mode instead of Flash. I am sure the experience is even better on other devices.
You are correct Leonid, which is what makes their mobile strategy so questionable. Users already had a “good enough” solution with the mobile site. What users were really excited for was the translation of that service into the new tablet medium. See Amazon’s Kindle app as an example of taking your current service strengths and tweaking them to suit the new users’ needs (offline portability, and tablet optimized readability features in this case.)My point really was more that the choice to pursue cross-platform development appears to have been a poor approach to the problem. The users were looking for an iPad translation of the Safari service, rather than a second “good enough” way to access Safari.I’m curious if the trouble experienced by the Safari App is due to the PhoneGap framework itself, or the implementation of those tools.
Interesting problem. As we speak, I am working on a very ambitious product that has many user interfaces, including a mobile interface. I’ll be at Cloudstock/Dreamforce in SF, by the way, if you by chance are planning to be there. I made the decision to go with multiple “skins” for my mobile interface. Unless you are on Android, I will show you an interface that mimics iPhone user experience. If I detect you as an Android user, I will show you an Android UI. The interesting part is that I have no intent to spend my resources on developing these alternate UIs initially. It’s in my “nice to have” portion of the roadmap and I’d much rather implement deeper Salesforce integration if given choice.My priority is to make my site feel like a native app for your device or better. iPhone absolutely nailed UX, in my opinion, which is why it is my default interface. It may look a little weird to have iPhone UI on Android, but chances are the users will figure it out quite easily.That is where the challenge with cross-device development comes in. I am showing embedded video on mobile phones, which in practical terms today means HTML5 and realistically mostly WebKit ( http://en.wikipedia.org/wik… ). Whereas desktop browsers are still behind the curve for HTML5 adoption, that is not an issue with mobile smartphones.I am defaulting to HTML5 unless I explicitly detect something ancient and it’s some desktop browser. My logic is that someone with an ancient browser simply does not have the capability to scan a QR code (discounting the option to process QR via MMS that no one is probably insane enough to use) in the first place. Unfortunately, due to the nature of my product, I will have to support all desktop browsers. Fortunately, for those people I can simply expose an HTML3.2 form to them and call it a day as it would likely be for some regulatory compliance and online video and AJAX will not have to be part of that experience.By the same token, I don’t really need to have a native app as I integrate with the phone far deeper with native capabilities, but I will make one strictly to take advantage of the distribution platform that the App stores provide. It will be a free app that will not be very useful unless you also subscribe to my product through my website.I think that’s what Fred is hinting at by saying that the paid apps are an aberration. Apple has a definition for each of its common use cases. Games are “Immersion” as they redefine the user interface completely and grocery shopping lists are “Productivity” that take full advantage of Apple UI. Immersion apps that use phone’s hardware features will stay paid, but productivity apps will likely increasingly migrate to being HTML5 apps and out of Apple’s 70/30 reach.
Well, Nitobi, the guys who developed PhoneGap, wrote the Safari app – so it’s a pretty fair bet that they used PhoneGap as best possible.There’s a big, BIG red flag for everyone pushing cross platform, right there.Also for people espousing HTML5, since PhoneGap can be regarded as HTML 5+.
Android is in a crazy growth phase. Neilson’s research on “desired” smartphone leads me to believe Android has lots of challenges http://bit.ly/gWb5EL. Add 40% of 120M “missing” iOS iPod Touch devices and iOS is in a strong position for next year.
Why not take it a step further & just make your iOS or Android ‘Apps’ a web browser that points to a mobile website.Also, depending what you’re doing – you don’t even need HTML 5. HTML 4 (with your back end language & database of choice) can do most tasks perfectly well.All we need is our existing web development skills and some nice mobile CSS templates…
Fred, great article and I do support your points fully. In terms of apps I believe that they have to serve a certain purpose and have to do something that goes beyond what HTML5 can do. As long one provides a mobile web experience across platforms with HTML5 I think one could get away with having an app just for one platform (that might be not true for game developers). I’m working a lot with retailers and in that space a mobile app is a barrier. It keeps shoppers from shopping on impuls. The app will only work if a shopper has a strong association with the brand and interacts with it on a regular basis.
Great Post, Fred. Also worth mentioning in this post:- Sequoia invested $14MM in Sencha, which is a merger of two open source frameworks for mobile development (Ext JS and JQTouch). This in theory should offer cross platform HTML5 mobile development, through a dual-license (one for commercial, and a non-commercial open source version).- The JQuery conference was all about their forthcoming mobile library. JQuery has become practically a ubiquitous standard in Javascript libraries, and their developer adoption is massive. So, a mobile version of this excellent library will likely be strongly supported.
huge, huge, hugeHTML5 ftw
You should also know one of big banks is developing with Sencha now. If you can get the stodgy to use it….
