Categories
Business

The Six Biggest Challenges for App Businesses (and what to do about them)

In our Developer Economics 2012 survey, we asked developers about their biggest challenges. Here we discuss six of them, with some basic tips on what to do about them. The challenges are split between marketing and post-launch app and user management. The three biggest marketing challenges were: keeping users engaged, targeting the right users and identifying the right revenue model. The three biggest post-launch challenges were: Tracking bugs and errors, getting users to review your app and updating applications in the field.

Keeping users engaged

Keeping users engaged was the challenge cited most often overall, by 39% of developers, irrespective of primary platform. This is consistent with data from analytics firm Flurry, who report that user engagement falls sharply over time, with only 24% of consumers continuing to use an app after three months from download. “Developers must focus on tracking user engagement & usage patterns rather than just on downloads” notes Jai Jaisimha, founder of Open Mobile Solutions, a brand-to-developer matchmaking service.

There are many techniques for improving user engagement and retention. Social buttons like Follow or Like, especially when integrated with social networks are known to increase engagement. “‘Follow is the most common social feature used by our users” notes Yiannis Varelas, co-founder of Weendy, a weather app for surfers, with 6,500 monthly active users and 80% retention rate (in May last year). Where direct social integration doesn’t make sense, push notifications are another tool to help keep users engaged.

Gamification is another retention technique that rewards users for achievements (e.g. FourSquare-style badges) or for inviting other users (e.g. for each user you invite to Dropbox, you get another 250MB free storage space). Moreover, Tom Hume, founder of Future Platforms, argues that developers need to fundamentally rethink user retention. “To improve retention, developers need to build up value for the user that increases with usage. A natural way to do this is to build in a history of usage data – for example in the Nike Plus the value and stickiness of the application increases as more data is recorded in the application”.

Targeting the right users

The second most oft-cited challenge is targeting and getting through to the right users – mostly because existing app stores offer little in the way of user targeting. App stores, for example, provide no means for developers to reach existing customers or gain information about them. The only way developers can target users via app stores is via coarse-grained methods based on app categorisation or keyword selection.

Consequently, we found that developers using app stores are more concerned about targeting (39%) and engagement (46%) than developers using most other distribution channels. The situation in carrier portals is even worse: around 55% of developers using them are challenged by targeting and engagement.
Customer information, as with any business, is a key source of competitive advantage. As such, app stores have little incentive to share customer data with app developers. Apple has done so in part, after considerable pressure, but only to Newsstand publishers, and only where customers opt in. There’s more to it than just control: app store owners are loath to jeopardise user privacy contracts, lest their platforms become marketing “wild wests.”

The inaccessibility of customer information will likely remain a thorny issue, and one that hampers developers’ marketing potential. For the moment, it generates a flurry of innovation, as evidenced by the proliferation of in-app and external app marketing channels. However, seeing as this will remain a pain point, there may be opportunities for app stores to differentiate, if they manage to balance their priorities against those of developers – as Apple arguably has with Newsstand publishers. In the meantime, app developers would be wise not to wait for the app stores to fix this. Try to reach the right users for your app wherever they are currently, via blogs, forums and more traditional media. Arrange cross-promotions with similar but complementary apps. Experiment with alternate discovery solutions and find out what works for your app.

Identifying the right revenue model

Developers were becoming increasingly confused (36%) about which revenue model to use. There are over 10 revenue models to choose from and no guarantees as to which revenue model will work best in the long run in terms of reach vs. monetisation. Moreover, the revenue model needs to be optimised to the platform and app category. The decision should also take into account factors such as customer paying propensity (which varies across platforms), competitor pricing and positioning (which varies by app category). For example, paid downloads are extremely unpopular on Android, whilst apps aimed at children often need to use that model, since parents are very uncomfortable with in-app purchase or advertising based models for apps their children are using. User needs should also come into perspective when considering
your pricing strategy. “You may only need one Facebook, sports or weather app, but you will want to play many games. Mobile games are like movies – users are always looking for the latest one,” notes Markus Kassulke, CEO at Germany-based HandyGames.

Overall, we found that pay-per download was the revenue model used most frequently, by 34% of developers irrespective of platform, followed closely by advertising, which was used by 33% of developers. Wherever possible, the best advice we can give for now is to try some sort of freemium or virtual goods model using in-app purchases. The growth of in-app purchase revenues across iOS and Android is significantly outpacing paid downloads.

