Here we are at the end of another year, and as usual it was a busy one for the mobile developer space. What were the stories that mattered this year?
Platform Battles Roll On
Android extended its dominance of smartphone market share in 2013, with IDC reporting it took over 80% of shipments in the third quarter, compared to about 13% for iOS. Android shipments were up over 50% since Q2 2012, with iOS shipments growing just 26% over the same period.
But things aren’t that simple and straightforward, with lots of other relevant data in play. For instance, Millienial Media’s most Q3 roundup of traffic across its ad network found that 56% of impressions went to Android devices, and 37% to iOS, continuing iOS’ trend of overindexing on usage compared to Android.
Meanwhile in the business market, iOS’ strength continues. Intermedia, which runs a large hosted Exchange service, says that Apple devices represent 76% of its small- and medium-sized business user base. Good Technology reported similar data from its network, with iOS accounting for 72% of devices.
Within the Android segment, a clear hierarchy is emerging as well: Samsung, then everyone else. Millienial’s data shows that about 60% of Android traffic is from Samsung devices, echoing data from Mixpanel and Localytics. This figure outpaces its share of device shipments, which sat at around 40% in Q3, according to IDC.
One other figure bears noting between iOS and Android: the revenue gap is closing. Vision Mobile reports that at the end of Q3, the average iOS developer was pulling in $5,200 per month from their apps, with Android up to $4,700 per month.
Meanwhile, at the other end of the market, things remain interesting. BlackBerry’s popularity has slid, with its latest OS release failing to get traction in the market, and the release of its BBM messaging app for other platforms was too little, too late. Some good news for Microsoft, though: Windows Phone market share grew by over 150% in a year, says IDC, but still accounted for just 3.6% of sales in Q3.
The Old School Continues to Fall
One of the bigger stories in the mobile industry in 2013 was the sale of Nokia to Microsoft. The Finnish company was an old champion of the industry, but its early leadership of the smartphone market never recovered from the 1-2 knockout punches delivered by iPhone and Android. The experienced hands out there will remember when Nokia held a 40% share of the mobile phone market. While it still commands 15% of the overall market, its position in smartphones stands much lower.
Meanwhile, BlackBerry looks like it will be the next Nortel: a Canadian one-time telecom giant that ends up being split up and sold for parts like a stolen car in a chop shop. A planned buyout of the company by Fairfax Financial (dubbed “the Berkshire Hathaway of Canada”) fell through, and while it did receive a $1 billion cash injection, things don’t look good. The BB10 launch has been a failure thus far, and with no user base, no quality apps are going to follow, either.
HTML5, Where Are You?
2012 ended with no shortage of hype around HTML5 for mobile apps, with headlines touting stats like 84% of developers are either actively using it or plan to soon. Of course, many of these figures come from companies with skin the game, such as the one above from a survey sponsored by a – you guessed it – HTML5 tool company. But where is HTML5 really being used?
Research from Developer Economics paints a slightly more restrained picture. It found:
Just over half of mobile developers use HTML5, compared to 57% for iOS and 71% for Android.
Of those developers that use HTML5, 61% of them are making either mobile websites or web apps, with just 31% using them to create native apps either through wrappers, conversion tools or on native HTML5 platforms like Firefox OS.
But perhaps the most interesting data point comes from comparing the second platform developers use (see the graph above). For instance, when Android is a developer’s primary platform, iOS is the second choice for 40% of them, followed by HTML5 at 29%. Just 20% of iOS developers turn to HTML5 as a second choice, with 69% looking to Android. Meanwhile, for HTML5-first developers, Android and iOS are the second choice of just 39% and 32%, respectively.
This creates the image that “HTML5 developers” are a fairly separate population from native mobile developers, with relatively few Android- and iOS-first developers turning to HTML5 in a substantial way. This would be backed up by the other data from the report, which can be extrapolated to mean that just 18% of developers are building apps with HTML5, as opposed to websites or webapps. The further question still remains, though, is how many of those developers would consider themselves an HTML5 developer, as opposed to a “PhoneGap developer” or an “Appcelerator developer”.
Hackathons Keep Holding On
In the developer marketing space, we were surprised by the resurgence of interest in hackathons in 2013. At the end of last year, it looked like hackathons were becoming a bit played out, but they were back with a vengeance this year.
We did notice several trends:
More non-traditional hosts, such as consumer brands or other non-tech companies. Some of this was simply jumping on the bandwagon, but others were more of a testing toe in the waters to see how they might be able to work with developers.
The “hackathon-as-a-job” crowd is still active, coming to events for prizes and free food over anything. Careful event and incentive design can help mitigate their impact, but there are still plenty of people out there who think hackathons exist just as a way to pay some lucky developers.
Hackathons are losing their effectiveness as a way to unearth new and great applications or startups. The reality is that the majority of hackathon projects die after the event, with little interest from participants to take them further.
Smaller events tend to deliver on specific objectives with much more success than larger events. Common thinking is that a bigger hackathon = a better hackathon, and simply that getting more developers participating increases the results through the rule of crowds. Our recommendation is that when there are specific goals for a hack (rather than just overall visibility or community building), it’s better to spend resources on ensuring you get a smaller number of the right developers, rather than a large number of any developers. Investing resources in scouting and invitations – or even paying travel costs to bring in highly qualified developers – can be a better investment than buying meals for hundreds of random developers.
Operators vs. OTT
“Network APIs” were getting a lot of publicity last year, as operators moved on to them, instead of app stores, in their latest attempt to win over developers. But things have definitely pulled back in the space, with many operators taking their APIs largely out of public view.
This doesn’t mean they’re totally giving up on them, but rather making them more of a “private” API than a publicly available one. For instance, Telefonica’s BlueVia unit has announced it’s shutting down all of its public APIs, and will only offer a private operator billing API going forward.
Billing remains a highly lucrative area for operators, but it’s one that continues to slide as consumers enter their credit card data in wallets with the likes of iTunes and Google Play. Outside of that, it’s tough to come up with an example of a space where operator APIs are beating out over-the-top players. Location and messaging have been two strong areas for operators, but that strength has eroded.
It looks like the API ship has largely sailed for mobile operators, thanks to a deadly combination of bureaucracy, slow movement, and unfriendly processes and policies. This begs the question of what’s next – if anything – for operator developer programs?