That’s exciting. I wonder if there are compliance issues with usingthe New Hottness in a banking IT stack?Andrew
I definitely wouldn’t count out RIM, as there are some very interesting pieces of technology that will find their way from the Blackbird tablet program to a next-gen smartphone. It could be a significant leapfrog if they can gain developer mindshare. That’s a big if, though.The other concern I have about Android over the long haul is the monetization model for Google. It seems that they haven’t yet turned on the advertising firehose yet, but it has to be coming, and it remains to be seen how intrusive and/or invasive that will be.Instinctively, there also seems to be an “enterprise smartphone gap” that someone’s going to fill in a very innovative way. Even in that segment, there are two distinct markets (large ones) that haven’t had much innovation in a while (business user smartphone) or are largely untapped (industrial/severe conditions user, such as construction, military, factory, utility workers, etc.).Remember that most people spend much more time interacting with technology at work than in their personal lives, so there is clearly great opportunity in that whitespace as well.
I’m a BlackBerry fan and not counting them out either, but they have a LOT of work to do.All I can say is, it’s getting old that I have to pay a huge premium (roughly $300 a year right now) to get fully synced Microsoft Exchange mail+calendar+tasks+contacts on a device with a fully functional and beautiful qwerty keyboard.An iPhone 4k (with keyboard) would do the trick…the software rocks. But alas, Steve Jobs is a stubborn man who cares more about beauty than market share.I’ve yet to see an Android phone that can do what my BlackBerry Torch can do with e-mail. But it’s feeling quite dated in many other respects.Droid Pro might actually fill the bill. We’ll see.
RIM is hostile to developersthat is a death sentence
Very interesting article, thanks.Android is as big as iPhone in the US (although IOS devices apps market with iPod & iPad is still twice bigger). In Europe, Android is well behind in France, the UK, etc…. maybe because carrier don’t have a monopoly on iPhone as with AT&T in the US.As for my personal experience as an iPhone & Android developer, we released our FREE Christmas Advent Calendar 2010 on the 1st December.http://bit.ly/Advent2010Our Android app got 12,000 downloads in 5 days versus already more than 800,000 for iPhone in only 5 days (#1 App Store in the UK, France, Germany, etc…) . For an App developer perspective, IOS platform seems to drive much more volume than Android Market…Emmanuel CarraudCEO MagicSolverhttp://www.magicsolver.com/
Really thought provoking piece, Fred. The numbers don’t lie, Android is on the up and Apple will have trouble competing for the bottom end of the smartphone market and it’s forcing developers to rethink their assumptions. However, I’ve been hearing two basic reasons why most of the app developers we work with STILL focus on iOS first, and then Android.1. Killer apps don’t launch on Android, they arrive on it. This is something Gruber pointed out a few months back. This could change in the future, but most of our members feel you can’t get the same kind of buzz on Android for a brand new app. If you want to get influencers excited about it, you need to be on iOS.2. Tough to get paid on Android. The numbers here don’t lie either. If you’re trying to sell your app, the Android platform is just less profitable. The average spend per Android user is 75% less than the avg spend per iOS user.
tough to get paid on iOS in a year or two
The move against Kik by RIM is interesting, because I believe it is similar to AOL move to protect it’s huge Instant Messenger install base. Where is AOL Instant Messenger today? AOL IM is part of a larger ecosystem being connected together with services like Meebo.I hadn’t heard about the lawsuit until reading this blog post, but any company trying to protect it’s messaging service from competition isn’t going to last. RIM does not have a vision for it’s future, it is destine to have a much smaller market share, much like Nokia.
sadly you are right
Why is it that Apple and Google have about the same % of market share, but every US-based ecommerce site I talk to shows ~80% mobile traffic from iOS, ~10% from Android and ~10% from everything else. The only Android devices that really have a comparable experience to iOS are the N1 and newer (devices you can actually run Froyo on) which is only about 40% of the Android total.
the android surge has just startedgive these numbers a few months and i think you’ll see the effects at playwe have to skate to where the puck will benot where it is
For purely selfish reasons, I wish you would convince your portfolio companies of this. Foursquare’s Android app is much weaker than its iOS app. Hashable has led with the iPhone. Zynga has as well.Regardless of the strategic wisdom of any of this, it makes life harder for Android owners who have to write about these companies and their products. Lay down the law!
why do you think i wrote this post?
