|iPhone? Android? Blackberry? or is it the new Windows Mobile 7?Over the last couple of months, I’ve been urged by a couple of people to write a “Tutorial for mobile app developers”. They figure that as I’ve been doing this since ‘99 I must know something that’s of interest to some other people. Now I’m generally a little too lazy to take on a project like that, so I figured I maybe take it on in installments, kind of one chapter at a time. But I didn’t really get around to it until today, when after moderating the Mobile Track at the Eye For Travel Distribution Summit I realized the audience still has some pretty big questions and promised I’ll do my best to answer them, so here goes.
Still wondering about an App vs. a Mobile website? Wait for the next article.
Frankly, that topic requires a discussion on its own, but that’s not what I want to talk about here. So generally speaking I should say that if you want persistence – whether it’s of information on the device (e.g. such that it’s available offline, or immediately available when the app is accessed) or just of your basic brand presence and UI, or if you want a best-of-breed user experience, an application is the way to go, at least at this stage. If you’re still not sure – you’ll have to wait for a subsequent article.
Mobile Platforms – What’s Out There?
Unfortunately – much too much. Speaking strictly about Smartphone OS – In North America, there are three dominant platforms and one potential – Android, BlackBerry, iPhone (iOS) and Windows Phone 7. Elsewhere in the world, Nokia’s Symbian is highly prevalent, and MeeGo may replace it soon. If we extend our view to feature phones, we have to consider J2ME, BREW, Samsung’s bada and DoCoMo’s iAppli. I will not discuss those latter platforms at this time.
A few words about each OS (listed Alphabetically to avoid misleading prioritization):
- Android – Google’s mobile Linux variant. Provided for free to many device manufacturers and heavily promoted by Verizon to counter the iPhone. Currently the fastest growing market. Supports a variety of device form factors and screen resolutions
- BlackBerry OS – RIM’s proprietary operating system. Drives all BlackBerries, but not the upcoming PlayBook tablet. BlackBerry is the leading platform in the US from a market share perspective. Even though there’s one brand, there are many OS versions, device form factors and screen sizes in the market.
- iOS (iPhone OS) – Apple’s proprietary OS. Harbinger of the App revolution and leads the App economy in many ways. Apple always has only one device model in production (currently iPhone 4) and there is great uniformity regarding form factor and screen size.
- Symbian – Nokia’s long-time mobile OS, once licensed (and co-owned) by many other manufacturers but since then dropped by all of them. Mostly drives mid-market phones in the non-US market nowadays, with dozens of form-factor / screen size combinations.
- Windows Phone 7 – Microsoft’s latest attempt to reclaim the mobile market, lanched last week. Openly licensed to many manufacturers (most of them also sell Android phones). This is a brand new platform, not really backward compatible with previous Windows Mobile OS. This time, Microsoft dictates a pretty uniform form factor.
Remark: At this point in time I ignore Palm’s Web OS – which is sort of “dormant” in the market right now.
Gartner presented this chart a few month ago:
Share of 2010 Q2 smartphone sales to end users by operating system, according to Gartner.
How The Platforms Differ in The Marketplace
While there are some meaningful technical differences, the most important ones from a business perspective are the different audiences they reach – geographically, demographically and psychographically. Here are the main point for each:
Android: Android is trying to be for Smartphones what MS-DOS and later Windows are to PCs (see for instance what Fred Wilson writes about it). This results in a platform that reaches a very varied demographic / psycho-graphic, with a world-wide distribution. While this is good for Google, for you it may mean that much of the audience is composed of people who are inclined to use their Smartphones as feature-phones – not necessarily App-savvy / quick to spend on Apps. This is why many developers who develop both on iPhone and Android, for instance, claim that iPhone Apps monetize an order of magnitude better.
BlackBerry: RIM is traditionally very strong in enterprises – as a tool furnished to executives and managers as well as many other professionals. This represents an upscale demographic. However, as many of these people see the BlackBerry as a tool provided by their employer, and also because of its inferior App search / download / install capabilities compared to iPhone for instance – the propensity to download Apps and especially to buy them is lower. Lately, RIM has successfully penetrated the consumer market in many countries, reaching new audiences – not as affluent, and not necessarily focused on the “Smartphone” capablities but more on the pure communication / messaging capabilities of the platform.
iPhone: Riding on the coat-tails of the iPod and Apple’s brand, the iPhone reached the US “creative class” first. Coupling this early adopter / high-income audience with a great App download / purchase / install user experience, iPhone users download more apps then any other platform’s so far. However with the iPhone moving further down the adoption graph and wider geographically, the demographics today are more varied.
