Forget Netflix and Disney+, which Connected TV device will win the streaming wars?

Co-Founder and Managing Director at FX Digital, Matthew Duhig, gives his take on which Connected TV device will win the streaming wars in an incredibly fragmented market.
by Matthew,  28th July 2021
Connected TV

While the likes of Netflix, Disney, Discovery and Amazon fight it out in their very own game of thrones to determine which of them is crowned king, there’s been another war waging that’s just as big and will have consequences for years to come. This is the war of Connected TV and which device will gain the greatest market penetration over the years to come.

For many years now Connected TV has been the less sexy, but still essential part of any serious streamers distribution strategy. To truly reach the levels of coverage that the likes of Netflix enjoy, content has had to be distributed to as many people as possible, and this has meant achieving a high level of device coverage. Up until now, Connected TV experiences have been relatively lacking, with users not wanting to do much else with their TV other than pressing play to start a stream. Frustratingly typing your username and password using a remote control pretty much sums up just how poor the CTV experience has been, however things are starting to change for the better. New, more feature rich devices have come to market, and these have been accompanied by more comprehensive and exciting platforms. Here I’m going to take a look at these devices, more specifically the platforms that run on them, to see if we can shed some light on who may be victorious in the battle of Connected TV.

Samsung Tizen

In recent times Samsung Tizen has been the platform that’s enjoyed the lion’s share of the market. It currently holds 13% of the global penetration of Connected TV platform share largely due to it’s impressive penetration into many international markets, where it leads in 22 of the 25 countries that were surveyed by a recent Strategy Analytics report.

Samsung Tizen provides users with a relatively slick interface along with all of the features you’d expect to see in a modern Smart TV platform such as; promo rails, universal search and voice controls.

Development on Samsung Tizen can be tricky. It is web based, requiring us to write to HTML5 specs. However, most developers tend to opt for the web hosted model, keeping the application on a self hosted URL, whilst including a reference to this URL in the application written for Samsung Tizen. Permission is needed from Samsung to use this approach, along with many of the APIs on the devices that can be tricky to explore (such as the APIs that drive the native AVPlay video player), but it’s in Samsung’s best interest to allow developers to use this method.

Development tools are also somewhat troublesome, with Tizen Studio being particularly buggy and the seemingly trivial task of launching a test application onto a TV via the Tizen Studio application sometimes causes immense pain.

The submission process for Samsung Tizen applications can also be difficult to navigate. The online self-service portal has had significant improvements recently and the interface is much more aesthetically pleasing now, but without a contact at Samsung or an experienced team such as ours to help encourage things along, it’s not uncommon for approval processes to take up to 6 weeks.

All things considered, I feel that Samsung Tizen has the coverage it does due to the sheer number of markets the Samsung TV devices are in, and just how mature Samsung are in each of these markets. It’s clear to see that Samsung has invested heavily in, not just the interface of Tizen, but also the back office tools. However, without further improvements in it’s developer and distributor tools, and a reduction in lead time as a result of improvements to their self service submission process, I think it could start to lose ground to rivals.

Find the right Connected TV platform for your brand.Contact us now



LG has often played second fiddle to Samsung, lagging behind it when it comes to market share. It’s another of the devices that benefits from having a presence in many markets giving it that global reach, so it still enjoys a larger share than the likes of Roku, Fire TV and the consoles.

Personally, I’m a huge fan of the LG WebOS operating system. It’s much more intuitive than the Samsung Tizen interface, where I usually get lost in the various menus I’m presented with. A couple of the notable features of the LG WebOS device is the handy “LG TV Plus” companion application and the “magic remote”.

