TheServerSide contributor and expert Barry Burd recently spoke with Skimble co-founder and CEO Maria Ly. Skimble is a San Francisco-based mobile health and fitness company. In part one, Ly discussed the basis of the Skimble fitness application and how the company came to fruition.
In part two, she delved a little deeper on the Android and iOS development process and applications support.
Maria Ly: We are. We develop natively for both Android and iOS development. It's a constant prioritization challenge because we want to do a bunch of things on both platforms all the time. We are constantly prioritizing features that we feel that will have the biggest impact -- and try to release them on both platforms as soon as possible.
No hybrid development? I hear arguments from both sides -- some people say nobody does native development anymore, and some say nobody does hybrid.
Ly: In the beginning, if you're scrappy, you will use a mobile webpage or two. Sure you might have to have a few mobile webpages for miscellaneous things like obscure settings. For us, we offer a native experience on both platforms for a variety of form factors. A native experience is the best experience, especially for the way to serve our unique, dynamic workouts. When people play a workout, they're guided through the entire workout move-by-move on their phone, tablet and watch -- this only works with a native experience.
You sound really comfortable with the fact that you've got two platforms to deal with. Many companies aren't comfortable with the need to support two platforms. But from the word "go" you decided to embrace both iOS and Android. You didn't think multi-platform development was something new that you had to tolerate and deal with.
Ly: We support both platforms. We're happy that we're on both.
What's it like launching on Google Play versus the iOS App Store?
Ly: As everyone knows, the iTunes App Store has a review process. When you submit your app for launch or update, it goes through a testing stage at Apple. So as a developer, you must schedule time -- a couple of weeks or so -- to get through the review process before you can release it. With Google Play, you can instantly release your app. Now, Google offers tools so you can do staged rollouts and A/B Experiments. In any case, it's pretty exciting to see your product go live and get used all around the world.
Has Apple ever failed to approve one of your releases?
Ly: Yes, once or twice.
And then what happens?
Ly: Here's one example: We were running a contest. It was a giveaway contest, and in the display of the rules and regulations we just needed to add disclaimer. I think it was one statement. We had to say that the promotion that we were doing was in no way affiliated with Apple. It was just one little thing. When that happens you can course correct. You do a resubmit. You wait a few more days and then hopefully your app will get approved.
So you can be nimbler on the Play Store.
Ly: Exactly. Both platforms will require a lot of planning to execute a successful app release or update.
What other differences -- between Android and iOS development -- come to mind?
Ly: There are so many. That was just one of them. For example, there are different design principles. We definitely try to leverage the design principles that each of the platforms recommends. For instance, on Android it's Material Design. We tapped into Material Design quite a bit with our latest update.
Is development slower on one platform than on the other? Does one platform require more hours or more developers?
Nowadays, more than ever, it's so easy to develop something that you are inspired to create.
Ly: No, I think it's equal -- just different. There are certain tasks that are easier to do one and harder to do on the other. For instance, Android was built first to support a multitude of screen sizes, so we got the hang of scaling up or down early on. On iOS, there were just a few device sizes to start, but, as you can see, Apple has released more sizes recently, and we have created our own ways to dynamically adapt to various display sizes. We don't really use interface builder.
Do you know what the split in your user base is? How many Android versus Apple?
Ly: When we first started, it was majority iOS. Just last year, we became majority Android. Most users are accessing our app via their mobile phones, 25% on tablet and multidevices. We're starting to see some adoption on watches too.
Do you have any thoughts on why Android users became a majority?
Ly: By 2014, eight out of 10 phones were running Android.
That's worldwide. In the United States, the numbers are different.
Ly: Our Workout Trainer app is available globally, so, as a whole, we're getting a lot of exposure internationally through the breadth of the Android platform itself and its overall growth worldwide.
We're doing this interview at Google I/O, so let me tell you a few great things that are specific to Android. Android allows us to do staged rollouts. This is a great feature because now we can roll out an app update at say 20%, and then we can check to see if there are any unforeseen issues before providing the update to our entire user base.
So only 20% of your customers see the update.
Ly: Yes. With over ten million downloads of our application on Android, it doesn't make sense to send an update to everyone all at once. Doing a staged rollout is something that every developer should take advantage of to ensure a successful app update.
Another great thing is the ability to do testing with Google's Cloud Test Lab. If you're just starting out as a developer, you might not have a ton of Android devices to use for testing. At Skimble, we have lots of Android devices for testing, but we can't have them all. There are thousands of Android devices, all with different form factors. The Cloud Test Lab offers us an opportunity to do automated testing. You can stress test the main flows for a multitude of devices. I'm excited about this opportunity.
There are already third party services that test an app on many different devices.
Ly: Yes, but the Cloud Test Lab is all built by Google, and it's free.
What other new Android features are you excited about?
Ly: We've tried A/B Experiment testing and having more oversight in the conversion funnels. We can do A/B testing of marketing. We've already run a few experiments including one for App Icon, Promo Graphic and App Descriptions in different launch languages. If we release a new feature, we can very quickly do A/B testing to see how the feature performs. Based on the test results, we can do another quick update. In other words, with A/B testing on Android, we'll actually end up doing more updates.
What's next for Skimble?
Ly: We're working on developing more tools to help trainers reach millions, including our community of fitness enthusiasts. If you know of any trainers who want to be a part of the fitness network we are building, they can get started by becoming a Verified Trainer on our platform.
Do you have any advice for someone who has an idea for an app and wants to create a startup?
Ly: Nowadays, more than ever, it's so easy to develop something that you are inspired to create. The hard costs involved are relatively low. You're not alone either. There is a great network out there to help support and mentor you. If you have an inkling of building your own business that solves a big problem -- not only one that impacts you or some small set of people that you know, but a problem with global impact -- then start experimenting and prototyping. It's always a challenge to get off the ground, but try to stay focused and make a product that people will actually use. Don't forget about design and the overall user experience. Get some feedback early on and adapt if needed. You may need to course correct a little or simply tighten your key value prop. As you launch your app, do whatever you can to understand who is actually using your service. Understand the community that you're serving. Community members will give you a lot of insight on how they use the app -- and how they don't -- so you can improve your product and provide an even more valuable service for them.
In terms of startup leadership, there are a ton of unknowns, and at times you may feel unsure, especially if you don't have a lot of experience or you don't have access to experienced mentors in your current environment. Perhaps you might also feel that you're not a stereotypical type of founder so you may not be successful. As someone who has gone through the "start," I can tell you that you don't need to be the most experienced person in the room or even the alpha female to be a great leader. Remember, there are so many different styles of leadership that can inspire a team to really do great work with you. I founded Skimble by working on something that I am truly passionate about and would be happy to do for a long time. I recommend that you do the same -- whatever your idea is. You don't need to be somebody you're not. Have conviction in your idea, keep iterating and learning, and remember to be honest with yourself. You will inspire others to do great work along with you.
What are your best practices for Android and iOS development and deployment? Let us know.