Hi, new to HA, been reading and tinkering with it for a few days now on my pi3 with a simple wifi switch.
I’m curious if anyone here is deploying HASS professionally into other people homes/businesses? If so, would you mind letting us know your experiences and if/what challenges you faced. Thanks.
3 Likes
Not a direct answer as such, but I have had a few approach me about such a thing knowing that I have set up my own Home Automation with HA. In fact, I had one person that supervises a local oil well site that was extremely interested in using HA to set up some security/safety features. When someone is on site after hours, integrate some IP cams, when certain doors/gates are open, that sort of thing and use it to send out notices so that it is known that there is someone there on site, etc.
That particular project, while has some traction, has not been finalized yet.
I am of the personal belief that as versatile as HA is and its relative ease of getting it up and running could provide a nice side business to some enterprising person in setting up small to perhaps medium scale automation and perhaps some residual income via remote access maintenance, programming, that sort of thing.
1 Like
That’s what I’ve been thinking. It seems like a better alternative to something like control 4, crestron or savant because it’s entirely open and the customers (or installers) barrier of entry won’t be be nearly as high/expensive.
1 Like
I do remember seeing a post about a user who was running HA to manage his farm. I’ll see if I can dig it up.
Hi Robiet, Rpitera,
Andrew here from AgSense (NZ) - Creators of Farm Assistant.
Thanks for your interest in our work.
We are close to a beta release for some test clients.
You can follow our progress and get in touch with us here:
Our work has not yet been totally open sourced, but some parts are:
best regards
Andrew
1 Like
I’ve been using HA at work. Mostly using it with Zabbix and commandline/rest sensors to monitor and report status to a slack channel for hosts and copiers. Basic presence detection to allow IT team to know if we’re in the building or have left and notify other staff members via slack, although there’s multiple ways that can be done without HA. Z-wave flood sensor to report temp, water detection and apcupsd to report if the power has gone out in our MDF and notify a slack channel. Copier sensors are command line for messages/supplies and rest for status from a Papercut server. Notifications are sent to a slack print status channel to inform users when those machines go offline and online or when supplies reach a critical levels using automations. If copier has an error status for certain period of time it will report the asset ID, status, last messages and supplies in a private slack channel that is forwarded along to the local Xerox group for service requests. Also using some custom scripts that allow me to reboot the copiers if needed. Home bridge to control some wemo switches for a couple devices and projectors in a couple locations to make sure they don’t get left on too long by users. Will probably add more stuff as I need it but for now this was pretty simple to setup, keeps everyone in the know and makes life a bit easier.
Currently trying to figure out how to setup Hubot-Homeassistant so I can trigger some scripts/switches from slack…
Now using hubot-homeassistant to get status of and control specific devices
1 Like
Awesome, glad to see its viability for your situation.
Interesting idea. My current system (not hass) is touted as enterprise but it’s definitely not. It is very closed and the bugs just got to me so much I started exploring hass and openhab.
I’m not expecting home assistant to be enterprise status but if it is that would be awesome.
I’ve had the opportunity to explore installing an automation system for a few people but never felt comfortable with my system (again not hass).
I’m excited the idea of “commercial” is being explored. It would definitely raise the bar from hobbyist to something more.
1 Like
Have you com ea conclusion about it?
I think that as long as the commercial approach allows for more and more people to have access to this kind of technology and the vendor rewards the community through donations and such, that would be great for HA.
At the same time, I fear that this grows so much it’d get bought by a huge company, making it unavailable like it is today…
Cheers,
Antonio
1 Like
Having worked since more than two decades in the AMX/Crestron world, I’m very interested in HA as open source solution, open platform and flexibility it brings 
For now my main issues not to use it for some customers are these ones (it only reflects my own point of view and experiences
-> HA is not enough reliable right now (once you have setup an AMX/Crestron system you can basically forget it and it runs). I have a lot of issues of resync in HA itself and updates after updates of HA, I find it less and less stable (in previous versions when I was changing heating temperature, it was instantly adjusting control of heater, now it takes often few seconds and sometimes just hang in idle state till I change again settings and it finally triggers…
-> No hardware to sell in term of value: it may sounds crazy for some users but it’s very difficult to sell some programming and setup time on some hardware that has no value (raspberry or even a NUC is worth 100 bucks wher competition is always some thousands bucks just for hardware !).
-> Big advantage of HA is lack of user interface so you have multiple options to interact with it and so you can really make what you want (which is not possible with business competitors) !
-> Multi-system is not really easy to implement from what I have seen in HA unless it can be done with MQTT as bridge between systems.
-> Needs to build a ready system for new setup (to avoid pain to resetup everything each time) but can be done easily by ghosting a system all setup I guess !
Just few points I got quickly out of my mind
2 Likes
Thanks for your insight! Since I started this thread, I’ve installed HA in a few clients homes nothing major, pretty much just zwave and WiFi lighting control with alexa integration and a few other security related things.
I haven’t run into any hardware or software related issues, only issues I’ve had were indecisive clients lol, but thanks to hassio and the configurator, I can remote in and make changes, etc.
As for updates, when I install/configure the system and everything works reliably the way its supposed to, the client is happy and there really is no need to do updates unless the client calls me in to add some functionality.
I’m not sure about Crestron, but with Control 4 it’s the same thing, there may be updates available, but doesn’t need to be updated unless the client requests it. And most of the time requires not only a software update, but of course another $400-$1000 hardware replacement to add functionality.
The hardware is the next thing, the clients I have installed systems for haven’t known that a raspberry pi is not a premium piece of hardware because they haven’t heard of a raspberry pi lol. My clients for the most part aren’t part of the DIY community which the raspberry pi is designed for.
That being said, controllers like crestron and control 4 are ridiculously marked up in my opinion. The internals are not much more capable than a raspberry pi and are I’m sure worse than a mini Linux box with a ITX board and a decent processor. It’s been a while but, do those controllers have built in WiFi or Bluetooth?
As I said, ive only istalled HA systems for a few clients, and it’s worked out so far, but time will tell. Also, in their systems it’s pretty much all automation with some alexa voice control, very little GUI use. I’m working on a custom user interface using floor plan that I intend to implement after I feel it’s ready, so we will see.
4 Likes
I’ll add a few tidbits.
The lack of a front in HASS is no problem IMHO. I like that there are lots of options. You can use something like RTI or a DIY system.
Stability. I don’t have experience with Control4 or Crestron, but I ran CQC for many years which is a paid Windows system and HASS is much more stable. I don’t recall any stability issues with HASS.
Raspberry Pi? I don’t think HASS has any business running on a PI if you’re being paid to install it 
1 Like
What would you use instead of a pi? It’s small and unobtrusive and lowers the barrier of entry for most which is cost.
A NUC at a minimum.
I need to get my hands on one of those. How do they work with hassio? Or is hassbian the only supported option?
hassio will run on a NUC, per the documentation, but I won’t run hassio on my NUC…it limits my abilities too much.
hassbian is a raspbian with home assistant pre-installed. raspbian is a special debian build specifically for a raspberry pi. hassbian will not run on a NUC, because it isn’t a pi.
You can run any linux distro you like on a NUC, and run home assistant via Docker, Python VENV, HASSIO in Docker, or you can just install HASSIO directly on the NUC per the documentation.
I realized hassbian was rassbian after I posted, but you beat me to it lol.
Out of curiosity, what limitations are you running into with hassio? Also, what would you realistically pay for a basic NUC with all components installed?
not having direct access to the host OS with the tools that I want makes it a non-starter for my preference.
I think I paid $250 shipped for a used NUC with 8GB RAM and 256GB SSD. What do you consider ‘all components’?