With it’s keyboard the companion app is very useful if you ever need to type in a considerable amount of text into an input on the TV, and if you ever misplace your remote you’ve always got another way of controlling the TV with the mobile remote in the app. I’m less sold on the magic remote. Not only does it not come with every LG TV sold, but I’m also yet to see a really useful application of the technology, which works in a very similar way to the point and click of a Nintendo Wii remote. The magic remote becomes that little more frustrating for developers, as LG requires that we support it in order to have applications certified when we submit them for approval, introducing an additional layer of complexity to an LG WebOS application build that we don’t experience with other devices.

Similarly to Samsung Tizen, LG WebOS requires us to develop using web technologies such as HTML5. Again, we generally choose to load a self hosted URL into an LG and use the web application approach to build. As aforementioned, the biggest challenge when building for LG is perhaps the implementation of the required Magic Remote functionality. Video playback can be a little tricky as we begin to load it with features such as DRM, but this is true of most platforms.

Unlike Samsung’s Tizen Studio, the WebOS IDE and CLI are actually quite reliable and loading a development application onto the LG TV from the IDE can be done without throwing the TV half way across the room in a rage.

The LG application submission process is very similar to that of Samsung, with a self service platform. Again it helps to be in touch with someone from LG, especially if you’re keen to work with them to try to promote the application.

I’m a little biased here as I have an LG TV at home and absolutely love it, but to be honest the IDE is more than tolerable and, implementation of the magic remote aside, I don’t have too many complaints from a development perspective either. I think LG has more than earned the share of the market they have, and I think there’s good reason to think they’re not going anywhere either!

Roku TV

For those of you that are unfamiliar with Roku, it’s a huge platform particularly within the US, and for good reason. As a result of it’s enormous footprint in the US, it takes an almost equal share of the market to that of LG, without being in the number of markets LG is.

I mentioned that there’s a good reason for Roku’s dominance in the US. The interface is very slick and applications run very smoothly on the platform. Not only this, but you can pick up a Roku streaming stick for less than £30 in the UK. Roku’s dominance may primarily be in the US at the moment, however there are strong signs to suggest that they are heavily investing in expansion across the UK and Europe. More recently Roku have announced a partnership with TCL to launch Roku enabled TV devices in the UK.

Development on Roku is where things become a little sticky. Unlike other platforms that require us to build using popular languages such as HTML5, Java or Swift, Roku created an entirely custom proprietary language for their devices, and if you want to build an application for Roku you need to write using it. The language is Brightscript, and it reads kind of like Python. It’s not the most desirable language for modern day developers to want to learn (Javascript anyone?), which makes finding good Roku teams quite difficult.

At FX Digital, we have one team that can build using our Javascript Connected TV framework that outputs to all devices and we have a separate internal team that develops our Roku applications separately to this. Having to build separately adds to development cost and time, and also introduces an additional codebase to keep on top of. Having said all of this, having their own language means that the development ecosystem and tools on Roku are very reliable, and that applications are optimised to run on the known hardware that Roku also controls.

Again the submission process on Roku is a self service model. One thing to note, however, is that Roku is becoming more and more strict with their certification guidelines. It’s not uncommon to be pulled up for doing things outside of their guidelines, so it’s important to give the Roku certification criteria a good read before embarking on the development of a Roku app.

Roku has such an enormous footprint in the US, so much so that they’re pushing LG for the second spot in the market share tables. With a potential expansion across the UK and Europe seemingly due, and the very affordable cost of their devices, we can be in little doubt that the platform will continue to grow.

Apple TV

Apple TV, much like any other Apple product, manages to capture the famously loyal Apple customer base. I’m not sure I know anyone that has an Apple TV who doesn’t already buy into the Apple ecosystem through a device such as an iPhone. It may not put people off buying an iPhone, but I think the high price does put people off buying the Apple TV, especially when there are much better, more affordable alternatives available on the market such as the Amazon Fire TV or Roku stick. For this reason, Apple has a relatively low share of the market with it’s Apple TV device.

Much like all Apple products, the UI of the Apple TV is incredibly smooth and after some recent updates the remote control no longer causes severe frustration. As you’d expect, Apple applications run incredibly well on the devices and this is largely due to Apple’s insistence that applications are written natively using either Swift or Objective C.

