Five Thoughts on the Amazon Echo Show
And what happens when a voice-first company makes a device with a screen
Formal reviews for the Amazon Echo Show came out earlier this week, and they made for interesting reading (you can find a roundup on Techmeme here). I pre-ordered a device for testing, and mine arrived Wednesday, so I’ve had it for a couple of days at this point, and wanted to share some of my own thoughts about it.
I don’t want to cover all the same ground as those reviews. Instead, I’ll focus mostly below on some of the other points that weren’t raised in the reviews I read, including some of the areas where the Echo Show could do better.
The Screen – hamstrung and laggy
The touch screen is the biggest single feature of this device that’s new, so I’ll start there. I’ve long argued that voice assistants need a screen for visual confirmation, for results to queries too complex to handle by voice alone, for recipes, and so on. So I was eagerly anticipating the implementation of that feature on this device.
What I found was that the screen does indeed add quite a bit of value, especially in categories like weather or shopping results, where either visual display or multiple options are useful. However, the implementation is lacking – while recipes are an obvious use case for a device many will keep in the kitchen, pulling them up requires a third party app – AllRecipes – which does an admirable job, but only for the minute or so it’s allowed to stay on screen. Getting back into the recipe requires asking Alexa to re-open the app (“skill”). That feels like an easy fix, but also an obvious shortcoming Amazon’s testers should have identified before the device launched.
But beyond recipes, there are two other ways in which the screen feels lacking: the inability to use touch as an interaction mechanism in many settings, and significant lag. First off, though the device has a touch screen, Amazon seems determined not to allow users to rely on touch as an interaction method.
That, in turn made me think about Siri on Apple devices. Yes, it’s got better over the years, and now works pretty well for many things, but I’ve always wondered whether Apple simply didn’t feel the need to lead in voice because there are so many other ways to interact with iPhones, iPads, and so on. When you’re not all-in on a particular interface, you can afford to have some gaps and shortcomings because you know other options will pick up the slack. Amazon, of course, is in the opposite position here – it’s decided that voice is the primary interface for the Echo line, and touch screens are an optional extra. But that’s tough to implement in a world where we’re all trained to use touch as a primary interface on devices with screens – Amazon is going against the grain here, and that creates unnecessary friction. There are many places in the interface where it feels like Amazon limited touch as an interface unncessarily.
But the other big frustration I’ve had with the screen on the Echo Show is lag. Very frequently, a request is met with what feels like a blank stare from the device for one or two seconds before it finally kicks in and does something. Quite a few times, I’ve assumed it simply isn’t going to do anything at all and started issuing a second request when it suddenly springs to life and responds after all. This feels like another area where Amazon’s disregard for the display hurts it – though there’s a visual indicator along the bottom of the screen when Alexa is actively listening, the main display doesn’t show any indication that the device is thinking or loading or anything else. As such, it’s not just the lag itself but the fact that there’s no indication anything is happening during that lag that’s problematic.
Communication and contact friction
One of the common themes in the formal reviews I read was that the reviewers had no-one to test video calling with except the nice person at Amazon PR. Given that these testers had in many cases already been emailing with that PR person, calling them was likely straightforward because their names appeared in the list of contacts on the Echo Show. But in real-world use, things are rather different, as I discovered when I started testing the feature.
Amazon has no history in communication or maintaining contacts or social networks on behalf of its customers, which means it has to hack those elements to make the Echo Show and Alexa calling and messaging work. The way it does that is to tap into the Contacts app on whatever smartphone app is running the companion Alexa app. Where contacts that appear in that app match up with registered users of the Alexa communication features, those users then show up in the Contacts list on Echo Show. So far, so good. But what you end up with in practice is a really random mix of people — mine included former colleagues, executives at companies I used to cover, old neighbors, and the odd family member. It’s certainly not the list of people I’ll want to use Alexa calling with.
And yet adding additional people to the list of contacts is a really tedious and friction-filled process. There’s no way to do so directly – you first have to add those people to your iPhone or Android contacts and then wait for the Alexa app to pick them up and recognize them and add them to the Contacts list on Echo Show. When I tested the feature, I had to go through this twice and appeared to confirm that there is a mutual element here – you have to be in someone’s address book and they have to be in yours before it will work. And of course you have to have the right phone number and/or email address in your address book too.
