Views on the new media market

Mobile application - sure, how do I get it?

Mobile applications and application stores are all the rage in mobile chattering classes at the moment, but it's worthwhile stepping back and looking at mobile through the eyes of a pragmatic, mass market consumer.

Sure, they have a mobile phone indeed surveys show that if forced to pick only one of TV, Computer or mobile phone 58% would pick the mobile phone.

So there should be a great market for mobile application producers, shouldn't there?

If you talk to mobile application developers there is one word that you'll hear again and again - fragmentation. The wide variety of different platforms, operating systems, programming runtimes, screen sizes, keyboard layouts and phone features, stores, certification programs and other differences make it expensive to produce even the simplest application on mobile. There are ways around these fragmentation problems, provided that developers are prepared to spend some time and money on solutions, so it's possible to overcome the problems.

Which leaves one problem remaining - how do you tell consumers about your mobile application, and what happens when they hear about it?

If we take a group of just 10 people in the UK market, discussing the latest and greatest mobile application we'll find:

Two of them have iPhones, two have Blackberries, four have Nokia smartphones, one has a Windows Mobile phone and one has an Android phone. Pretty tech savvy lot - especially as the other half of the room have non smart phones.

They are swapping stories about the latest application that they have discovered, and recommending it to their friends - great viral marketing. The problem is - how do they each get the application? Is it listed in the application stores on the devices? Which store? How about they search for it? Will Google find it? How does it then link to the application?

Even for an application that is available on all the platforms the chances that these consumers will quickly succeed in finding the application is small - there is no clear way to get applications for mobile phones that doesn't first require answering questions like "what phone model? what operator?"

This problem repeats itself when you look at any form of mass consumer advertising or marketing - the probability that you'll successfully convert customers for a mobile application is very low.

So what is likely to be the best way of marketing a mobile application?

How about marketing a mobile web site? If you can educate people to go to a mobile site first - then the application provider can provide the logic to navigate to the appropriate application store for a device. Thus, marketing your mobile application becomes a question of how would you market your website - mobile, or desktop.

Friction in mobile application provision

Firmly believe that "applications" will win on mobile devices - an application that can take up the full screen of a device, and has an instant response to starting the application is more convenient for a mobile phone - and a browser can't provide this as easily as an application can.

However browser based solutions provide far fewer points of friction to a developer who wants to make functionality available on a handset.

Developers frequently complain about fragmentation being a cause of problems in delivering mobile applications. This is a natural view, as developers spend most of their time dealing with issues caused by different platforms, languages and other technical features of phones. Yet fragmentation is only one form of friction for mobile application developers. Other forms of friction in getting an application out are:

• Multiple application stores - to reach a consumer in the UK it's likely you will have to support 10s of different application stores, and keep versions up to date across all of these stores
• Marketing differences between different stores - there is no consistency between the image formats or screen sizes required for screen shots for the same application on different stores - so even with one app there will be multiple versions of marketing assets required
• Approval times for different app stores vary
• Digital signature processes differ for different application stores, so application authors need to maintain multiple identities to support this
• Testing and QA processes differ for different stores, so application authors need to manage multiple different QA processes

All of these elements get in the way of an application developer getting their solution to market on time, and on budget, and being able to spend money on creating new features, rather than managing the distribution of the application.

Hardly surprising that the people who are making money in mobile applications are the large aggregators who can generate economies of scale by managing these solutions for multiple applications with multiple application stores.

The economics of a mobile web application are quite different however. While fragmentation in the mobile web space is just as bad - or arguably even worse than in the mobile application space there are fewer points of friction.

• A single web URL can be made to work for all mobile phones. Try m.facebook.com
• Consumers expect either direct marketing, or search engine marketing for web sites - on mobile as well as on desktops
• Few external approvals are required to launch a web application - there may be some if operator specific APIs are integrated with (for example for online mobile payments)

The result - it's quicker and easier to reach mobile devices from a web solution than with mobile applications. Apple's App store managed to significantly reduce the points of friction for an application developer bringing their app to market - and saw huge success as a result compared to previous application stores. The challenge for other application stores is - do they bring similar reductions in friction.

Apple's App Store careful where you tread

Apple's App store continues it's awesome growth, with headline figures of 2bn downloads.

The rush to create new applications and get them onto the market continues, as does the rush of other platform providers to copy the success of Apple.

It's easy to extrapolate from the very large numbers being posted to the app store to assume that there are lots of developers making vast amounts of money from the App Store, and that Apple are in turn making vast amounts of money off it.

The truth might be somewhat different though - talking to developers who have applications on Apple's App store you get a somewhat different story - most people seem to be making a small amount of money - better than previous application store offerings, but not enough to build significant business on (by contrast the web, has a reasonable number of mid-size businesses, and a reasonable number of large businesses that sustain themselves solely by revenues generated from the Web)

So where is the gap? Is it that a very small number of iPhone application developers are making all the money, or do we need to look a little deeper at the App Store figures?

If we look at the headline number from Apple - 2bn downloads things look great.

However - how is this number counted? It's a total of downloads from the App Store. What might make up a download?

1. New, paid app that is bought, then downloaded
2. Free application that is downloaded for the first time
3. Updates to applications that have been previously downloaded automatically downloaded by iTunes

Application updates frequently get forgotten - these are a very important part of an application lifecycle - they enable application authors to incrementally add features in response to user needs, and to update to fix bugs - the App Store handles these particularly well - not quite as well as the web (where application updates happen server side) but better than any other App Store.

So how do Application updates work? The app author submits a new revision of an application, and it is approved by Apple's reviewers, and listed as a free download. Apple application review sites then typically pick up on updates, and add new reviews, and iTunes automatically tracks updated applications and downloads them.

The net effect? Application authors get a sales bump from submitting new versions. So some applications have very rapid version changes to maximise sales - some applications have update frequencies as often as once a week, once a month isn't uncommon, and it's reasonable to assume that on average many applications have an update once per quarter.

Each of these updates drives another download - to all the people who have previously downloaded the application - even if that application is not installed on their iPhone - iTunes will download the update to the host PC or Mac anyway.

If we believed the headline numbers from Apple we'd see

• On average the 50m iPhones + iPod Touch devices have 40 downloaded applications installed

This isn't true - though on average the 50 million host PC or mac devices have downloaded 40 application packages

If we assume that each application has had 4 updates during the year we can now assume that there have been 500M unique new app downloads to devices - 10 per device. If 80% are free downloads compared to paid - then that's 100M paid application downloads. With an average revenue of around $2.50 per application that's $250M revenue flowing through the app store system - $175m being paid out to application developers, or, if there are 40,000 paid applications, an average of $4,375 per year application launched. Some hits will be higher, but many will be lower than this. If an iPhone developer wants to earn $50,000 per year then she needs to be creating a new app every month.