Despite requiring us to use Swift or Objective C, the tooling available for developers in this space is considerably more advanced than that of Roku. Similarly to iOS on mobile, we’re able to use frameworks such a React Native TVOS to allow developers to build in Javascript, if that’s their preference. Using this tool means that we can use that very same build for Android TV.

Here at FX Digital, we go one step further and extend React Native to introduce support for the other web-based platforms such as Samsung Tizen, LG WebOS and STBs (set-top-boxes). This means that we can get significant device coverage with one codebase. If Javascript just isn’t your thing, that’s ok, you can stick to Swift or Objective C if you choose.

The submission for Apple TV applications is all entirely self-serve, and it’s incredibly rare that you’d have a close relationship with Apple during submission… not even Fortnite has that luxury. With some grit, determination, and patience, it can be one of the less taxing submission processes.

As great as Apple TV is, I do think that the price point is an issue. Understandably it doesn’t stop people buying an iPhone, but TV is different, and as mentioned previously whilst there are much better and more affordable options available, I can’t see Apple TV gaining much beyond its current position.

Android TV

As you may have realised, the Apple strategy of owning the device and software that they have implemented so well across the iPhone ecosystem is the same strategy they use for Apple TV. Android’s TV strategy very much mirrors their mobile strategy too. Android developed Android TV, and rather than installing this onto Google-only created devices, they allow the likes of Sony, Amazon, TCL and NVIDIA to install this operating system onto their devices too. This means that the footprint of Android TV is large, and growing quickly. When you consider Amazon Fire as an Android TV device (which it largely is), the combined market share means that Android TV nearly takes the crown from Samsung Tizen.

When it comes to usability and experience, I would consider Android TV to be one of the best available platforms. As well as having a very usable, performant and comprehensive user interface, Android TV is also blessed with Google Assistant which can be used to find content and even control video playback. Android TVs biggest strength is also it’s biggest weakness however, with their strategy of allowing third-party device manufacturers to use the platform meaning that the device you are on really impacts the experience you have with Android TV. Using it on something like the NVIDIA Shield is an absolute dream, while using it out on a low-end STB or Sony TV can be laggy and frustrating at times.

Android TV uses the same development tooling and approach as that found in Android for mobile. Android Studio is used along with the developer tools to give developers the ability to launch applications onto any Android TV device they wish. Furthermore there are many emulators available to test the app on, and while they don’t give an entirely accurate reflection of the app’s performance on a device, they’re useful nonetheless. When it comes to documentation, as you’d expect, Google has done a great job here too. Information is – for the most part – easy to find, and there’s lots of support online in forums such as Stack Overflow, such is the popularity of the mobile platform.

At FX Digital, using our React Native TV framework means that we can write the application once for Apple TV, and also benefit from this same code deploying to Android TV and web based TV’s too.

Similarly to Apple TV and the process for it’s Android mobile counterpart, the Android TV application submission process is self service. It’s not uncommon to receive some relatively obfuscated messages from Google regarding ‘issues’ with a submitted application, but with enough experience, Googling around, and some persistence, you can generally unblock such issues.

Android TV’s strategy of allowing telcos and device vendors to install the operating system onto the STB and devices they ship is going to give it a major advantage over rivals, and we’re already seeing the effect of this strategy in the explosive growth that was reported at this year’s Google I/O conference. It’s not just this device domination strategy that supports it’s claim for the throne though, the Android TV experience is amongst the best available across TV. For these reasons I’m expecting to see rather sizable growth for Android TV over the next few years.