I disagree. These numbers will flip when the iPhone is available on Verizon. You cant do market analysis like this based on the assumption of non-compete due to agreements which have not yet come into play.And even if the iPhone never came to Verizon or anyone else, if your app gets no hype on iPhone its not going anywhere. The angry birds model is the one developers will be following…sorry VC guy.
i’ll revisit this post six months after iPhone has been on verizoni bet that android will be in front in market share
FW: “One thing I am sure of is that developing solely for iOS, which is a very common thing I see out there, is not the right strategy unless you only want to serve 25% of the market.”Agreed completely. But I’m interested that more folks don’t discuss the volume of apps available for a particular device. This can HURT you if you are a young startup or mobile developer. You have to have either an absolutely game-changing concept, or a known name behind your app, or good luck getting your application found among 300,000 others (iPhone) or 150,000 others (Android).If you don’t have the next huge thing, or an established company behind you, I believe developing for any one device specifically means you better have a heck of a marketing team. You need to a) convince the folks on your chosen device that they HAVE to have your app, and b) try to explain to owners of every other device in the mobile world why they can’t use your product.I eat, sleep, breathe, and bleed mobile. In my opinion, successful startups will have a concept that works perfectly, on as many devices as possible AND on desktops, on day 1 of public beta.
Although there are still some cross-device glitches today, I think eventually Android will get it right and we will see the same dynamics as PCs and Macs. A “standard” platform, lower cost and with great market share (Android), vs. a “sexier” device that survives thanks to innovation and “vertical integration”, i.e. better control of all the components of your platform to deliver a easier to use and more stable system. I think it will take longer than one might think for Android to achieve standardization (meaning a stable system across platforms) because the emergence of tablets (iPads etc.) will make the task more difficult.Right now both Apple and Google are running very fast and are eclipsing the competition. RIM’s strength is the corporate installed base but Google and Apple are already chipping away at that. If I was a developer I would focus on iOS and Android, and I would add RIM if I was developing an enterprise-focused app. For anyone else … pay me porting fees if you really want me/need me.
hi Allesandronice to see you in the comments!i think this is largely correct
One thing I am sure of is that developing solely for iOS, which is a very common thing I see out there, is not the right strategy unless you only want to serve 25% of the market.From a developer perspective, do you care what percentage the market share is of the platform you are developing for if you are likely to earn better revenue/profits on that platform compared to the majority share platform?
yes if the market will move to a different model driven by the eventual market leader
I think it is pretty simple if you are a business: make sure you have an awesome mobile web experience first. Then, provide added value via apps.I also wouldn’t downplay WP7 and Palm just yet. I just bought a WP7 phone and in terms of UI, it makes both iOS and Android look like dinosaurs. If they can start filling some feature gaps and find a decent marketing strategy, I think they could garner 10-15% market share in the next year.
If you account for the iPad & the iPod Touch, and allow that some of these Android devices are lower-end devices that people will get because they were free, and aren’t exactly cutting edge, iOS is a much better developer platform. At least, if your goal is to sell apps to make money.Writing an app for iOS gets you iPhones, iPods touch, and a headstart on the iPad. Writing an android app means… what? You have to work harder to get it to work on multiple Android devices and maybe no one will buy it?Surely, at some point total Android smartphone marketshare will be higher than the iPhone but it doesn’t really matter when iOS is a platform that started on the iPhone and has already moved to 2 other devices. There is still no iPod Touch like device that runs Android.
Personally, I think until you’ve got at least a minimum viable product working on a web-based platform )optimized for mobile) you shouldn’t even be picking a platform. Then, and only then, should you wade into a proprietary platform if a) your users are demanding it AND b) there’s a critical/key feature you just can’t accomplish within the limits of the web-based app. But going proprietary should be an absolute last resort. Especially for a start-up.As the market fragments across platforms anything that’s built for only one platform instantly starts at a disadvantage when it comes to getting critical mass, especially if the app includes a social component.
