Author Archives: adityaponnada

Ideation tool for “Internet of Things”

Though many have considered it a “buzzword” in the world of technology, “Internet of Things” has certainly exposed us to enormous opportunities to build more ubiquitous and connected systems around us. While engineers around the globe are building more efficient protocols and platforms to host all the “things” that are meant to be connected, User Experience (UX) designers on the other hand are engaged to sketch relevant experiences around these complex systems. While the development phases for engineers are often well defined or are structured enough to start with, our UX folks still find it a very vague problem to ideate on. Especially, for a UX designer who is just going to be a part of an IoT related project for the first time, an ideation exercise may appear like writing a science fiction. In fact, as a UX professional myself, I had faced this problem of ‘how to start ideating for IoT?’

To begin with, it is important to understand why this whole deal appears vague and all over the place to UX designers. Unlike traditional information systems and currently aggressive mobile application designs, IoT simply allows more physical objects/appliances to be a part of the system in addition to different users. Moreover, the core purpose of IoT also entails the fact that technology should not be on user’s face. In other words, all the communication between different physical objects is happening in the background. Being such an open and all encompassing eco-system, IoT use-cases tend to overwhelm us. Therefore, to begin an ideation exercise, it is first important to set a structure.

This was a quick and dirty method that I came up with during my brief industrial stint at Samsung Research. And it was particularly helpful when designers have carried out a preliminary user research with their potential end users. Once you have gathered user needs(either from primary or secondary sources), insights and other critical information related to your target users, you should bucket them into four categories. These categories are — Users, Context, Things and Information. Let’s take a quick look at what these mean individually.

1. Users: Who are the different players of your system? This is certainly the most important thing a UX designer is concerned about. Since in an IoT setting, there could be many possible users, it is imperative to list down all the relevant ones. E.g. if we are to design an IoT experience for preventive health, our target users could include doctors, patients, family members, insurance officials etc.

2. Context: Context here can be defined as where and when this IoT system can be useful. You can define different locations (e.g. outdoors, indoors, car, hospital, kitchen etc), which have been derived out of the user research or secondary research conducted as part of the project. Context can also include time of the day and even as narrow as a specific activity (e.g.Medicine Adherence). However, this information should be grounded in your prior user research.

3. Information: Information in this case essentially comprises of information exchange between physical objects, exchange between objects and users and exchange between users (in case of computer mediated communication). Secondary or primary research carried out before the ideation phase can easily give a sense of the kind of information that is required. E.g. the study may reveal that diabetic patients require the date and time of their last vaccine, sugar content in recently bought yogurt etc.

4. Things: In the end, IoT is incomplete without identifying the most relevant “things” that need to be connected. You can populate the list of things that users use in their daily life or interact with for some important task. And yes, things can range from their personal devices to shared objects such as television sets. An essential aspect of knowing user’s “things” is to also notice how these interactions happen and when. e.g. a sedentary behaviour could be watching TV while eating. In this case, there is a possibility of a communication between the oven and the TV!

Before starting the ideation session, you should draw four columns on a white board. Each column should be titled as one of the four categories discussed above. Populate these categories with as many items as possible. Make sure that these items are purely derived from your prior study related to the project. Once the items have been populated, invite some of your team members. These members can include engineers, product managers, interaction designers, domain experts and even some of the end users. It is always good to have a more collaborative as well as a participatory ideation session.

Once all the team members have arrived, ask them to pick one item randomly from each list and then coin a single use-case out of it. E.g. if we pick ‘elderly’ from users, ‘Living Room’ from context, ‘medicine consumption’ from information AND ‘Television’ from things. The possible use-case one can coin is “The last medicine consumed by an elderly will be displayed on the television when he is sitting in the living room”. Since, this is a broad level use-case, it can be further detailed out to cover different workflows and possibilities. This way, with each set of randomly selected items from all the four categories, the team can generate a large number of crude ideas. Further based on the business goals and criticality of user needs/challenges, ideas can be prioritised for further development and detailing.

I hope this turns out to be a useful method. Please do share your experiences in comments or simply leave a note. 🙂

Generating Design Heuristics for Emerging Technologies

User Experience designers have been relying on design heuristics to quickly evaluate their designs for usability bugs. While it is undeniable that other methods of usability evaluation have evolved to be more robust and accurate, heuristic evaluation puts forward some of the high-level bugs at the early stages of design. For years, UX designers have been using Nielsen’s heuristics to evaluate information systems, however, with the increase in variety of platforms, the traditional heuristics have not been able to evaluate designs accurately. Traditional heuristics though cover sufficient guidelines that encompass human experience with an information system, they are often limited to web, desktop or control panel based interfaces. Emerging platforms such as mobile, ambient technologies, virtual/augmented reality, wearable technology and other natural interfaces require some dedicated guidelines that specifically target the critical aspects of these platforms. Attempts have been made to design dedicated heuristics for mobile platforms such as NN Group’s Mobile Usability Guidelines. Likewise, dedicated heuristics have been designed by Korhonen and Koivisto (2006), to evaluate mobile games in particular. This clearly suggests that UX designers are required to formulate special heuristics tailored to the platform they are designing on.

However, it is important to understand what makes a good heuristic set. A heuristic set becomes useful if it is able to identify more usability bugs (and of course more relevant and accurate bugs). However, what does it take a UX team to generate these heuristics? The answer lies in this 2003 CHI publication on “Heuristic Evaluation of Ambient Displays”. In this paper, they have demonstrated a simple yet a very effective way of generating as well as validating design heuristics for ambient displays.

This post is my attempt at generalising their findings, so that UX designers are able to generate heuristics for their respective platforms. To begin with, you should invite interaction designers and product engineers of the platform (e.g. VR Headset, Wearable Band etc) and discuss the core features or utilities associated with the platform. Then identify the primary goals of the platform and check which of the existing Nielsen’s heuristics are able to evaluate them. Then formulate a rough guideline for the remaining of the platform goals. As a result, you will have modified set of heuristics, which essentially comprise of core Nielsen heuristics as well as additional heuristics tailored to your platform.

Now since you have an additional set of guidelines, it is time to validate them. For this purpose, you can again invite usability experts (especially the ones who have some experience in heuristic evaluation). It is preferable if you have at least 4–6 experts, so that you can divide them into two groups. One group will be asked to evaluate the platform using existing Nielsen heuristics and the other group will be asked to evaluate them using the modified heuristics. If the modified heuristics are able to churn out more usability bugs, then we can reasonably conclude that modified heuristics can be used to evaluate the emerging platform at hand.

With increase in the variety of platforms in current technologies, it is essential to have improved and dedicated heuristics that can evaluate our solutions more accurately. If you found this post useful or have anything to add, do comment or simply leave a note. 🙂