Tracking bugs and errors

Tracking bugs and errors was, by far, the most frequent post-launch headache, as reported by 38% of developers in our survey – and particularly so for WP7 developers. There is no direct feedback channel between users and developers, and no out-of-box
means to monitor the performance of an app. App reviews work and feel more like post-mortems, rather than a live feedback tool. As a result, developers will often find out what’s wrong with their app too late, through users’ negative feedback. “Our biggest headache after launch is the lack of a two-way communication channel with our users” notes Hong Wu, Director of Android Engineering at Peel, makers of a personalised TV guide app.

The first line of defence here is to remove as many errors as possible before launch, both through good engineering practices during development and extensive beta testing. The second line of defense comes in the form of crash analytics and bug tracking services. These services track app errors by monitoring crashes and reporting the type of error, platform, device and environmental variables like location, time and transaction flow. As such, they can provide useful insights, helping find and fix errors before they drive users away.

Updating applications in the field

Updating apps was highlighted as a challenge by 25% of developers irrespective of platform. Interestingly, the difference in the update process between iOS and Android has no impact on developers’ attitudes – as both iOS and Android have their own update challenges. On iOS the process requires full certification and approval by Apple, plus explicit opt-in by the user. On Android, the update process can be automatic and near-instantaneous. This however requires that users opt-in for automatic updates for specific applications. In effect, these challenges with the update process on both iOS and Android increase the average application “age” and escalate both code maintenance and customer support costs for developers.

One solution to this is to have the app check for the availability of a newer version at launch. Although it may not be possible to have the application download the update, it could prompt the user to do so. Another option here is to track application versions via analytics and send push notifications to users with sufficiently old versions, highlighting the benefits of updating to the latest version.

Getting users to review apps

Last but not least, another frequent post-launch challenge was getting users to review apps, reported by 30% of developers irrespective of platform. At the same time, there have been some success stories of apps boosting their review numbers, usually by nagging users after they have used the app for some time. For example, to solicit reviews, DrawSomething shows a motivating alert where “Rate 5 stars!” and “Remind me later” are the only two options, wrapped in a friendly pop-up box. The example shows that the runaway success of DrawSomething was more science than luck. However, DrawSomething’s grossing ranking was declining in the run-up to the all-important Christmas sales season, showing that even successful and well funded apps with highly social components can struggle with our first challenge – keeping users engaged in the face of all the other shiny new offerings in the app stores.

Categories
Platforms

Our 2012 Platform Shootout and Implications for 2013

In our 2012 Developer Economics report we included a developer sentiment barometer for the various mobile platforms. As we move into 2013 several new platforms are on their way but all of them look very similar to existing platforms from a developer perspective. We can use this existing data to predict how developers will perceive the new platforms.

Here’s the key developer sentiment table:
Developer-Sentiment-2012

Not all factors are equal in generating developer activity

The iOS scores show what really matters to developers in a relatively healthy ecosystem. While  learning curve and development cost are important pain points for iOS developers (here developer sentiment echoes the difficulty of Objective C and the cost of buying a Mac to develop for iOS, respectively) and ease of coding and prototyping is poor, there is good access to APIs and reach and revenues are great. As long as those factors stay positive we can expect the iOS ecosystem to continue to thrive.

Android had healthy developer activity in early 2012, mostly due to great growth of the platform and developers’ belief in its future. The good API access has only improved and Google & partners took steps that have significantly improved the revenue generation issues. Android v4, launched in 2012, was the first version to ask users for a credit card on sign-up and the Android Market was overhauled and then relaunched as Google Play. Android looks set to continue to grow developer activity in 2013

Windows Phone lies at the other end of the spectrum. Developers are very happy with the ease of coding and prototyping, cost of development & learning curve, whilst access to features was only average. Unfortunately, although developers enjoy building apps for the platform, reach is bad and revenues are poor. There don’t appear to be any major issues with the Windows Phone Marketplace compared to rival stores and competition for user attention is lower, so we can assume that revenue issues are primarily due to the lack of reach. Although the introduction of Windows Phone 8 has improved access to features, early indications are that it has not meaningfully impacted the reach issue yet. Microsoft and partners have had to financially incentivise a significant amount of the developer activity for Windows Phone and that is likely to continue unless the platform gains much greater traction with consumers.

