Clutch spoke with Markiyan Matsekh, Product Manager for mobility and wearables at ELEKS, as part of a series of interviews about wearable technology.
Could you describe your company and your role there?
Our company offers full-cycle digital product development services. We’ve been on the market for about 20 years. We are based out of Eastern Europe and have offices in London and New York and about a thousand employees. We focus on six main service offerings: core software product development, data science, mobility and wearables, digital, quality assurance, and IoT [the Internet of Things].
I am a Product Manager, responsible for mobility and wearable offerings.
When did ELEKS begin working with wearables?
We began designing applications for wearables two years ago, when Google Glass was first released. We managed to get ahold of the device on its first iteration and started experimenting with it. Our experiment actually won awards, and since then, we’ve tried to implement solutions for our clients using different wearable platforms.
After Google Glass, we experimented with Pebble, Android Wear, and Apple Watch. Now, we continue to experiment with other wearable platforms, but those four are the main areas we work with in terms of creating wearable applications.
Wearables Platform Differences
You have worked with four major wearables platforms. Can you describe the differences between these platforms?
We’ve done a lot of research on the platforms. We also consult our customers about which platform would be the best choice for their needs. The smartwatches are obviously very different from smart glasses, and the smartwatches themselves are a little different from one another in their approach and philosophy.
Google Glass, in its current state and in the state it was for the last two years, is better suited for the enterprise environment. Although it was promoted as a consumer device, we didn’t believe in it as a consumer device because of a number of issues, both technological and social.
In the enterprise environment, Google Glass has some very bright use cases, which we’ve been working to capitalize on with our customers, but it’s still a very immature product. Because of this, none of our experiments on Google Glass have been deployed for mainstream production. Many of our customers have high expectations for a new version of Google Glass.
For the other smartwatch platforms – Pebble, Android Wear, and Apple Watch – the most popular device, the Apple Watch, is obviously a leader among consumer-oriented devices. It creates use cases for businesses to develop apps for the Apple Watch, in order to support different ways of communicating with their customers.
We’ve seen use of consumer watch apps rising in industries such as, retail, transportation, navigation, hospitality, banking and finance, and the media sector. Apple Watch has sold about 3.5 million devices already, which creates a strong incentive for businesses to support this platform by investing in wearable applications.
Android Wear is similar to Apple Watch but less popular for a number of reasons. From the development point of view, Android Wear is much more customizable. You can do many more things on Android Wear than on Apple Watch. But, we haven’t found many businesses that want to support Android Wear as a platform for consumers. However, sometimes it makes more sense to use Android Wear with enterprise, employee-oriented applications.
Pebble puts utility first and fashion second, while Apple Watch does the opposite. With Pebble, you have a stronger battery and four physical buttons instead of a touchscreen, which is more convenient. Also, the new version of Pebble has a microphone and color screen. We used the older version of Pebble as a platform for workforce on one of our projects.
Pebble is well suited for enterprise because of its simplicity, battery-life, and lower price. Buying a few hundred Pebbles for employees would be a better choice than buying a few hundred Apple Watches for employees. But again, it all depends on how the business wants to use the device.
Could you describe some of the wearable applications you have developed?
We have done more than ten applications for wearables.
Tesla Apple Watch App
The most famous wearable application we did was an Apple Watch application that controlled Tesla cars, which got a lot of media attention. That was one of the first third-party apps ever for Apple Watch, and I think the first Apple Watch application in the automotive industry.
We have to mention though, this was our initiative, not a request from Tesla. We created it as an experiment to see how far we could push the Apple Watch’s capabilities. The goal was to control the main functions of a car, like air conditioning, lights, map location, opening the car, and so on.
The app was not published on the Apple Store initially. We did it as an experiment, and we open-sourced our work. Later, the community grabbed our project, improved it, and published the application on the Apple Store.
Watch app for Havas company event
The second big wearable application was for a huge advertising and communications company called Havas. The application was more of a promotion than a product. Havas had an exclusive event with industry consultants, where they presented their company’s current state and plans for the future.
They wanted to impress the audience and bring them a more personalized experience during the event, so they decided to buy everyone a watch with an exclusive personalized app on it. We cooperated with them and developed a wearable application that included an agenda, so the guests could see who was speaking, when, and about what, receive personalized notifications, and view a customized watch face with the company’s logo. Each watch had a different colored watch face, and the color was based on the data we analyzed from the wearers’ public social profiles.
We also delivered notifications at specific times during the event, to alert people of a new speech or helpful information. These notifications were the most difficult features to implement. At that point in time, there was no Wi-Fi on the watches, so we had to implement custom Bluetooth architecture to make the notifications work.
Time Tracking App for Apple Watch
The third project, which we are working on right now, is a time tracking application for one of our customers. It is for Apple Watch OS2. The application, even though it may sound simple, tracks time and can be used as a time management tool. We agreed with the customer that it makes sense to develop an Apple Watch application for this purpose. A year ago, when Android Wear was out, we developed the same application for Android Wear. Now, it’s an obvious fit to build one for Apple Watch, as well.
