At the end of March, we launched an app for Apple TV. Our video team is small but mighty, and they produce great content. We’re glad they’ll be able to bring their talents to a native experience on your TV. However, this app didn’t come to fruition in the tradition sense.
***
We’ve been intrigued by Apple TV since Apple announced the new version at its fall product launch in 2015. We had added it to our road map, but it had been far enough down that it likely wouldn’t rise to the top, given the uncertainty around the opportunity. Slate produces short, fun, offbeat, original videos every day; could we find an audience on Apple TV?
***
Although we launched our iOS apps in 2011, we didn’t hire our first full-time iOS developer until last year. Hiring Chris Jeane, and giving him ownership over the code base, has substantially improved the reading experience. Since Jeane started, Slate’s app store rating has climbed from 2.5 stars to 4.5 stars. Traffic has grown accordingly. The app is now more than 99 percent crash-free. The code is cleaner and faster, making it easier to add new features (more on our future iOS plans in an upcoming post). Download it here. Hiring is hard, and we got lucky.
And that’s how we got our Apple TV app. Jeane decided to prototype a version of the app during a few days off. He showed up on a Monday and said, “I have this Apple TV app. Do you want to see it?” (Obviously.) It’s difficult to imagine a contractor, working on an hourly rate, doing this sort of forward-looking work. So if you want experimentation and fast, relatively inexpensive tests, having someone in-house is really important. It was a nice way to learn that lesson.
Because we were uncertain about Apple TV as a platform for Slate content, we knew we wanted to treat this as a minimum viable product, or MVP. In shipping an MVP, we want to make sure the UX makes sense, the product reflects a basic implementation of Slate’s branding, and we have a way to measure how the app is being used. We checked those boxes and submitted it to the App Store. We wanted to see if people would like it and use it, and if so, we could do more work on the app. After a month in the wild, here’s where we are.
***
We look at a few high-level metrics for all native app products: interest (downloads), engagement, and retention. Here are the questions we had before we submitted the app:
- Is there an audience for Slate’s video content on Apple TV? Are users downloading and opening it with any regularity? How much are they consuming when they open the app?
- What sort of content are users watching? Should we prioritize newsy content over evergreen content?
- The app comes without ads to start. Do we have enough interest from users and advertisers to make this a sellable product? And do we have the interest to invest more time and resources into it?
The app has been live for a little more than a month, and most of the app’s usage metrics are consistent with what we thought: Visits and usage spike on the weekends, with users consuming more videos per visit on the weekends (3.5 video plays per visit versus 2).
Visits have remained relatively flat over time. We’ve amassed roughly 5,000 downloads after being featured on the App Store’s home screen and news pages, but our daily visits have remained around 250 per day. Our hope is they would have increased as the app has been downloaded more and more. This is a problem with the app ecosystem in general: People download apps, use them once (or not at all), and never open them again. It’s why push notifications are so useful on mobile and tablet devices. There isn’t an equivalent on Apple TV at the moment. We also don’t force users to sign in, and there isn’t a news feed that we can use to push people into the app, so we don’t have a way to reach users who haven’t opened the app in a while. We’d love to discuss push notifications on tvOS if you’re listening, Apple.
Unique visitors also doesn’t present a particular interesting story; the folks who use the app open it once per day. We think that’s a direct result of two things. First, we’re only publishing about one or two new videos per day right now (more on that later), so there’s not much incentive to come back multiple times throughout the day. Second, tvOS as a platform doesn’t lend itself to daily repeat visits, which will be a difficult or impossible habit to break. Therein lies the major challenge with this platform: How do we make the app compelling enough to get the first bite at your attention when you turn on your Apple TV, and how long can we hold it?
However, there are two pieces of data that jumped out as us.
First, video-watching across mobile, tablet, and Apple TV takes place at very similar times during the day and week. Without having spent a lot of time with video usage data, I would have guessed mobile video–watching occurs during the week and tablet video–watching during the evenings, but save for a few spikes, usage is fairly consistent.
Second, our Apple TV users are the most engaged video-watchers on Slate, with an 85 percent completion rate on average versus tablet at 55 percent and mobile at 45 percent. Despite some of the retention numbers, the folks who use the app are really engaged, which better makes the business case for spending more time on it. It also suggests that these engaged users might be interested in supporting Slate by joining Slate Plus.
***
Here’s what we hope to do now that we have this data.
Traffic and engagement is higher on weekends.
Users are coming for evergreen content. The most watched videos are from our science section. We want to deprioritize live video, specifically around breaking news, as a feature and focus on unique, smart explainers rather than commodified breaking-news coverage. That’s great because it’s our bread and butter.
Users don’t open the app more than once per day.
We’re going to add more content, close to 100 percent of our videos. We’ve always seen content equals traffic across Slate properties. We’re also making improvements to the app to showcase more of that content (see below). These two improvements will be the first steps in establishing a real retention strategy in making the app more attractive to come back to and prompting users to spend more time with it inside of a visit.
We don’t have a large user base yet, but it is very engaged.
The app generated about 20,000 video plays in the roughly 30-day period we’ve been looking at, and video completion rates are at roughly 85 percent. While that’s a drop in the bucket relative to total video plays across Slate properties, getting more content into the app and establishing a retention strategy should grow that number.
Because there is interest, we need to integrate advertising.
More video means better business for Slate, so we’re going to be including preroll advertising at some point in the future. Couple that with our high video completion rate and this can become a compelling product for our sales folks.
We came into this without expectations and are pleased with this start. We’ll have a new version live this month, but in the meantime, open your Apple TV, download Slate in the App Store, and send us feedback at ios@slate.com. Thanks for reading (and watching).