don’t forget that the Windows Mobile numbers don’t reflect the new Phone7 platform. It’s great to see how strongly Microsoft has got back in the game, with free developer tools (that don’t require a Mac or a degree in rocket science) and a simple marketplace…. and a ready made pool of .Net / Silverlight developers who already know how to build for the platform (and don’t forget the dedicated XNA game capabilities)No-one has been buying WM6.x phones in the last couple of months waiting for WP7 to ship… now it’s here (and riding on the halo of Kinect as well) it’s going to be interesting to see how the results look in 3 months time
Why focus on WHO WINS? The data clearly shows fragmentation. Microsoft, Nokia, and HP will add to the fragmentation. Even Android is not one platform.It’ll take a dozen year for the market to sort down to two winners – and only when the pace of change slows. When the mobile ecosystem changes weekly, why focus on the tusk of the elephant?Competitive markets are unpredictable. Who would guess that Microsoft would displace IBM? And Google displaces Microsoft? And Apple and Facebook are threatening Google? Can anyone name the database companies that dominated mainframes? The future is unpredictable even by the best opinion leaders.App developers need to diversify their risks.
Most companies should continue with “iOS first”. Unit share is important, but so are mind share, usage share and dollar share.
As my CTO pointed out- “By the way, the chart is misleading. For non-engineers and folks who’ve only owned one Android phone, they don’t realize that under the label of Google, there really should be subsections named: Verizon Android, Sprint Android, AT&T Android, T-Mobile Android, and Miscellaneous Android. Unlike the iPhone, which is one platform, Android is actually multiple platforms (even within one provider). So a lot of product managers will mistakenly think iPhone is one development endeavor and Android is a second; where in fact, iPhone is one development endeavor, and Android is N development endeavors.”All the whining by developers is moot. Granted, there is a Lowest Common Denominator subset of Android features, so that when developing for the iPhone, we also get much of the Android LCD for free because we are using an app in development that bridges the gap between iPhone and Android.I think your basic statement about the reasons why Android will continue to gain market share are valid and must be taken into consideration when developing for the mobile space.Granted, there is a Lowest Common Denominator subset of Android features
Yes, but there are two issues dogging Android today: fragmentation and monetization.If these are not real issues, then Google needs to market away the perception by highlighting developers who made fortunes off Android. Google needs to create Android Donald Trumps.However, the fragmentation issue remains a concern to me … how will Android avoid Java’s fate (i.e., wildly pervasive but subpar user experience and reach due to differences across devices)?I’m also surprised Google isn’t pushing more of an HTML5/Javascript approach, where the Android/Java coding acts more like a shell to access underlying phone features (e.g., camera, contacts), but the core app logic and appearance resides in HTML5. The way I think Android prevails is to offer an HTML5-centric platform wrapped inside this device integration layer (which is also more closely aligned with Google’s vision of a web-based world).Yes, mobile economics == web economics, but this also means people will pay. Just not for commodity sites and services. Today, people do pay for web sites like Salesforce, Pandora, and Dropbox, but these are properly considered services with the web as the main interface. Services that emerge with mobile as the main interface can continue to monetize via “paid apps.”
but they all have a freemium modelyou start with a free product
An alternative to target multiple mobile platforms is to use a technology that binds to native api’s, so that you get a native application, rather than the write once, run anywhere school of thought. With the former approach you can reuse non-ui code such as the object model, networking, xml/json parsing and web services, all of which can be substantial. You still end up rewriting the ui with such an approach, but at the end of the day you have native apps that don’t feel out of place in user experience, capabilities or performance. For example you could use MonoTouch, MonoDroid and the native WP7 .Net stacks and target iOS, Android and WP7 respectively. (Disclosure: I’m writing a book on MonoTouch).
I would consider Apple over Android if their customers are more of the paying types. If low price is what is driving android sales, they these customers may not pay to get an app.