Symbian: As shown above, Symbian has the largest worldwide footprint – with North America taking a very small part in that. In actuality, much of the Symbian volume goes to mid-market phones used much less as smartphones than the other platforms, resulting in a varied, dispersed audience with (on average) a mid-market demographic. Couple that with the relatively weak experience of its Ovi Store, and you have a platform where the total use of applications is probably no bigger than the other platforms’ (with smaller volumes), dispersed worldwide (from the UK to China) and much less willing to spend on paid apps.
Windows Phone: A brand new entrant to the game, it remains to be seen what audience it will draw and where. An informed guess would suggest it will be similar to Android’s, as we’re looking at the same set of manufacturers more or less, and the same target markets.
What’s Hard, What’s Easy
Without diving too deep into techno-speak, there are two main factors affecting how much it costs to develop and maintain an App on each platform:
- How easy it is to develop software on the platform – which depends on the language used, the quality of the development tool suite, the richness of the “canned software” provided as libraries with the OS, and the idiosyncrasies of the platform
- How uniform is the target phone base – for instance iPhones all have the same screen resolution, vs. Symbian devices that exist in at least 6 resolutions and 3-4 form factors (actually iPhone 4 introduced a new one but it’s still backward compatible and the same aspect ratio)
Some of my favorite developers like to use the following rule-of-thumb:
What costs 1 man-month on iPhone, will cost 1.5 man-months on Android, 2 on BlackBerry and 4 on Symbian
… Windows Phone is new, but as it’s Silverlight based and has a uniform screen / device layout, I guess it will be somewhere between iPhone and BlackBerry – once some developers develop the skill set.Relative ease / difficulty of development defines your fixed cost. But how hard it will be for you to penetrate the market effectively will affect more important variables – your actual success, or at least your variable cost (if you’re going to pay for marketing). This makes the market dynamics for each an even more important factor.
One area where they differ is the hurdles you need to jump in order to distribute your app. On the iPhone, an application has to be certified by Apple, which will take awhile to test it and may reject it (sometimes without even notifying you) for a number of reasons, the chief ones being objectionable content, not enough real functionality, or business conflict with Apple’s goals. On other platforms while there are submissions processes to the store, in actuality there’s very little filtering. In most cases you can also directly distribute (via a mobile web download) without ever going through the store. While this is a marked difference between the platforms – unless your application is in direct conflict with something Apple is doing, it’s not really a significant point.
More significant are the competitive environment, and how the distribution channel (mainly the App Store / World / Market) is managed.
The Competitive Environment for Apps: Circa 2010
iPhone claims 200-250,000 Apps in the market. While this shows that some people have been successful there (drawing a lot of other who haven’t), it also means that almost every idea is already represented there, often with mature products of high quality. The sheer volume of apps also means that yours will be buried down a database where serendipitous discovery of it is not very likely – kind of like a new website on the Internet. If you are a brand that people will search for anyway, or if you have an established audience you can direct to your App, you are likely to acquire users easily. But otherwise you will need to resort to any of the key digital marketing methods known on the web – advertising, PR, viral marketing etc., and your success, at least initially, will be predicated on these.
I say “initially” because the App Store has a memory, and the best way to draw downloads is by being ranked in the “Top 25” charts. This constitutes some of the most valuable exposure to end users. If you can break it into the Top 25 in a category (or better – overall), then you can remain there for awhile with limited further investment. And if your App and / or Brand are very good – then only meaningful competition can unseat you. However, do note that this is not easy. One of the reasons is that the iTunes App Store takes into account two main factors – Velocity (how fast has your app been moving in the last 24 hrs more or less) and Total Downloads. So incumbents have an advantage – unless the new App shows people are crazy for it.
Taking all this into account, one would argue that it may make sense to first try in another, more virgin territory, where the competition is not as fierce, and the incumbents are not as strongly established. With Android claiming about 60,000 apps, and BlackBerry a fraction of that, you theoretically have a better chance to compete. That is right, but you also need to consider the target audience and market dynamics. Android users tend to buy much less. This is due to a combination of factors including iTunes being more effectively hooked to a payment mechanism (all iTunes users must input their credit card details and payment is a one-click process) and the iPhone audience being more conditioned / more open to paying. So if that’s your business model, the returns may not be as good. BlackBerry users download less. However those who do, are willing to pay more. Whether or not that offsets the lower downloads, will depend on your specific case, and the overall product and marketing mix. With Symbian the distribution game is more complex – with the Ovi Store being just one channel to the market, and alternative stores (e.g. GetJar) and direct downloads being in many cases an important channel too. However in general the propensity to pay for Symbian Apps is much lower than on the other platforms, in my experience (and WorldMate has millions of pretty up-market Symbian users…).
So how to factor all of this into a decision? In my next post I will try to give some examples of how such a decision should be arrived at.
... Questions? request? post your comment below and I’ll try to address it in the next installment!