All of this and the other missteps with the Alexa calling and messaging feature previously documented just highlight how inexperienced Amazon is in the communication sphere and how little it understands about how people actually want to communicate. The Drop In feature feels like another example of that – I can think of no-one, even in my immediate family, that I’d be willing to enable to drop in on me unannounced in my home and listen to whatever might be going on in the few seconds it might take me to notice that they’re there and respond.
On the other hand, once you’re in a video call, it works great, the quality is good, the picture and audio are clear, and it’s a nice, frictionless experience. It’s just that getting to that point is far too painful. Amazon needs an easy way to add contacts directly to the device, perhaps through some kind of PIN system.
Speakers are great but intimidating
The speakers on the Echo Show are, as promised, quite a bit better than those on earlier Echos and the Google Home. The loudness and clarity are both much improved, and sitting right in front of the speaker is uncomfortable when the volume is all the way up.
That raises an interesting point, which is that when the device is playing music loudly, it feels quite intimidating to try to speak to it, like trying to talk to someone who’s yelling at you. But this isn’t just pychological: when the device is playing loud music, it really doesn’t hear the Alexa command as well, and you do have to raise your voice to be heard. It’s notable that Apple explicitly mentioned this scenario in introducing the HomePod earlier this month, and it’ll be interesting to see how well it handles this in practice later this year. It’s a minor downside to better speakers, but one that I did encounter in everyday use.
A video device with no way to browse video
One of the things the screen on the Echo Show can do is play video, currently only from Prime Video and YouTube. As such, you might expect voice commands such as “Alexa, browse Amazon Prime Video” would bring up a selection of videos to watch, but you’d be wrong. There doesn’t seem to be any way to browse Prime Video at all. There isn’t even a Prime Video skill or app. However, if you know exactly what you want to watch, you can call it up – “Alexa, play Man in the High Castle” brings up a listing of episodes for the Amazon original, and similar commands work for other videos found in Prime Video and YouTube. This feels like another example of the limitations of a voice-only interface — Amazon doesn’t seem to think users will want to browse anything, merely invoke specific content by name.
Naggy home screen
I should note here that I’ve had the Echo Show sitting next to me on my desk while I work, in order to enable frequent testing as things have occurred to me throughout the day. One thing that’s made me very aware of is the home screen on the device, which is always on unless the device is effectively turned off, and which displays a continuous cycle of messages throughout the day. Those messages, in turn, combine the somewhat useful with the completely inane, as shown in the time lapse video embedded below:
Note that “Insiders Piece” is a reminder from my calendar to write my weekly piece for Techpinions subscribers, and you’ll see that show up frequently. But much of the rest is inane news items such as “Spicy Food Keeps You Cool”, “Video: Spider Shoots 80-Foot Web”, and something about Rob Lowe seeing Bigfoot. The remainder is mostly nags to try this or that feature using Alexa.
Here’s the thing: pretty much every device with a screen that we use has the screen turned off by default: we explicitly turn the screens on when we need to use them for something, and then we turn them off again. The Echo Show is about the first device with a screen for the home that I’ve encountered where the display is always-on. And what’s on that screen is a remarkable combination of spam and nagging which is almost unbearable if you spend much time around it. That feels like a huge abuse of the trust someone exhibits by putting a device with an always-on screen in their home.
One of my big questions about the home screen is whether the nags to try this or that feature will eventually go away once the device is in a home long enough, or whether it will just continue ad nauseam. Arguably, one of the biggest benefits of a device with only a voice interface is that it fades very nicely into the background when not being used – there are no distractions, no nagging notifications, or anything else on the original Amazon Echo or the Google Home. If you’re going to put an always-on screen on a device like this, it should ideally stay fairly static and show useful information such as the time, weather, upcoming calendar items and the like. Having it cycle through those and the various other useless bits of information the Echo Show currently serves up isn’t the way to do it.
As I mentioned at the top, I was intrigued by the notion of a voice speaker with a screen, because the screen could potentially add a lot of value. And there are glimpses of that added value with the Echo Show. But for today, the implementation is still flawed in some pretty big ways, from the laggy screen to the selective use of touch as an interface, to the nagging home screen and the high-friction process of adding new contacts. Much of this could be easily fixed in software updates, and I’d hope that Amazon will make quick changes to deal with some of these frustrations. Whether it does or not will indicate both how serious it is about this device, and the degree to which it’s willing to sacrifice its voice-first approach to true usability.