4 Keys to a Successful Interface Design
The only exception to this is on panels large enough for the environmental controls to be somewhat distanced from the source controls, and they should be available from every panel page to maintain consistency (which is covered in the next section).
There is another aspect to panel efficiency that is very important, but quite often overlooked. The interface should be designed to work well on the specific touchpanel device it is intended for. Device attributes such as screen size (not to be confused with screen resolution), how the device is physically interacted with and the location of “hard” buttons are of utmost importance when thinking about the efficiency of your interface.
A panel design that is 320 pixels by 240 pixels that works great on a 6-inch, handheld device might be very difficult to use on a 4-inch, in-wall device of the same resolution because the objects on the screen will be much smaller. Spend time playing with the device you are designing for and take notes on things you notice about it. Is it held with one hand or two? Will it be operated primarily by the user’s thumbs?
If so, perhaps the buttons need to be larger than you would normally make them. In the case of an in-wall or kiosk-style touchpanel, think about at what height the panel will be. Will most users be looking down at the panel or will it be closer to eye-level? If you normally put text labels below the buttons, perhaps they should be above them instead so the user’s finger doesn’t block her view. As you can see, it’s very important to keep the physical attributes of the device in mind as you design your interface.
Consistency
Any well designed touchpanel interface will be consistent in several areas. As mentioned in the segment on efficiency, the location of frequently used buttons should remain consistent across the entire panel. This may seem like common sense for controls such as “volume up/down” and “mute”, but this logic is often forgotten when it comes to controls for individual devices.
Many devices share a common set of commands. Placing these commands in exactly the same location on the panel across all devices that share them will make operation of the interface much easier for the end user. If the user learns that the “channel up/down” buttons are in a certain location for one device, she may be frustrated if those commands are in a different place for another device. This usually occurs when device pages are combined from several projects or a programmer makes pages for a new device without looking back at previous layouts.

The most commonly used buttons should be made more obvious through color or size variations.
This problem can be easily avoided by simply “proofreading” your touchpanel projects carefully from job to job. It also may be helpful to take notes in the form of a spreadsheet that you can refer to every time you create a new device page. This will help to keep button labels consistent from one device to another too.
If you’re not paying close attention, it’s easy to label a page flip button “next” in one place and “more” in another. Not to mention the common mistake of placing a “go back” page flip button in one location on one page and somewhere else on another. It’s important to keep in mind that the overall objective of being consistent between touchpanel pages and projects is to develop a system where learning how to operate one device makes learning subsequent devices much quicker and easier for the user.
Tailoring
This last objective is perhaps the most important because it determines the figurative “tool set” you will have to accomplish the other three objectives. The process of tailoring the interface for its audience, the clients, should begin very early on in the sales process with a meeting between the programmer, and the client. This is another situation where a carefully thought out spreadsheet or checklist of variables to be determined can be very beneficial.
Among other things, this meeting should be used to determine the overall look or theme of the panels. You should discuss what device controls are necessary for the customer and which ones can be eliminated to reduce panel clutter. Perhaps oversized buttons and text need to be used if the client has poor eyesight. Ultimately, the goal of this meeting should be to establish a game plan or “blueprint” for designing the best possible interface for the client in question. Many integrators rely on the salesperson or project manager to gather this information, but most often programmers will ask questions or think of potential problems that their peers will not.
Another key factor in tailoring a touchpanel design for a particular client is determining the style or “skin” that will be used. Most programmers have several templates or themes for their clients to choose from. It is important to make sure that you can achieve all four of the aforementioned objectives with each of the templates you offer.
For example, if one of your templates doesn’t have any oversized, button graphics and a client with poor eyesight requests it, you won’t be able to tailor the design appropriately to that client’s needs. Some projects might require custom pages that need to be created. In this situation it is a good idea to make some simple pencil sketches of what the page will look like before creating the actual interface.
This allows the programmer to give the client a preview of what they will get in case they have concerns or suggestions. Perhaps more importantly, it shows the client that his satisfaction is of utmost importance. Tailoring the interface to the client is no doubt time consuming and depending on the disposition of the client, it can be frustrating at times; but it will almost always pay off in the form of referrals, reduced instruction time and service calls.
There are no hard and fast rules when it comes to designing touchpanel interfaces. Many strong and varying opinions exist on the topic and what works for one user may be completely wrong for another. No matter what an interface ends up looking like, or what techniques were used to devise it, the four objectives outlined in this article will almost always need to be addressed — and the best interests of the client should be the highest priority when addressing them.
Subscribe to the CE Pro Newsletter
Read more Business Resources stories
People On the Move: JL Audio, NACE, Azione Unlimited, Crestron10 Reasons Coax, Not Wireless, Is Future of Video Distribution
3D: Tips to Reignite Consumer Interest
The Home Depot: Your Friendly Neighborhood Distributor?
13 Wire & Cable Tips from the Pros
More in Business Resources
About the Author