New Developer Platforms for 2013:

BlackBerry 10

BlackBerry 10 is due to launch at the end of January. From a developer perspective, the platform is cutting off the existing installed base of devices by not providing support for the legacy BlackBerry Java environment. Instead, BlackBerry 10 supports a native environment with an optional Qt-based application framework, an Android runtime and mobile web apps. The Qt platform was very popular with community of developers using it when it was Nokia’s primary environment and the developer satisfaction is reflected in the scores in our table above. It was one of the strongest platforms all round with the main weaknesses in reach and revenue. As the BlackBerry 10 platform is starting from scratch, reach is going to be an even bigger problem there and thus lack of revenue will also be a major concern for developers, despite BlackBerry App World’s historical record in this regard. The addition of the Android runtime gives developers a lower cost entry to a curated app store with lower competition, extending the reach of their existing investment. This suggests the Android runtime will be the most popular option for developers on BlackBerry 10, unless they can get an excellent start to sales of the new platform, driving developers to invest in native apps to establish themselves early with a superior user experience.

Firefox OS and Tizen

Mozilla’s Firefox OS and Samsung’s Tizen are both expected to debut this year and both aim to run mobile web apps as first class citizens. The Firefox OS has some backing from network operators and targets low cost smartphones. We’ll have to wait and see whether Tizen does better than Samsung’s last platform, bada, which had a promising start but little developer traction. Both platforms add more API access for mobile web developers and if they achieve any scale in the market, should also improve revenue generation. Overall, success for either of these platforms should make mobile web developers happier. The key question here is that if the web runtime is not offering a sufficiently compelling user experience on the existing platforms then how likely are web-only platforms to succeed, particularly with lower cost hardware. This strategy didn’t work for Palm with WebOS – can it work in 2013 or maybe 2014? Games are the most popular category of apps and they typically require the performance that only native apps provide.

Sailfish

The Sailfish OS from tiny Finish startup Jolla is due to launch a first device early this year. It’s also pinning its app ecosystem hopes on a native Qt-based application framework and Android app compatibility, much like BlackBerry 10. Developers should be very happy with the technical aspects of these environments. Jolla’s challenge is a lack of resources to get reach and thus solve developer revenue issues. Their plan for this relies on alliances with Chinese OEMs and networks, with western open source fanatics as a smaller secondary market. The ability of this platform to survive depends upon Jolla’s ability to do deals in China far more than their technical execution.

Ubuntu Mobile

Last but not least, newly launched Ubuntu Mobile also has a Qt-based application framework, this time paired with mobile web apps as first class citizens. As you can see from our table this is probably the best combination for developer satisfaction but they currently have no concrete device plans to share and don’t expect anything to ship until 2014. They do have a much bigger existing brand in developing markets than Jolla and theoretically a similar offering (both in terms of user experience and technology).

Developer Platforms: Conclusions

Android and iOS look set to continue growing their application ecosystems. Windows Phone appear to have another year of uphill struggle to gain acceptance. For any of these new platforms to generate developer activity on a similar scale to iOS and Android they need to achieve massive reach. That seems highly dependent upon Chinese OEMs deserting Android in search of differentiation (for all but BlackBerry 10) and consumer acceptance of platforms that, at least initially, have much smaller app ecosystems in every case. We’re not expecting any of these to gain major market share in 2013. If one of them does take off, history suggests that the developers who got in early will have an advantage. However, with so many relatively similar offerings on the way, it’s not clear where developers should invest their effort.

Categories
Platforms

How Much Time Could You Save With Backend-as-a-Service?

Backend-as-a-Service (Baas) provider Kinvey published an interesting infographic on the average time taken to build an iOS or Android app (with a backend service) this week. The data comes from a survey of 100 developers with their estimates averaged. Before interpreting the data, it’s worth bearing in mind the following:

  • This is only to build a Minimum Viable Product (MVP) – it’s nowhere near the total that would be spent on a successful app.
  • The features included create an app with a relatively rich backend service and a fairly basic client.
  • This is only building a client app for one platform rather than several.
  • Building a robust API versioning system would not normally be part of an MVP.
  • This survey has been created specifically to promote the benefits of BaaS.

Even so, the the items included in the infographic would be common to a wide range of applications.

Categories
Business

Growth Lessons from LinkedIn

Elliot Schmukler from LinkedIn spoke at a recent Growth Hacker conference about the strategies they’d used to grow the site since he joined in 2008. His advice was very helpfully summarised by Sandi MacPherson, Founder at Quibb and is general enough to be applied to mobile apps. Here’s our take on his main points.