Join the Connected TV revolution.Contact us now


Amazon Fire TV

Amazon Fire TV is the name given to a set of devices ranging from a small streaming stick, to a complete TV device, that all run using the Amazon Fire OS. Fire OS itself is a fork of Android TV, and for this reason the experience and capabilities of the Fire TV devices are not too different from that of the Android TV devices. This fork is often a few version behind the latest Android OS version, having said this, Amazon Fire TV devices are generally well resourced and run the Fire OS fork of Android TV particularly smoothly, which is more than can be said for some of the lower-end, less resourced, Android TV devices. Such is the affordability and performance of the Amazon Fire TV devices, that they can currently boast an equal share of the market to that of the Android TV devices they’re based on.

There’s not much more that can be said of the Amazon Fire user experience that we’ve not already mentioned about Android TV, however your preference on voice assistant could influence your purchasing decision. Fire TV devices come complete with Alexa as opposed to the default Android TV Google Assistant. Furthermore, when you buy an Amazon Fire TV device, you can be pretty confident that the devices are well resourced and can therefore run the OS and the applications relatively smoothly.

Amazon Fire TV uses an almost identical development approach and toolset to that of Android TV, even requiring the use of Android Studio to build and examine your applications. Where it does differ however is in the video player. Most Android TV developers will opt to use ExoPlayer as their chosen video player, and in the case of Amazon TV there is an official port of ExoPlayer that developers should use for playback support on Fire TV devices.

Any Android TV application that you build should be capable of running on an Amazon Fire TV device with little to no changes. At FX, thanks to the efforts of our core development team, we’re able to provide Amazon Fire TV support through a specific configuration of the Android TV application that is built from our React Native based framework, complete with everything that’s needed to get you up and running on those Fire TV devices.

Similarly to most others, the Amazon Fire application submission process is a self service process. It’s not often you’ll speak to Amazon (or need to) during the submission process as even promotional campaigns can be managed using their self service model. Passing submission can be a little easier than with Android TV, as application performance is generally a lot more predictable (and generally better) on the Amazon Fire TV devices than it is across the vastly varied Android TV enabled devices.

Amazon has made a relatively cunning move with Amazon Fire TV by taking one of the best and fastest growing operating systems in Android TV, and installing it onto a powerful and relatively affordable device. If you like the look of Android TV and want to be confident that the device you purchase will give you a smooth experience, it’s likely you’ll opt to purchase an Amazon Fire TV device, especially if you’re an Alexa fan over Google Assistant. Furthermore, for less than £40 you can essentially take whatever HD TV you have at home, and plug a Fire TV stick into it to enable Smart TV features, or to provide yourself with an alternative to another operating system such as Samsung’s Tizen.

It remains to be seen what the recent Chromecast and Google TV devices will do to Amazon Fire TVs market share. Common sense tells us it will start to gain ground, but Google often notoriously fails with direct to consumer products like this… Stadia anyone? One thing is for sure, Amazon is pretty big and powerful, they are the gods of ecommerce. If anyone can sell tons of Amazon Fire TV devices (or give them away) it’s them, so I’d expect to see strong device sales, even if it doesn’t have a significant impact on their market share.

Sony PlayStation

Having grown up in the 90s and 00s I have fond memories of PlayStation, with my first PlayStation device being a PlayStation 2, after many years spent on my Nintendo 64. When we started working with PlayStation devices it was like the child in me had been reignited. Receiving one of the Playstation Test Kits in a neatly packaged box brought back plenty of memories of unpacking my own Playstation 2 as a child. With regards to penetration, the PlayStation was and still is primarily a gaming console, but despite this it commands a relatively sizable share of the market similar to that of Amazon Fire TV and Roku.

Before I started using my LG WebOS TV, I used my Playstation 4 as a streaming device for a considerable number of years. I grew used to the relatively simple operating system and can’t have many complaints. Applications did seem to be a little more buggy on the device (using All4 was always a nightmare with it crashing after each ad break), but I put up with the issues at least until I switched to using an LG WebOS TV.

