Help us improve your Home Approval Factor!

Hey Home Assistant users! Matthias from Nabu Casa here. We are working on improving the Home Assistant user experience and we have spent the last month researching how to increase your Home Approval Factor.

One thing we have learned so far is that it’s important to make better UX not only for you - the creators of your smart home - but also for your users, the family members or roommates who live in your home.

Would you like to help us with our research? We would appreciate it if you could invite your family members or roommates to participate our interviews, and help us help you set up your perfect home. Interview will be in English.

We fully understand that you may have some concerns about them sharing their feedback with us, but we assure you that we value your honesty and we want to improve our products based on their insights.

Please let us know by sending me a PM and we will arrange a convenient time for the interviews. Thank you for your cooperation and support!

2 Likes

I live alone, so approval not required. However, visitors’ reactions are usually “Wow!”, closely followed by “Why?”. Just saying.

2 Likes

Please, add the ability to create proper users. If i’ve a son, or a guest, I don’t want them to be able to check into everything by using the search function, activate automations and stuff.

User limitations should be a thing from a long time, this is one of them greatest flaws in home assistant at the moment.

2 Likes

Hahaha, thanks! I will write this down in our research notes.

Great idea!

Along the same lines, how about getting some input from us who install and support HA? I don’t want to sound negative here, but there have been a couple of UI changes lately which were neither asked for nor well received by this community. Even if I’d prefer something different, an open discussion, and some indication that thought is being given to user input, would go along way toward making the change more palatable.

My 19 year old son probably never has seen the UX, as everything still can be controlled the ‘old fashioned’ way. He just knows lights turn on/off automatic, and I can yell at him from my work if he is still asleep at 7.

3 Likes

Over the past month we interviewed several Home Assistant users and gotten some great insights. We think we can also learn from users of our users. That’s why we asked our community.

Talking about a general opinion of the community is not fair to people with a different opinion. That’s why we do research to find out what the needs for different kind of users are. Topics on platforms like this give us insights. Feel free to open discussions, but we know that people rarely discuss with people that don’t like something.

Knowing what people want is one of the hardest parts of UX. We learn everyday and make changes accordingly. The conclusions of our research wil be shared with the community later.

2 Likes

Coming from another home automation platform, my biggest reason to start over was a lack of them fixing things that ‘improvements’ broke along the way. We all love to see what’s new every month, but not every month warrants a GUI enhancement or re-orginization. Sometimes just fixing things that have not worked as well as they used to (Sonos, for example) would be a welcome update for that month. I read the forums every day and think there are more than a few Integrations that could use some love.
Overall, Home Assistant is by far my most favorite platform for controlling my smart home and I have no plans to go anywhere else, but I do think that constantly evolving the GUI, how we build and maintain our workflows and the overall design layout should be secondary to a ROCK SOLID foundation, where everything works to its optimal performance.
As a power user, I code exclusively in YAML. I know this is not everyones cup of tea, but for me it is the most efficient, searchable way to keep all my code under control. The GUI tools are great for those that prefer not to code YAML, but it should not be the ONLY way to accomplish a task. EVERY integration that has moved away from YAML to GUI setup should have parallel capabilities in YAML.
Just my 2 cents worth.

1 Like

As far as you know. 19-year-olds can be tricky.

Agreed. It sometimes seems as if the imperative to update every month is distorting longer term development.

1 Like

The OP asked about non-technical user experiences with HA, so I’ll leave glaring (and well known) issues like constant breaking changes, lack of LTS release and all those out.

The first and most important thing one of your involuntary users (ie family members or guests) will be faced with is the UI. That’s with any non-expert software, from phone apps to TV menus. The difference here is that you build the dashboard, so ultimately you are responsible for the UX of your family members. Building a good UI is not an easy thing, it’s an art by itself. The big problem is that HA doesn’t make it easy for you, in fact it seems to be doing everything it can to throw sticks and stones in your way.

So as far as I see it, before doing any research on how non technical users perceive HA, take one step back and fix the frontend so that people can actually build a good UI easily. The HA frontend should support users in creating their dashboards, supplying tools to add, move and delete cards in a predictable and repeatable way, inviting experimentation while always being able to revert. Building a dashboard should be fun.

Right now, the frontend building experience is the total opposite. It’s by far the worst I have ever seen in a web based application. This horrific masonry layout thing, that auto arranges the cards in a completely unpredictable way, makes building a dashboard a frustrating chore. Things never end up where you want them to end up and it’s almost always entirely trial and error. Basically randomly click up and down on various cards and hope for the best. And then adding a single entity to a card will shuffle things around again. This is not fun and it is not productive. You can work around this by nesting tons of stack cards, but this becomes a monster to maintain as soon as you want to modify the slightest little thing in your dashboard.

People will start building their first dashboard with a lot of motivation and good ideas, then realize how tiring and frustrating that experience is. It usually ends up with a ‘looks kinda OK’ dashboard they won’t touch anymore because everything may fall apart with the slightest change or even with a HA update where the frontend is modified again. And that leads to suboptimal UX for all other users that (have to) interact with the HA frontend in your home.

7 Likes

A good Home Automation should not need to build an UI for family members or guest. For me, the UI is only to set things up, family members or guest need to be able to do anything they need without having to look at a dashboard.

3 Likes

I tend to agree, but that depends a lot on the individual use case. Just look at the number of people who hang up tablets on their walls.

And the op looks for feedback on family/guest UX. The only way these users would directly interact with HA is the UI. Any other interaction, through automations, voice or traditional control like wall switches, while arguably the most important interface with your house, all these are ultimately a function of how you design the automations in your home and are independent of the home automation platform. Your guests don’t care about what platform turns on the bathroom fan when they take a shower.

That’s exactly the reason I do everything in yaml and use the custom layout card for any extensive dashboard config. For a minimal dashboard I can fairly well predict where things will end up using yaml so the layout card isn’t needed.

I will say that the huge waste of space and time that are the updated and now ironically named “more-info” dialogues are a step in the wrong direction for a good user experience.

2 Likes