Designing Mobile Applications with Customer Data

Techniques that Work for Mobile Platforms

CHI 2004 SIG

At CHI 2004, a formal SIG provided a forum for discussing how customer-centered techniques can best be used to gather requirements for mobile applications, especially considering the possibilities that new technologies offer. After initial presentations by a panel, over 80 people broke into five small groups to brainstorm issues and solutions. The chosen topics were:

1. How do we collect data when users are mobile moving from place to place?
2. How do we engage users in inventing new solutions especially with technology not yet available?
3. How can we keep navigation simple enough for the small form factor?
4. How do we effectively use the small display space and limited input modalities available on a mobile device?
5. How do we include branding considerations on the small screen?

The results are displayed below.

Topic 1: How do we collect data when users are mobile?

Issues

1. We need to collect data in different contexts including different seasons, locations, etc.
   • The very nature of the technology means that users will need to utilize the technology in many different contexts
   • Ideally we would want to live with the users; but obviously can’t
   • Nevertheless, we do need to be able to follow users around into many different contexts
   • But we are constrained by limits on time and money regarding the contexts in which we will study users
2. The contextual dimensions we may want to vary include:
   • The individual’s surroundings
   • Who they are with
   • What types of devices they have with them
   • Who their provider is
3. We could use a video camera to capture the user in these different contexts, but cameras are not always capturing the right stuff or the capture method makes the participants feel uncomfortable
4. Issues involved in capturing data on mobile users include:
   • How you capture data in relatively personal contexts without unduly influencing the user
   • It is very difficult in an unobtrusive fashion to see what the user is doing with a small device
   • Users may give different interpretations of the situation than what you have observed
   • How you choose situations without losing interesting data from other contex

Solutions for collecting data on mobile users

1. Use logged data to supplement other data collection (challenge is to synchronize different types of data that are collected)
2. A diary might help determine what to observe (but hard for them to do it at same time)
3. Anything that reliably captures context is good: camera, microphone, GPS, interaction log. But we would want to supplement these data capture techniques with prompts and probes about what the user is doing
4. Record audio interactions while users are using their mobile phones
5. Rely on self observation and reporting, but we would have to motivate users to do it
6. Use public observation cameras to limit the extent to which we influence the users we are observing
7. Enlist the assistance of friends or colleagues of the user to serve as confederates to less obtrusively observe the user
8. To validate our interpretations, we must check our analysis with participants
9. To sample the most important or significant contexts, we need to know what are the most important data flows
10. To ensure that our data is not unduly influenced by one method, we should combine methods to triangulate on an accurate understanding of the user
11. Use wearable mini-cameras to capture data less obtrusively

Topic 2: How do we engage users in inventing new solutions?

Issues

1. Users don’t know the technological possibilities
   • What they think is possible is limited by what works for them and their knowledge of today’s technology
   • If we share technological possibilities, how do we know if new technology ideas generated by users support real needs or are just ways to use the technology?
2. Unrealistic expectations may set unrealistic expectations for customers
   • If we encourage users to think “out of the box” we may set unrealistic expectations on when new capabilities will be delivered
   • How far out should we envision the future; will that set unreal expectations?
3. How do we help users to be creative and invent new possibilities?
   • If we provide users with examples they can both be suggestive but may also limit their thinking
   • Do we restrict users’ visions?
4. How do we evaluate user suggestions
   • If we engage users in brainstorming, when they move into evaluative phase how do we get positives not just negatives?
   • If we know there will be a problem with new technology, how can we assess the impact of these problems on the user?
   • How do assess whether users will adopt new technology?
5. How can we combine getting users to vision with collecting data about their daily lives?

Solutions for enlisting users in brainstorming new solutions

1. Ways to get user data on current mobile user behavior
   • Field studies in context to see what they do now
   • Use a natural group like a community, friend, family, or work group to collect data on mobile group interaction and leverage relationships
   • Talk to customer care for issue
   • Observe and interview people about unintended us with mobile devices
   • Keep diaries of daily activities and see where frustration
   • Go to where the technology is already used a lot and study it there
2. Ways to get user’s to imagine using new technologies
   • Group visioning process after normal field interviews and introduction of possible technology
   • After diary — invite the users to think about what did during the last 3 days and have them brainstorm how they would communicate if they only had images
   • In a focus group ask them what they want to do in different locations
   • Use a metaphor they can relate to (mechanical production of things) and ask them to imagine how this could be done through mobile devices
3. Ways to affect the scope of the brainstorm
   • Tell them to imagine that they have a magical genie to get them to what if about a different future and broaden their scope. Then follow up questions
   • Start with a broad scenario and then tell them what if you only have a phone. Help them open up and then narrow to something doable
   • Present a wide scenario to brainstorm. The brainstorm opens up wide thinking, then narrow the brainstorm just to desktop tools or other software
4. Ways to test your design and future concepts
   • Present your scenario based future idea in a 1-1 interview getting their reactions
   • Start with a Contextual Interview but bring along your mockups of the new system or a current prototype, and then get their reaction after they become more conscious of their processes through the CI
   • Build and present proto with new tech – and get feedback. See what they say if your prototype is so good it exceeds expectation.

Topic 3: How can we keep navigation simple enough for the small form factor?

