Taking home automation to another level
This article explores designing the home automation systems, how beneficial it was working in pairs, and what we have learned about our pricing.
Introduction
We were asked by Qest, a Prague development studio, to help them with a home automation project for Haidy. We were looking to expand our horizons and play with hardware, so this came at the right time 👊
Haidy is a Czech market leader in smart homes. They can help you when you are building your house or want to upgrade an old “dumb” house 🏚. The goal is a full control of an app, saving the energies, and for the house, learning how to, within due time, be perfect without any intervention of yours.
Challenge
A smart home simply means that you as a user do not know about it. It fully adapts to your daily routines and addresses you only when it’s really needed.
The questions are:
- What are the people's routines so that we can adapt the app to their behavior?
- What are the key elements which should always be devised in a few clicks?
- What structure would enable the users to be always on the top of things?
Process
We endeavored to include all perspectives. We started with the stakeholders' interviews, then we moved on to talk to people working in the company, from technicians to salespeople. It turned out that the technicians were fountains of knowledge for us. And the users too, from those who had been using their system for 9 years to those about to move into their new houses.
After this round of interviews we drafted the first sketches and talked internally about the app structure and the main goals which we wanted to fulfill:
- Creating an interface through which the house can fully adapt to the users' needs.
- Equipping the users with up-to-date and within-an-easy- reach information.
- Structuring the information so that the users can easily access anything they might need.
- Conceiving an immediate feedback to all the users' actions.
- Producing long-term views, smart analytics, and findings for an easy translation into savings.
We elaborated on our sketches and turned them into low-fidelity prototypes to get feedback from the users. We faced an interesting challenge: Qest company didn't work to plan, and developed a web interface internally. That created a consistency requirement for the emerging interaction and visual style of the application.
After guerrilla testing — we tested the black & white prototypes with 5 people to verify the structure designed. We discovered some minor usability problems, but since the structure was working well, all users were able to complete the tasks according to the scenario, and we could move on to the visual part 🎨.
Since the developers had started implementing Android before iOS, we prepared a simple visual style, developing the basic rules of Material Design defined by the Design team of Google. Our starting point was the original version of Material Design. The new rules for UI development on Android allow for a more open interpretation of various visual styles. Unfortunately, the original version did not meet with the stakeholers' approval.
“The design was working perfectly. But it was dull”.
That was the feedback from the stakeholders. So we swapped the grey background for a white one, added some attractive patterns, unified and brightened up the colors, and spiced it up with a decent amount of skeuomorphism for the key elements.
You can have a look at our deliverables for Android phone because now it is under control of the Qest development team. The first version will be available in September 🚀.
Case in numbers
20+ devices integrated
3 months+ spent to deliver design for iOS/Android platforms
83+ number of screens in UX part
16 people in the team (4 SQ designers + 12 Qest developers)
Findings
I think we originally underestimated the project on almost every level, as it seemed rather easy at the beginning.
“You know, it will be only a couple of screens and we can move on.”
The more we understood the domain, the more problems started to arise. This was welcome because we could face a lot of edge case problems, such as: some users might want to create various groups of devices situated in various rooms of the house, or they might want to have different rules for each device while also including it in various groups.
“Knowledge of visual design is for nothing if you do not have specifications and moodboard approved. And that despite the trust a client has in you.”
What worked well, in this case, was that we created pairs of designers — UX & UI, so that one person was leading a part of the project and had a second pair of eyes challenging everything he had designed. This helped us to deliver a very good experience to the end users.
Conclusion
Home automation as any kind of automation stands for an invisible worker or assistant.
“We have the role as referees. The less we step into the experience of users the better work we deliver.”
Working in designer pairs is rewarding. One is leading a part and the other is kind of a “devil advocate”, which may be sometimes tough, but in the end, it pays off.
A golden rule: The more time you spend getting to know the domain you are in, the less time you will spend figuring out solutions to the problems which will arise later on in the project.
Last not least, from the business point of view, always try to estimate the project scope as soon as you can. In our next projects, we will try to develop an adequate pricing for our service, to be more flexible according to the findings in the course of the project.
This article wouldn't be possible without the help of Jiří Zlatohlávek, Vladimir Mokry, Stanislav Mokrý, and the crew of Status Quack. And our business partner Qest.
