The State of Video App Development 2015
Real estate always has a price, and a digital land grab in app stores and on set-top boxes is no different. There are more than a few things to consider about app development for mobile devices, connected TV, and game consoles. This article will attempt to capture the most important concerns for today’s competitive market.
There’s the audience, there’s the technology, and then there’s the content itself. Getting the three basic components in place is really critical.
We asked industry experts for advice about building next-generation streaming apps. There were a lot of opinions, and we’ve tried to offer as many as would fit.
Here are our experts:
- Michael Dale—Director, Product Management, Kaltura
- Derek Eichele—Director, Cloud Services, Kidoodle TV
- Eric Elia—Managing Director, Cainkade
- Barry Hartman—Product Director, Ooyala
- Ziba Kaboli-Gerbrands—Director, Content Acquisition, Roku
- Kaliel Roberts—SVP, Product and Technology, U.S. Digital, Discovery Communications
- Jeff Taper—Lead Architect, Digital Primates
- Nate Thompson—CEO, Ratio
- Simon Jones—VP of Marketing, Conviva
- Chris Xiques—Director Video Infrastructure, CBS Interactive
Step One: The Audience
There might be a perception you can publish once and run anywhere, but that’s not the case when it comes to user experience, analytics, monetization, authentication, and optimizing the user experience across different platforms. So the first thing to identify is how many platforms you want to deliver to.
“Be realistic about development costs, because the reality of the insane device fragmentation driving the conversation that we’re having is publishers [saying], ‘I need to be on more places than just the desktop,’” Xiques says. “If you want to be on Chromecast, Roku, Apple TV, and Xbox, you’re talking about four environments, and most likely you will need a different developer with a different skill set to get each one done.
“It’s almost inevitable, given the early maturity of these platforms, that you will need to do some custom work; there won’t be an out-of-the-box solution that does everything you need it to.”
The different platforms have various strengths and weaknesses. Consumer behavior for a TV is quite different than for a tablet. A 10? lean-back experience using a Roku is nothing like what you will have with a mobile or tablet device held right in front of your face. You can’t duplicate the same use cases on devices that are used in radically different ways.
“It’s not just mobile first, because you’ll hear that a lot,” Hartman says. “Mobile first has been a mantra lately, because people built desktop experiences and they tried to modify those or take them to a mobile experience, and there have been a lot of poor or failed approaches.
“I think being device-specific is really the view you have to take. It may be that you’re trying to create the same user experience or use cases, but the devices may dictate you do something different. The things you can put in front of the user will really vary a lot. Look for a solution or a platform so you can get a uniform, rich set of capabilities against all devices.”
However, before we get into the development perspective, remember that it’s about the viewer first. It goes without saying you would never build something for a platform if the audience weren’t there.
“We look at the audiences on the different devices to see which makes the best sense. Generally we do iOS and Android,” Roberts says. “Then we look deeply into whether we develop for Fire TV, PS4, or Xbox, depending on the audience that we’re trying to reach.
For instance, we have several animal sites we don’t develop for game devices. Then we have Phil DeFranco, a famous YouTube star, and his entire audience is on Xbox and PS4.”
The differences among platforms don’t end there.
“When you develop apps,” Roberts says, “it’s harder for user acquisition in comparison to driving people to a website. Each of the platforms is different. With iOS you can include a banner to the app store. With Roku and Android, it’s really a good idea to work with Roku and Google to try and get promotion within their respective marketing spots. It drives a lot of downloads.”
A key starting point for evaluating new app platforms is to identify what your goals are. If your goal is to reach across the widest array of platforms, you might face a different set of challenges than developing a marketing partnership with a specific company.
“One of the first questions I would ask is, ‘How do you want to measure success 6 months after you launch the experience?’” says Eric Elia of app development company Cainkade. “Is it PR? Is it marketing partnership? Is it viewership? Is it revenue? Each one of those may dictate which platform you’re going to roll out to first, second, or third.”
Many development companies say their clients reference the HBO GO experience. But what happens if you are just starting out, or deciding which platform to go for next and you don’t have the resources to build another HBO GO?
“The big three are IOS, Roku, and Android,” Elia says. “We don’t hear people talk about smart TV or connected TV platforms. We’re starting to hear about Amazon Fire TV.” The development environment for gaming consoles is less open, so smaller companies are focusing on easier platforms at the moment.
“The first thing that comes up with small- or medium-sized companies is an educational phase where publishers are looking to learn what it takes to get across different platforms,” Elia adds. “For a smaller brand it’s going to require a lot more development of that audience and marketing to build up awareness.”
Amazon Fire TV, Kindle, and a variety of phones and tablets all use Android. Roku is open and straightforward, while game platforms and Apple TV are fairly closed systems that require you to have a business relationship.