Before you get started: Understand your channels

It’s important to understand how new users come to your product. Although it may be distributed exclusively through an app store, if that’s the only way users discover your product then you’re unlikely to succeed. This aspect of app marketing is still evolving rapidly, so keep testing different channels but focus your resources on the ones that generate the best results for you.

Categories
Business

The “Onboarding” Problem

With some types of mobile app, getting a user to download it is just the beginning of the problem. If the application is going to be personalised to a user’s preferences, or allow them to interact with others via some online service, then they’ll need to provide some data before they can start using it. Typically the more information a user provides about themselves, the better job an app or service can do of tailoring the experience to them. Unfortunately, the more steps a user has to go through before they can start using an app, the less likely they are to complete the signup process. Getting this wrong can catastrophically alter the economics of user acquisition.

Categories
Business

Which apps make money?

[This post by Andreas Pappas, Senior Analyst at VisionMobile, first appeared on the VisionMobile blog on 13 November 2012.]

[Andreas Pappas takes another look at the results of VisionMobile’s Developer Economics 2012 survey and comes up with interesting new insights on app monetisation: how does app revenue vary by app-category and by country? Is there a correlation between time spent developing an app and they money it makes?]

VisionMobile - which apps make money

In Developer Economics 2012 we discussed app revenues and how they vary across platforms. We found that overall, around half of all app developers that are interested in making money did not earn a sustaining income, i.e. they were below the “poverty line”, which we drew at $500 per month per app. Of course the real poverty line will vary widely across countries and regions: while $500 per month may not be enough for a San Francisco-based developer, it could be more than enough for a developer based in Bangalore where average living cost is less than a third, according to Numbeo.

Categories
Business

Different Ways of Winning on the App Stores

A recent report from Canalys highlighted the extreme concentration of income distribution across the iOS and Android stores in the US. The top 25 publishers make 50% of the revenues. 24 out of 25 of those are games publishers (the 1 exception is the Pandora music streaming service). During the first 20 days of November these 25 publishers made $60m from paid downloads and in-app purchases in the US alone. Is there still room left for smaller publishers? How can smaller companies succeed and start winning on the app stores?

Categories
Business

Advertising to Existing Users – could re-engagement reap rewards?

Currently the vast majority of mobile app advertising is used to generate new installs. At the same time, for some of the most successful revenue models, a small fraction of the most active users generate the bulk of the revenue. Free-to-play games are a good example of this model but there are similar in-app purchase driven schemes in other categories. Whilst a user is still very engaged with an app it’s likely that the most cost effective way to increase their spend is within the app. However, if an existing user stops using an app regularly ,then might there be more value in re-engaging users than acquiring a new user?

Categories
Platforms

From Mobile to TV: The companion screens opportunity and the role of apps

[This post on Mobile to TV, by Peggy Albright, guest author for VisionMobile, first appeared on the VisionMobile blog on  17 September 2012.]

[The latest trend in app development is targeting companion screens, as a way to bridge a multi-screen experience. Guest author Peggy Allbright investigates the future of app development on companion screens -and TV apps in particular – and discusses how TV advertising has found a whole new screen to engage users on.]

The Future of TV Apps

TV applications are opening up a new business frontier for the mobile industry. But we are still in the most nascent phases of the TV app industry’s formation and it needs to evolve on many fronts. Fortunately, early startup activities are revealing some of the roles that devices, apps, developers, merchandising and advertising can play in this industry, as we’ll see in this article.

The industry is well aware that consumers want to be engaged with their devices while watching TV and that many consumers are beginning to use mobile devices and apps as interactive “companions” to supplement the TV viewing experience. New research released by Google in August provides some of the latest data to characterize this trend. Google found that in a typical day, 77% of television viewers use a second device, such as a tablet, smartphone or PC, while watching TV. More than one-fifth (22%) of these consumers are using the TV and their second device in ways that complement each other, even if it is only a simple search related to the live TV programming.

Categories
Platforms

The yellow brick road of app store monetisation

[This post on App Store monetisation by Andreas Pappas, Senior Analyst at VisionMobile, first appeared on the VisionMobile blog on  31 July 2012.]

Apple and Google dominate the app store game – but only in terms of size. Senior Analyst Andreas Pappas discusses the key success factors for app stores, why Google is lagging behind and how Amazon fits in the whole picture.

With Amazon challenging Google’s app market and Apple allegedly offering the best monetisation potential, several developers and analysts have put these claims to the test by comparing monetisation data across the three app markets. Differences in monetisation potential can have a significant impact on developer mindshare: developers will seek to leverage platforms that will make them more money.