One big problem is the fact that a lot of Android apps are hacked and shared, whereas Apple is very aggressive against this behavior. Therefore it is not only market share but size of actual paid market that is more important. In this Apple still rules
Fred, I don’t understand how you can compare iOS to Android when you only count iPhone numbers in your stats? Adding just the iPod Touch to Apple’s numbers and the iOS share almost doubles. Add in the iPad and iOS dwarfs Android.Developers want to know the total size of the platform capable of running their software. To only give numbers of a small segment of the platform which then casts that platform in a negative light compared to the opposition smacks of dishonesty or a hidden agenda.With Google admitting with the release of Gngerbread that Android growth has now flattened out at around 216,000 activations a day (1.5 million per week), iOS is growing at the far faster rate of 275,000 unit sales per day (almost 2 million per week).With an installed base of 157 million, iOS continues to draw away from Android which sits down around the 30 million mark.With far more income ($1 billion) paid out to iOS developers compared to the $21 million paid to Android devs over a similar timeframe, Larva Labs states that “This really indicates how much of a cottage industry the paid Android Market remains, with insufficient sales numbers to warrant full-time labor for paid content”-Mart
I should note that Google’s Android figures include tablets like the Dell Streak and Samsung Galaxy Tab, so one wonders why commentators continue to ignore the other elephants in the room – the iPod touch and iPad – when reporting on the size of the iOS platform?Developers want to know total size of their potential market if they are going to make educated decisions.-Mart
(edited)
As others have pointed out, the numbers you show were compiled leading up to the Windows Phone 7 launch. All you have to do is follow the #WP7 hash on twitter to find out what an impact it is making.I think on major thing that is not being discussed on this thread (at least not to any great extent), is what fragmentation does to a device. As a developer, it is difficult to maintain code bases across many different devices and OSs. Windows phone saw this first hand for versions 6.5 and below (which is what the above numbers represent). Android is starting to see what happens when you have many different devices and OSs out in the market at the same time. Frustration from developers is the first thing that happens, as time goes on, users start to see the downside when the phone they are using cannot run the cool application that their friend just downloaded. Microsoft has learned from the past and delivered a great new smartphone with some very unique features and a development platform that is easy to use.
i like your blog. Your comments are indicative of someone paying attention ;-). My take on the iOS vs Android is this: iOS is closed. You cannot (without significant effort) run emacs on iOS. You’re not supposed to run JVM on iOS. What made the Apple recovery possible was the openness of Mac OS X. It was Unix-based and it was full of open source. This won the hearts and minds of developers who were not in the immediate Apple family. Since the recovery we see the development of the iPhone, and we see a move away from this openness. It is at least plausible that on a phone (where reliability may be an issue of life and death or other mission-critical factors) more control over the OS is a reasonable stance. Further, when we look at the lead iPhone has over other SmartPhones/OS’s it’s likely to be a tenable stance for some time in the future. However, when we look at the touchpad applications things change. There is no mission-critical argument. These are leisure computing devices even more so than laptops. Hence freedom of development trumps any sort of control arguments.When we look at this in the context of what’s happening with the development communities we see that Apple is moving behind the curve. The developer world is moving towards functional languages as a palatable solution to deal with the concurrency requirements from above (Apps must be web and mobile) and below (hardware is multi-core-per-die chips bundled into machines that are bundled into racks…). What is enabling this move for developers is the ubiquity of robust VMs (JVM, .net, LLVM). The iOS is taking a step back from these enablers which means it is taking a step back from the platforms that developers are moving to in droves. That’s very dangerous for the class of computing devices Apple is spearheading. That’s where i think the Android OS has the most opportunity. Android on a larger touch device allows for a whole new world of UI and application design and supports the new world of development techniques that are emerging to meet this opportunity.i have been so impressed by Apple. They have amazing design sensibilities. They get computing-as-lifestyle better than any other company. It would be very sad for them to have to play catch up in this space. They’re much better as leaders than followers. Control, however, has always been their Achilles’ heel. Part of being a truly great leader is enabling others to do great things and one of the most important ways of doing that is allowing people to work the way they want to, including using the tools that give them leverage.
We’re seeing as much Android interest as the other platforms (iOS, BlackBerry) and maybe (not) surprisingly Windows Phone 7. At first glance, Windows Phone 7 is showing promise with great devices and a development environment rivaling the best tools from Apple.RIM is not out by a long shot, and has great support and help for dev shops like ours. Windows Phone will be huge (and no one writes dev tools like MSFT), and will have a great dev. ecosystem. I think Google will get their act together on the dev side (better and better support for app devs) and Apple will ride it out :)Farhanwww.xtremelabs.com
Great analysis. Precisely what we’re implementing towards with the PhoneGap project. Cheers – brian
Blackberry may need this – run Blackberry OS as a VM under/alongside Android – press a button and switch from one to the otherhttp://www.networkworld.com…interesting case study, whether Blackberry should roll out as a sandboxed app on other platforms, fight migration or help switch.
Parallels for android?