With this app, we learned a lot about the Apple Watch platform. Even after building the Tesla application, we learned even more with this application. Watch OS2 is available in beta release now, and there are new functionalities that are not as simple as OS1.
The problem we faced was that Watch OS2 was still in beta, and every two weeks, a new beta came out, so some parts became broken in the platform. Another challenge relates to syncing the Apple Watch with the iPhone in real time. But, the ultimate challenge was to create a meaningful experience for the watch with a small screen and no keyboard.
Could you describe the process of developing these wearable applications?
The process for developing wearable applications isn’t too different from the typical software development process, although it has more emphasis on UX [user experience].
We started with some UX strategy and design sessions that included predicting how people would use the app, when, and in which contexts. The wearable platforms obviously bring new parameters, new attributes, and new challenges relating to how people use these devices. There are new guidelines to follow, but even the guidelines cannot tell the whole story of what the smartwatch experience is like.
The development process in general is,
- UX design,
- Testing, and
- More design and testing
We go through multiple iterations, until we make something that’s appealing.
Duration and Timeline
Tesla App - When we created the Tesla application, we spent about a month, maybe a month and a half. When we did the same application for Android Wear, it took a month and a half as well. When building a real product, you need much more time for testing in order to tailor the user experience.
Time Tracking App - With our latest application for time tracking, we built the basic functionality in a few weeks. Then the Watch OS2 came out, and we decided to add more features, but we had a lot of problems with the OS2 beta. We did a lot of testing and a lot of UX/UI [user experience/user interface] changes. It’s been about three months in development, and the basic version will be out soon. There is still a lot of work to do though.
At our company, building a smartwatch product, which is not just a prototype, requires a lot of testing and the creation of multiple iterations. With that in mind, I think it would be somewhere between $30,000 and $45,000.
From your experience, what are the ideal application features and functionalities for wearables?
The number one feature that I think is applicable for most wearable apps, and which can bring most of the value, is the watch face. It’s called ‘Watchface’ on Android and other smartwatches, and it’s called ‘Complications’ on the Apple Watch.
It allows developers and designers to show the most relevant information on the first screen that users see when they glance at the watch. The user does not need to launch any application or do any extra actions. She can just see what she needs to see with a simple glance. That’s a very important thing because it saves a lot of time.
In the Tesla application, we used the watch face to show the car’s battery status, which is actually the app’s most popular feature, with the most feedback.
In the time tracking application, we show the current tracked time for today and the specific timer that is running.
What challenges did you face during the projects’ rollout, and what steps did you take to overcome these challenges?
With Google Glass, the challenges are overheating, the battery, and the camera’s limited capabilities. You cannot complete the tasks you desire because, for non-trivial processing, Google Glass heats up and the battery dies within a few hours or less.
Development Environment - For smartwatch development, the biggest challenge is the development environment. It can be really painful because most of your time is spent on deploying, debugging, and rolling out the app, instead of actually building stuff. This is especially true with Android Wear.
UX - In terms of the development process, UX is a much bigger challenge for wearables than for other projects. Fitting the required functionality on a small screen, without a keyboard, with new input methods, and with short interaction span, is challenging.
Mobile v. Wearable Use Differences - It’s difficult to explain how different designing for watches is from designing for mobile apps. What is obvious now, but wasn’t obvious at the beginning, is that people use the wearable app for a couple seconds only. If the interaction lasts for more than three-to-five seconds, the user is more likely to take out her phone. So the challenge is to make the app as small, easy to use, and predictable as possible. We need to predict user input and then give them the information they need as quickly as possible.
We made this mistake when building the Tesla app for Apple Watch. It just had too many screens because we tried to copy the mobile app. Thankfully, we realized this while building the app for Android Wear and performing user testing. After this, we focused on the watch face and notifications.
Small Screen and No Keyboard - Another challenge is the absence of a keyboard on smartwatches. We either have to use voice or buttons for interactions, but we can only fit about three-to-four buttons on one screen. From a UX perspective, the small screen size is a huge challenge.
Mobile v. Wearable Interface Differences - There’s also a UI [user interface] issue specific to Apple Watch apps. The guidelines require that the apps have a black background. For mobile apps, which have a light UI, using the same color schemes on a black background doesn’t always look good, so you have to come up with a new style.
Voice Dictation - One challenge that we couldn’t overcome was how to make voice dictation work. For smartwatch apps, the perfect way to launch the app and interact would be with voice commands.
- ‘Okay, Google; start my timer number 1234,’ or
- ‘Hey, Siri, start my Tesla car.’
But, the platforms don’t allow this type of interaction. You need to say, for example, ‘Hey Siri, start my Tesla app,’ then wait for two or three seconds, and only then say, ‘Start my car,’ or ‘Turn on the lights.’