Issues

1. Orientation:
   • Users need to know where they are in the application relative to where they have been and where else they can go in the application
   • Users need to know how to get back to those parts of the application that they have been at
2. Users apply specific mental models when using an application that affect their understanding of:
   • If I do “x” — what will happen?
   • What are the results of action?
   • Overall understanding of how navigation within the application works
3. Mental models are influenced by user’s familiarity with other applications that they have used on both similar devices (e.g., other mobile applications) as well as in other context (e.g., Internet). Users’ mental models are also affected by cultural differences.
4. Mobile devices have a limited number of buttons that can be used. Effective use of these buttons must be achieved without overloading function of each button. Designers must determine:
   • What operations to associate with each button actions
   • What functionality to associate with hard buttons as oppose to soft keys
   • Whether to associate multiple functions with a soft key and what and how many functions to associate with a soft key
   • How to communicate the functionality associated with a specific button to the user
5. Users need to understand how to select an item and must be able to clearly see what item is selected
6. Designers need to determine how best to structure content on a screen e.g., into tabs; panes; multiple screens.
7. Users need to be able to easily navigate between primary function of the mobile device (i.e., answer telephone) and secondary functions (access data)
8. Users need to be able to easily navigate between accessing and viewing data and inputting data
9. Users need to understand how to recover from error conditions or be guided through error recovery

Solutions for simplifying navigation

1. Help users get or maintain their orientation within an application:
   • Bread crumbs can be used to help users track where they have been; but bread crumbs are less than ideal since they use valuable screen real estate
   • Organizing into higher order categories may be helpful since by showing the users what higher order category they are in; the users may a better understanding of where they are at any one time in an application
2. Designers should be aware of the mental models users bring to an application. However, not all “mental models” can be appropriately applied to the small form factor.
   • The model of a browser may have limited utility for small form factor applications since much less screen real estate is available and there are more limited navigation controls
   • The model of a TV remote control device may be more appropriate since many mobile devices have the same 4-navigation controls as a remote control device
3. Providing users wizards or guided tours may be the most effective way to ensure that users understand the mental model they need to apply to navigate through an application
4. To address the limited number of buttons available on most mobile devices developers can:
   • Use a number pad
     • If appropriate, allow user to interact through a touch screen
   • Use voice input
   • Use icons
   • Use a combination of techniques
5. To facilitate users ability to easily move between taking a call and using the data functions of a mobile application; the application needs to:
   • Save information when a user needs to switch to taking a call
   • Provide users a simple toggle between the telephone function and application function of the device

Topic 4: How do we effectively use the small display space and limited input modalities available on a mobile device?

Issues

1. Devices may limited input modality to selecting items and operating on the items
   • Keying input may not be an option
2. The same application may need to be able to run on devices supporting different and multiple input modalities including select and click, key-in input, and voice
3. For output we must choose the best mode for displaying information e.g., text or icons; when and if to use graphics; the use of voice output
4. The best principle to follow in displaying output on a small form factor is “less is more”; i.e. it is better to focus on the displaying the minimal information required by the user
5. Other issues that must be taken into account in designing for the small form factor include:
   • Limiting the amount of time required to upload or download content
   • The limited amount of memory available on a small mobile device; must determine which data can be presented as static and what information needs to be downloaded on an as-needed basis
   • The need to localize the content to the specific demographics that is utilizing the application

Solutions for the limited display space and input modes

1. Provide redundant information by communicating the same information through multiple channels (e.g., text, graphic, voice)
2. Provide the user with optional methods for input/output
   • E.g., select and click; touch screen; keys
   • Provide different input modalities depending on context
3. Leverage the ability to display content from handheld device on a larger output screen when available
4. Make effective use of graphic metaphors that can convey considerable information through their layout e.g. a diamond to convey players positions on a field
5. In designing for the small form factor it is important not to simply copy UI solutions for larger display screens
   • If porting a design from large screen to small form factor, it’s better to design from scratch
   • Apply approaches for designing for the small form factor from other UIs designed for the small form factor
   • Study how the user interacts with the design

Topic 5: How do we include branding considerations on the small screen?

Issues

1. How do you include brand so that it does not disturb user experience?
2. One application may need to display and/or promote several brands each with different requirements including:
   • Service providers
   • Operator
   • HW Manufacturer
   • Application Developer
3. Different brands may be promoted in different places on the application using different language and visual design
4. Different applications on the same device may promote brands in different ways including:
   • Interaction paradigm
   • Billing mechanism
   • Buying process

Solutions for Branding

1. If there is one brand for the application, use color or tiny logo to denote brand presence
2. If there are multiple brands, but the brand can be tied to a product or service displayed in the application, tie the brand to the product or service directly. For example, application logo for the main sponsor or product maker, a logo associated with an ad, a logo associated with a product on a merchandise page a logo associated with a service page like ticket buying. But design the graphics and interaction design so there is one coherent visual design.
3. Creating a good user experience enhances any brand. This is the most powerful branding device since color, images, and logo take up valuable real estate. So design from customer data.

Acknowledgements

This article reflects the effort of the three panelists who shared their experiences and then acted as discussion group leaders at the CHI 2004 SIG:

• Karen Holtzblatt, PhD — InContext Enterprises
• Pekka Ketola, PhD — Nokia Multimedia
• Thea Turner — Motorola Labs

Our thanks to everyone who attended the SIG