On a phone, if a feature is a bit awkward, people tap around until it works. On a TV, they put the remote down and watch something else.
At Travelxp we shipped the OTT app on Web, iOS, Android and several Smart TV platforms. The TV apps were consistently the hardest to get right, and not for the reasons I expected going in. The screen is bigger, but almost everything else about the experience is more constrained.
The remote changes everything
A TV remote has directional buttons, select, back, and maybe a few others. No touch, rarely a usable keyboard. Every interaction has to work with up, down, left, right.
That affects product decisions in ways people outside TV work don't always appreciate:
- Search is painful. Typing with a remote is slow, so anything that avoids search, like continue watching or good recommendations, matters more.
- Focus has to be obvious. Users need to know at a glance which item is selected. Subtle highlights that look elegant on a monitor disappear at viewing distance.
- Deep menus hurt. Each extra level means more button presses and more chances to get lost.
- Back must be predictable. If back sometimes exits the app and sometimes goes up a level, people stop trusting it.
Apple's tvOS design guidelines and Google's Android TV design guidance both go into detail on focus and navigation for the ten-foot experience. 1 2
Fragmentation is the real cost
| Area | What makes TV harder |
|---|---|
| Platforms | Several TV operating systems, each with its own SDK and rules |
| Hardware | Older models with limited memory and slow processors stay in homes for years |
| Release cycle | Store review and certification can take much longer than mobile |
| Testing | Emulators miss problems you only see on the real device |
| Updates | Some users don't update apps for a long time |
The hardware point is easy to underestimate. A TV bought several years ago may still be the main screen in the house. If a new home screen design runs well on a current model and stutters on an older one, a real chunk of your audience gets the stutter.
Plan releases differently
Because TV releases are slower and harder to roll back, I treated them differently from mobile:
- Bigger, less frequent releases, with more testing before submission.
- Features behind server-side configuration where possible, so we could turn things off without a new build.
- A device lab with the specific models our users actually had, not just the newest ones.
Our streaming metrics, broken down by platform and app version, were how we caught TV-specific problems quickly. Overall averages often looked fine while one TV platform was struggling.
Design for the living room
TV is usually shared and usually lean-back. Several people might use the same profile. Someone might be half watching while doing something else. That shapes priorities:
- Profiles and "who's watching" matter more than on mobile.
- Autoplay and previews need care. Helpful for some, irritating for others.
- Text needs to be readable from across a room, which limits how much you can show.
Parity isn't always the goal
There's often pressure to make every platform identical. I've come to think that's the wrong target. A feature that's great on mobile can be pointless on TV, and some things, like a strong continue-watching row, deserve more weight on TV than anywhere else.
The goal is that each platform does its job well for how people actually use it. On TV that usually means fewer features, done more carefully, tested on the oldest device you're willing to support.