I have been inspired by at & t mobile applications and models,i just now started learning j2me,Java 2 mobile edition..I have read in some best articles by ceo’s,that java is the best and powerful application which can be used for developing mobile applications..Now a days android is striking the market in a good way..awaiting to see some more company’s coming up with mobile application..ecommerce web developer ecommerce web development
Fred, I don’t think this is such a cut and dry strategy based on sheer volume of handsets with a specific OS. I recently posted a pretty detailed comparison of both the iOS vs. Android app stores on quora:http://qr.ae/7LcrI think which platform you target is largely dependent on your frame of reference. If you are a more traditional web-based service that sees mobile as a natural extension of your online properties then HTML5 first makes a ton of sense. If you’re taking a mobile-first strategy of distribution and rare doing paid apps or freemium then Apple is still really the best place to prove your product and gain initial traction. If you’re doing free ad-supported business then Android works just as well though you’ll face a host of other distribution and development challenges.- Ish Harshawat
Totally totally North American centric figure and article and advice. Everybody in North America is obsessed that they are the center of the universe, you’re not and neither are North American sales figures representative of larger market demographics.
True Android is starting to reach the critical mass (clearly) for various reasons and may have more market share than Apple in 2011, but there are several hurdles from a developer standpoint.1. the app storeThe biggest issue with Android is the app store and how it’s near impossible to filter quality from quantity. This piece on the NYTimes beautifully summarizes why Apple will remain dominant in the app space next year, too: http://goo.gl/D01Vu2. target marketit is near impossible to target the tech savvy/early adopters, since most -if not all- have iPhones/iPads. These are the constituents for beta testing and MVP app marketing.3. product marketinghow does one market to the critical mass, that generally use their device for Facebook and texting?With tech, product is always superior but because of those three points, I still say iPhone first, Android second, then perhaps HTML5.I also firmly believe entrepreneurs should spend dollars on basic business courses. It blows my mind how 1. so many brilliant engineers fail to deliver products answering the keys to fundamental success factors; the who, what, when, where and whys. and 2. have horrific biz dev skills.So where should mobile developers focus their funds? iPhone development and personal development courses since successful mobile products will remain reliant on word of mouth marketing through 2011 and perhaps 2012. Or until Google betters the app store UIX.ps: one more link re: Angry Bird’s and Rovio’s success http://www.mediabistro.com/…
While HTML5 is terrific and has tremendous potential – is nobody concerned about protection of intellectual property? While reverse engineering is a risk anywhere – at least with native you can make it really difficult and cumbersome to try and get the core. With HTML5 you’re sending all of your IP along with the application as javascript which at best you can obfuscate a bit.What are people’s thoughts?
Don’t forget WAC/JIL – http://www.jil.org which is the Industry backed “write once, run many” web app library that was initially started by Vodafone, Verizon, CCMC, and Softbank, but has just emerged from a merger with WAC which is industry standards body made up of 20+ mobile carriers around the world. The plan is to deploy the JIL runtime on the handsets, and run their own Appstores. JIL runtimes exist for Android, RIM, Brew, and Symbian.Will it gain any traction in the marketplace? Currently WAC/JIL only support HTML4/CSS2, but you can bet its not going to stay that way.. However, we have seen this behavior before with the music industries’ creation of SDMI. Seldom do this industry backed consortiums ever gain traction.. But this time, they might dig their heals in.. Depends on how much they value the growing App market and position in the App distribution value-chain. With webapps.. the idea of allowing Pre-paid phones and lower income demographics access to webapps that run as native apps at a developer or OpCo set price .. (think ‘QuarterApps’).Disclosure: I was the JIL Developer Support Lead @ Vodafone from 6/2010 – 10/2010 and was responsible for architecting the latest Developer Application Submission wizard allowing for scalability in required metadata fields for 20+ carriers in 100+ markets.
I got my new phone a few weeks ago and it’s an a Droid. I like it a lot, I have the capability to do anything from that phone. I will interested to see what the Droid tablets do to the market as well.
Things are moving so fast! I wonder if the question does one do HTML 5 or native… is still valid. The questions should a startup write their own blogging software has long since been answered.If the app is NOT the core of their business, then they should consider looking at one of the many “mobile publishing” platforms popping up.I am the co-founder of one such platform ( http://goodclic.com ). We spend our whole day (and most nights) thinking about what features our clients need next. What is the next hot phone? What does this version of Android do to the look of the app? Will the next windows phone browser support webkit (a necessity if you are going to do cool HTML5)… then there is iPad… Apple TV (soon to do apps)… GoogleTV…We make sure you can connect to any social network, any video platform, sell tickets to your event, and set up an mobile store to sell your t-shirt at that event. We think if your company is not the mobile app itself, then it is silly to spend time thinking about it. And, if we do our jobs right we hope you won’t have to!We want to be wordpress for mobile and let our clients focus on their differentiators. (I hope it works!)
Agree with you. Of course the one caveat is that the browser you download doesn’t come with a two year contract. 🙂