It’s the same with Android. That’s the big drawback of both platforms. We’ve addressed it, and Google is aware of it. We’ve communicated with them on this, and we hope they will fix it in the near future because that’s a big showstopper for user experience.
Bluetooth Usability and Stability - Regarding Bluetooth: we wish it would be more stable, with both Android Wear and Apple Watch. The phone can disconnect from the watch at any point in time. During the development process, you have to spend a lot of time to make it work as you expect.
In what ways did your initial conceptualization of these applications change throughout the development process?
With Tesla, we had to change things after we started the development process. We envisioned a basic design, started developing, and then realized our idea could not be done, so we had to go back and change things.
Havas Event App
With Havas, we did research for a day or two, and then we completely changed the picture. Back in April, Android watches didn’t have WiFi. But, we needed to communicate between multiple watches in a big room. Bluetooth signal was not enough because it only spreads about ten, maybe 15 meters, and the room was about 30 meters long.
We ended up taking four Bluetooth servers, putting them in different corners of the room to cover the whole area, and connecting them through WiFi. It totally changed the architecture of the app and influenced the experience.
So yes, it’s a very experimental platform with watches. You need to think of ways to overcome the challenges because they’re everywhere. In our case, we had to cut off a lot of functionality because we realized it wouldn’t be possible.
Time Tracking App
With the time tracking project, there was a huge overhaul of what we initially envisioned. When we started building the app, we imagined it as a copy of a mobile app, just on a smaller screen. We started with five screens on the main app. But, after prototypes and user testing, we realized this was not what we wanted. After the overhaul, we had one major screen with three buttons: start/stop new timer, apply timer template, and add voice comment.
Another factor that changed the functionality for the time tracking app was the release of Watch OS2, in the middle of our development cycle. With Watch OS2, we were able to add complications, time travel, custom voice commands, better animations, and more.
Could you share any statistics, metrics, or feedback that might demonstrate your applications’ performance?
I don’t think I can give specific numbers. All I can share is user feedback, our impressions, and our customers’ impressions.
Havas Event App
For the Havas application, the users were gifted with the watch. They used it for the first time at the event, and it was a very new experience for them. They didn’t know how to deal with it, so we had to instruct them and show them the basic gestures, voice dictation, and things like that. After a short period of time, I think they liked it. They’ve tweeted pictures with it, and the general impression is that it’s fine. I’m not sure they would use it on a daily basis though.
Time Tracking App
I think the time tracking app has much more potential for daily usage. My impression is that it will save a lot of time for the people who use it because they won’t need to get their phone out. If they start, stop, pause, or dictate some things on their timers ten or 20 times per day, they could save maybe up to a half hour each day. That’s what the smartwatch is about: saving time when interacting with technology.
With Tesla, we had a huge amount of feedback from both the users and the community. I would say that people were mostly delighted with the ability to check their car’s battery status on the watch face because that’s the information they need to know the most. Usually, a car owner would have to take the phone out, open the mobile app, check the car’s battery life, and then put the phone back. With the watch, it’s a very different scenario. It’s not life saving, but it’s convenient.
What lessons did you learn from your experiences, and what areas would you like to improve upon or do differently in the future?
We learn something new every time we work with wearable technology. It’s a very fast-changing platform. Things that were not available three months ago are available now, and you always need to learn something new.
I think one of our major findings is not to compare smartwatch development with mobile development. They are two completely different things. That’s one of the biggest challenges that we were aware of, but we still made this mistake subconsciously.
What we would like to improve upon in the future is timing: making app iterations faster.
We also want to have better user feedback. The experience on a smartwatch is something completely new. The way that the design and development teams foresee the final product and how they expect it to behave can be totally different from how users actually use the app. This information is beginning to be studied now.
What advice would you share with another organization that is seeking to develop a wearable application?
First of all, justify it. Not everyone needs a wearable application. It doesn’t always make sense. There needs to be a use case that proves that people would rather use a watch than a phone in a specific scenario.
Second, do not fall into the trap of replicating the mobile app functionality. It’s a completely new experience, so learn about it, research it, and then build for that experience – for that new context.
Future of Wearables
What role do you see wearable technology playing in the next 6-12 months?
Wearable technology is a misleading term because there are activity trackers, smartwatches, and smart glasses, all of which solve different problems, especially when comparing activity trackers to smartwatches.
I mainly talk about smartwatches because for me, it’s the most interesting area. I think that thanks to the Apple Watch, the category is growing quickly, and the general public has started to see some worthy use cases for buying it, charging it each night, and wearing it every day.
In the future, I hope there will be new use cases, apps, and experiences with smartwatches that will make them an even better platform. I expect voice control and other functionalities to be available on the platform, and I expect watches to deliver a better overall experience to both consumers and businesses that are trying to satisfy consumers.
What I didn’t cover too much is the Google Glass. I’ve mostly talked about consumer applications, but there is also a big wave of enterprise wearable applications. I think in the next six to 12 months Google Glass will probably rise as an enterprise device.