6 Comments (displayed in order by date/time)
CE Pro—
It is a sad state of affairs when you try to teach interface design via a magazine article.
While an interesting idea, you have to consider the platform that some in the home automation biz are saddled. It is no wonder that the iPhone makes every home automation panel (that means you Crestron, AMX and Control4) look like stooges from the stone ages.
An interface is built upon the hardware and operating system. When you start with crummy hardware and 1980-style operating systems, it is hard to build an intuitive interface. Clearly, some developers do better than others given the limits, but still the state of the art in the home automation realm stinks compared with what is running on today’s home computer and smart phones.
Let’s agree that the sooner Crestron and AMX go the way of the horse and buggy, the better off for home automation customers.
—JTD
Great article, Aaron, and thanks to CEPro for publishing…
Unlike another reader, I came away from that article with a handful of invaluable concepts and considerations to keep in mind when approaching touch panel projects. Not sure how it could be taken as an attempt to “teach interface design”, as anyone actually familiar with the products and the slightest inkling of what is involved with designing an interface for their users would realize that you couldn’t achieve that task in two pages…
Since this article was about the user interface, I’ll leave the questionable-at-best slams against the hardware and operating systems as a discussion for another place and time. Suffice it to say, if a client has the money to spend, they can get any kind of interface they want for their control system, and they’d be hard pressed to find manufacturers with a hardware platform to implement and support the control functionality that - when all is said and done - is better than Crestron or AMX’s. (And if Microsoft’s Project Natal technology is ever made available as a spun-off peripheral, I believe it’s a pretty safe bet we’ll be able to do “Minority Report” style interfaces with both of those vendor’s systems)
- Chip
So i guess Databoss likes to use his iphone as a remote control…
With iphone:
Slide to unlock…
Select app…
Select room…
Select source…
Select play…
Finally!
With Crestron:
Press Blu-ray Button… DONE!
I would probably have just as hard a time trying to use my touchpanel as a telephone. BTW.. apple can claim the rights to the ‘main menu’ user interface all they want but FYI.. we have been doing it for a decade.
I can’t wait for the day when I can go to my local AT&T;store for an iPhone, StarBucks and home automation. Maybe I can get that new Krups interface (what will StarBusks say?).There will always be a need for proprietary solutions, just as markets comoditize, they will also seek an upper limit. Thank You ADA, AMX, Control 4, Crestron, Elan, Home Logic, and all others who continue to drive the markets which do not comoditize my expertise and career.
Thanks for a well thought out and useful article on a topic that is critically important to client satisfaction and often the weakest link in the chain.
I certainly agree that programmers should participate in UI design, though ultimately I think the design should be owned by neither programmer, sales person nor project manager. Ideally, I think the requirements, specs and acceptance testing for the UI should be the responsibility of a separate role squarely charged with ensuring that the system delivered meets the functional and operational requirements of the customer.
In conventional software development, this role is often called “product manager.” Without having a specially trained and qualified person playing this role separate and apart from project management and programming, the likelihood of delivering a system that isn’t sufficiently intuitive, efficient, consistent, etc. is just too great. (In small shops, one person can play multiple roles - i.e. the project manager can also be the product manager - so long as this person is properly trained and qualified.)



Great article. It’s nice to hear a professional explain the process. I feel like im cheating on a test when i come across articles like this
.