Things SmartThings Got Right

We’ve tossed a lot of well-deserved grief at the SmartThings perpetrators :laughing: lately. Some of us have been around the HA (Home Automation) world essentially since the beginning, and have different expectations than folk who come to it fresh, expecting iOS-like simplicity and effectiveness. And, to be clear, ST has scored too many own-goals…

  • No effective support. And terrible messaging of known issues and of pending fixes.

  • Incomplete/inaccurate development docs (especially on items such as device IDs/network IP addresses, HubActions, running code locally)

  • Cloud reliability

  • Terrible implementation for adding apps/device handlers for non-developers: find-and-paste code into the graph.api IDE or peck around the non-intuitive portion of the mobile app.

Really, these qualify as “polish and support”.

But… I’m not thinking of leaving SmartThings. I’ve used/built many HA systems since the early X10 days. There’s a lot ST gets right…

  • Cheap. Let’s face it, you paid $99 for the only proprietary bit. A typical dinner and drinks for two in any big city will cost you more than that. If you have to throw it away, you’re out less than the cost of three Zigbee appliance modules.

  • Low On-going costs. No required subscription fees. Did you notice what you get for that? Some cloud storage, automatic (forced) updates, central management (graph.api), SMS sending, servers that talk to the iOS/Android app… they’re providing actual infrastructure, bundled into that low acquisition cost.

  • Compatibility. Unlike many of the systems out there, ST came with IP, ZWave and Zigbee functionality. Had I been given a vote, I’d probably have swapped out Zigbee for UPB, but still, a pretty good setup. You can’t add Zigbee to HomeSeer (probably the big dawg of the non-proprietary industry), and adding Insteon (which you can’t do for ST) would run $40 for the plug-in and more than that for the hardware… meaning your ST V2 hub costs less than adding a second protocol to an expensive HomeSeer system.

  • Programmability. Most HA systems are expandable in some way. I’ve written HomeSeer plug-ins. It’s not for the faint-of-heart and requires a complex set-up. ST is basically like scripting in JavaScript. I’d prefer more power, but the entry bar is lower than on pretty much anything else.

  • Community. We can gripe about ST right here, and other than the occasional closed thread, I haven’t seen any censorship. Some of the other HA systems have similar fora, but not many.

  • Expanding Functionality. This is because of the development features and community, but features such as my LANdroid enabling speech and alarms on Android devices and @rayzurbock’s BigTalker connecting events to speech (and adding speech to DLNA devices) happen here. And you reap the results free of charge.

  • Stability. Okay, yeah, it’s terrible for some. And the cloud reliance has been a problem. But mostly it’s been pretty good for me. For a device that costs nearly nothing.

  • Relatively easy to use. When it works, of course, but that’s most of the time. Some things are way too hard, such as dealing with broken devices, but nearly anyone can get an HA system based on ST up and running fast.

I do hope they don’t get discouraged or driven out by the focus on what they’ve failed to get right. There isn’t a better value out there, and the concept is great. I personally am frustrated by some of the support failures, but that doesn’t mean the hub isn’t functional and useful, just that it so easily could be better.

17 Likes

I agree.

The basics are working for me (if I open x door, then y lights turn on, albeit with a slight delay, but it’s manageable).

My reliability trails off when I start trying to automate mode changes (if I open x door while in Night mode, change to Home mode and turn on y lights). Not a critical issue and I can fix it myself with a couple of cheap androids mounted around the apartment with SmartTiles on them and a super basic dashboard.

One of the other big items that I like with ST is the fact they are continuing to add and certify new devices. Staples failed to do that and now that hub is sitting on a shelf collecting dust. Sure, the community can create custom device types and SmartApps all day, but getting official integration gives us a channel for requesting official support. Outside of that, the community is supporting each other (which is fantastic in itself).

1 Like

I agree, but I think they have / had potential to solve some of their own issues that could be easily solved.

I would pay twice as much for a system with a few more bells and whistles, particularly local control, maybe better presence detection via Bluetooth, etc.

People pay twice as much for a couple of bulbs, or significantly more for an Amazon Echo or a Sonos speaker.

I like that ST has a low entry point for the hub, but I just feel like a few more bucks could have made such a big difference.

1 Like

I would add four more:

  • Triggers. This seems so simple, and yet there are many systems out there, including HomeKit, which still don’t have a real trigger concept. Something as simple as the contact sensor opened.

  • IFTTT channel. Yes, it’s almost always better if there’s native integration instead of the IFTTT channel. But in a space that’s evolving so quickly, with new devices and services every day, having an IFTTT channel gives you immediate integration of some kind with many new offerings. As I’ve mentioned, I used the IFTTT channel for voice control via voice texts before the echo was available, and still use it when I’m out in the yard. And some devices, like window coverings and sprinkler systems, can be handled very well through an IFTTT integration.

  • Notifications. True, it’s not perfect. It would be nice to have more options. It would be nice to have a cellular add-on. But even so, notifications are again a piece of functionality missing from many competitors. In today’s mobile world, they add a lot of value.

  • Virtual Devices. If you use IFTTT, or indeed if you want to connect SmartThings to any outside services, the ability to create virtual devices is really helpful. There are people who never use them, or people who will always see them as just a Kluge. But as an interface mechanism, they are very powerful.

These four things together, plus the community, keep me with SmartThings even though I have moved most mission-critical use cases to other controllers. Something as simple as getting a text and a colored light notification when the front gate is opened or the guestroom window was left ajar are quick, easy, and economical to set up with SmartThings.

I can’t trust that they’ll work every time, but there’s still real value there. :sunglasses:

7 Likes

This, in a nutshell, is exactly how I feel about ST.

2 Likes

Is not much else to say they got right, besides delivering a platform that allows developers like @bravenel and many others to rock the boat of HA with exceptionally flexible applications.

If the platform didn’t exist, we would not have anything to talk about. Sure there are lots of improvements that we all want to have happened yesterday, but instead, they are happening at a slower peace. However, the foundation is here to stay and we can actually build on it today, the things that were not possible just a few years ago.

Kudos to the engineers who crafted it, and for the founders who dreamed about it!

4 Likes

The customer support is beyond stellar.

When you get a response.

From all the posts I’ve seen on here lately, it seems a lot of people aren’t having much luck with support

1 Like

I have had to contact support at least 3 times. Each time they were personable, knowledgeable, friendly and took time to fully research the problems.
I couldn’t ask for a better experience. One-One contact. This is the reason that I have stuck with them thorough their growing pains.

4 Likes