The development approach for PlayStation is considerably more complex than all other platforms. Firstly you need to be an approved partner of Sony in order to be able to develop streaming applications for PlayStation, and you need to purchase specifically designed Playstation Test Kits, with the typical consumer model not giving you access to what you need to get development applications on the device. Once you have a device you then need to familiarise yourself with the PlayStation nuances for building streaming applications, such as the requirements to use a proprietary Sony PlayStation framework to build and configure the application.

As it is with the other web based devices, at FX our TV application development framework builds a distributable web application that can be deployed onto the PlayStation devices.

The PlayStation application submission process is a somewhat self-serve model that can be a little fiddly, however this is accompanied by a great and supportive team that are prompt at answering any questions you may have. PlayStation can be a little selective about which applications they allow onto the platform, as they seek those that suit their gamer audience well. Considering this it’s always best to try to engage them early on in the process of any new application build.

We hope that one day the global chip shortage will improve and we will once again be able to buy PlayStation 5 devices without having to seek out one of the many opportunist console scalpers. In the meantime I can’t see PlayStation’s share of the market growing at all.

Microsoft XBox

In the wars of the consoles it is Playstation that currently leads, with the Playstation 4 outselling the XBox One by more than double, furthermore early signs also suggest the PS5 will outsell it’s next-gen counterpart at XBox. For this reason the XBox takes a lower share of the market than PlayStation and most of the other devices we’re looking at here.

Despite its lower share of the market, the XBox One device is actually quite pleasing to use, especially for developers. The interface is slick and smooth, and applications tend to run on these powerful devices particularly well!

The development approach when it comes to XBox is also relatively good, with developers needing to use the Universal Windows Platform (UWP) when building. In order to test an application there’s no requirement to purchase specialist equipment such as a Test Kit, instead you can simply switch a consumer model XBox One into developer mode and use Visual Studio to launch your application from a Windows PC. UWP applications can be written in C++, but we tend to find that building the application using the web based route is much more effective. There are a couple of gotchas you have to look out for when building using this approach, but once you’re aware of them it’s generally one of the more robust devices to build an application for.

Being a particularly powerful device, we can run a web based application on the XBox One with no problems, so at FX we opt to configure and build another distributable web application from our framework to deploy onto XBox devices.

The application submission process for XBox is entirely self service, and for the most part it’s often one of the least painful to experience.

With the XBox One being considerably outsold by PlayStation 4 and that set to continue with the next gen consoles, I can’t see the XBox devices taking much more of the market share than what they currently have, on the contrary we could start to see people using it less and less to for streaming applications, a trend I wouldn’t be surprised to see extend to PlayStation as users opt to use a Smart TV or streaming stick instead of their console.

Which device will win the wars?

I’m not sure we can look beyond Android TV. Google’s strategy, which has worked so well for them in the mobile space, looks like it could do the same for them across TV, with various vendors and telcos opting to use the Android TV operating system. Google themselves also now sell a Chromecast and Google TV for a relatively affordable £59.99, which will only add fuel to the fire. We’re already seeing signs that the strategy is proving successful, in the huge growth numbers that Google are sharing for Android TV. All of this, coupled with a relatively enjoyable developer experience and submission process, make it one of the least frustrating platforms to work with too, which will only further help in it’s growth.

If I were to point at some devices that I think will lose share to Android TV, I think the consoles are a clear bet. As the Android TV devices begin to flood the market, users that own both a console and Android TV device will begin to opt to use the latter for their streaming needs, leaving their console to run the games it was engineered to.

Let’s not completely dismiss the likes of Roku however, who I’d definitely highlight as a ‘one to watch’ as they continue their expansion into the UK and Europe. I don’t think they have the might of Google, or a strategy that supports a hugely competitive market share gain, but we will definitely begin to see their streaming sticks in more households outside of the US.

With the introduction of more and more devices, one thing is for sure, we look set to continue in a fragmented world of platforms and operating systems. It is for this reason that we decided to architect our Connected TV framework to be able to achieve coverage across as many devices as possible, with a single codebase.