“Roku has been extremely viable for publishers of all sizes. There’s something interesting about that platform being a dedicated video viewing device where people buy it specifically for its streaming purposes,” says Elia.
Roku has the most straightforward app development process of any of the major platforms.
“The three things that make a Roku channel successful are the UI, the content, and promotion. Don’t make it difficult to find your content, refresh it 2–3 times per week, and be sure to make it visually appealing,” Kaboli-Gerbrands says. “Sixty percent of our content is ad supported, and we encourage publishers to have as much content free in front of the paywall.
“We have a Roku channel playbook that we make available to the partners,” adds Kaboli-Gerbrands. “From the time you gather the technical requirements, backend CDN, CMS, analytics, ad server, all the major feeds, it takes 2 to 3 months for the actual development—the scoping of the project, the concept, the design, actual build where the engineers build the channel. Then it gets submitted to our QA team. From there it takes 2 to 3 weeks [to go through] the certification process.”
While the Roku approach presents a very specific development path, other app platforms generate many more choices for development. The choices are varied and deep. We start to examine a few of the concerns next.
Step Two: The Technology
“It’s important to get something out early and then iterate,” Dale says. “An example would be getting your app out without ad monetization and then expand from there. Or you could launch with just progressive downloading, but not adaptive.”
But don’t rush it too much. “The temptation is to get something out there quickly with HTML5. You should resist that temptation a bit for now,” Dale says. “Understand what can be accomplished with both (HTML5 and without). Kaltura’s approach is to provide turn-key, templetized solution. For example the player component we use in our templetized application can also be used in your own custom application. So it varies whether they want to do app development themselves and have more control over the experience delivery, or whether it’s something that they want to get up and running as quickly as possible.”
So how do you iterate quickly? It starts with deciding on your required features. Do you need to handle advertising, analytics, DRM? Each piece increases the complexity.
“What we learned pretty quickly is that an app that plays video is a little more complicated than 90% of the apps in the app store,” Eichele says.
Kidoodle TV’s first app was very basic with few bells and whistles. Then it planned what it wanted to add, including feature sets, efficient ways of communicating with servers, updating data, and sharing information across different devices and platforms.
“Key for Kidoodle TV was knowing in advance the upcoming features to get efficiency for backend development,” Eichele says. “Having a road map guided this. You’ve got to understand what data you’re going to have to store to make those features possible.”
A common feature request from many publishers, including Kidoodle TV, is pausing your content on one device and then starting it up later on another device.
“I think the reality is that’s not actually all that hard to do so long as the services are architected with that in mind,” Tapper says. “It certainly is more complex to add this in after the fact if you haven’t considered it from the beginning.”
So iterating, identifying, and mapping out features are the first pieces of advice from the experts. Other advice centered on finding developers you trust and the debate over native or nonnative development.
Most everyone we spoke with said that you need to have a great working relationship with your developers. You want them to be experienced with publishing apps within the streaming space, and you want them to be in business for the long haul when you need major changes. This is true whether you choose to use a developer who will create a native or nonnative app.
With native code, you’ll have to write it separately for each platform you’re on. So iOS, Android/Chromecast, Roku, and gaming devices will each require almost completely different code. Invariably it will perform better, but it will cost more to develop because you’re writing it several times.
“The other alternative is to use a cross-compiled solution like Titanium and Phone Gap, where you can make an application perform well, but not as well as the native application,” says Tapper. ”The idea is you can write it once and use it in many places instead of writing it over and over again. We do a fair amount of each. More often it’s about the budget and not just the initial development budget, but also the long-term maintenance budget.”
“Native is typically considered to be more expensive to invest in, but it’s when you get the best performance. The user interaction, the responsiveness of the screen, all those sort of things can really be tuned to the device if you go down a native approach,” Hartman says.
Other effective development practices include being sure the connective tissue is contemplated before you start. Be clear that the third-party partners for your ads, analytics, and ecommerce will move out across all platforms.
Because Apple TV is a closed platform, it’s necessary to have a business relationship with Apple to develop an app for the platform.
“Look at platform partners and ensure they scale out against all platforms. Map dependencies out before you build an app,” says Nate Thompson of Ratio, a streaming app delivery company. “For example, how will you process payments? The way the different platforms do that vary quite a bit. Some of those types of things work differently on each platform. Assuming it will work the same way on all platforms would be a bad path forward.”
Related Articles
From ad insertion to analytics, successful video app developers focus on more than just the content. Here's a roadmap to creating a successful app.
04 Apr 2016
Creating jaw-dropping slow-motion and timelapse effects is now easier and more affordable than ever. Here's how to speed up or slow down time like a pro.
08 Jun 2015
Companies and Suppliers Mentioned