Published by the IEEE CS n 1536-1268/08/$25.00 2008 IEEE PERVASIVE computing 81
Standards & Emerging TechnologiesEditor: Sumi Helal n University of Florida n email@example.com
Bridging the Gap between Research and IndustryBrian David Johnson
To raise new questions, new possibilities, to regard old problems from a new angle, requires creative imagination and marks real advance in science. Albert Einstein
R evisiting old problems and ask-ing new questions, discovering new possibilities, and creating new approaches is the cornerstone of inno-vation, and it not only advances sci-ence but also fuels industry develop-ments. Although innovation occurs daily in universities and research groups around the world, sadly, much of this great work gets written off as mere theory, with little application to the wider high-tech industry. Innova-tion is essential to high-tech compa-nies and their development processes and business models, yet these com-panies often miss research that could have led to the next big thingor at least to intelligent and thoughtful improvements to current technologies. The tragedy of this gap isnt that the innovations arent applicable to indus-try but that companies and develop-ment teams lack a context for imple-menting them.
To better illustrate this need to incorporate theoretical and scientific innovation and execute it in industry, let me recount a recent talk I attended. Then, Ill introduce the Consumer Experience Architecture (CEA), a new framework for applying scien-tific insights to product and service development.
IllustratIng the gap: MobIle MappIng r&D Last September, in Ulm, Germany, the 2007 Intelligent Environments confer-ence was in full swing. The auditorium buzzed with excitement and the fresh ideas that can only come from the worlds great universities. I sat in the third row of the auditorium and lis-tened to the papers.
Katherine Willis from Bauhaus University was presenting her paper, Understanding Mobile Spatial Inter-action in Urban Environments.1 It was a fascinating paper that explored the differences in the effect of tradi-tional versus mobile maps (a map on a mobile phone or Internet-connected device). The results were unexpected. Her team showed that the people with traditional maps found their way bet-ter and faster than those who carried a mobile phone or GPS map.
Coming from industry, I was already envisioning the broader implications from this for consumer products and servicessuch as for smart phones, mobile Internet devices, and satellite navigation products. Excited by the promise, I asked what the team planned to do next. Williss response mirrored the teams future work section of the paper:
Further work is planned in order to provide a more detailed analysis of the experiment results. This will include an investigation of the effect of both route sequence (in the learning phase of the ex- periment), and the number of route legs between start and destination on the performance in the estimation task. Additionally study will be carried out that will reconstruct subjects cognitive maps of the various locations using combinations of estimates for each location.1
Essentially, the team planned to fur-ther analyze the results, conduct more research to confirm the results, and perform comparative field studies.
I was crestfallen. Although Williss response was theoretically sound, I had hoped to hear a more direct link between this interesting theoretical and experimental work and its appli-cation to the wide range of connected devices that have become so popular. The teams findings could impact how mobile mapping applications and services are developed and presented to consumers. Potentially, it could improve services and help people more quickly find their destination.
82 PERVASIVE computing www.computer.org/pervasive
STandaRdS & EmERging TEchnologiES
STandaRdS & EmERging TEchnologiES
As it was, there was no way for the team to contextualize their results in such a way that industry development teams could directly use those results. It was an unfortunate illustration of the gap between theory and industry.
the ConsuMer experIenCe arChIteCture To bridge this gap, the User Experience group inside Intels Digital Home group has developed a framework for apply-ing scientific insights to the product and service development process (see Figure 1).
CEA is a standardized methodology for developing products that accommo-dates multiple inputs during the prod-uct design process. Such inputs include computer science research and early development, theoretical computing, and social science research, along with traditional market analysis, demo-graphic profiles, technological surveys, and competitive analysis. Additionally,
the framework helps companies iden-tify, document, and validate specific experience metrics, derived from these multiple and varied inputs. CEAs holistic approach to product develop-ment lets the integrated development team not only gather and use inno-vative research but also validate the application of these ideas throughout the development process.
At the metalevel, we can divide the CEA framework into four stages, each serving as a key point of intersection between research and the industry development team. Additionally, each stage produces a set of standardized documents, practices, and workflows to inform the industry-accepted prod-uct- and service-development process.
Stage 1: Experimental and theoretical insightsThe initial information-gathering stage provides input into the planning cycle, and you can gather research and
insights from any number of sourcessuch as from Williss research papers. The goal is to determine who your consumers or users might be along with what they value and how they understand your product or applica-tion. At Intel, a team of anthropolo-gists, ethnographers, technologists, human-factors engineers, and design researchers work together to gather important and influential informa-tion. Then, the development team uses the early research and development to develop a deep and well-rounded vision of end-user needs and reservations. At this stage, insights into human behav-ior and needs as well as technological advances can influence a product or systems early design.
Using Williss findings, a mapping service provider could incorporate the fact that consumers found and under-stood their destination better with a traditional map as compared to a constantly updated mobile map. What effect would this have on how the development team decides to present a map to the service providers customer? Could a static map be more affective than a mobile map? Should the con-sumer have a choice? The development team would need to analyze these ques-tions, but Willis results certainly chal-lenge the notion that a mobile map is far more affective than a traditional static map.
Stage 2: Experience definition As the planning cycle moves forward and the product offering becomes more defined, the development team and research experts create a set of stan-dardized documents that outline the specific consumer experience with the product or service. These documents have a particular resonance, because theyre based on academic experi-mentation and real-world consumer insights.
This stage provides every member of the development team with a holistic understanding of the desired consumer experience. Consumer experience is
Whois your customer?
Whatis the experience?
Howwill you enable the experience?
Stage 1: Consumer research andethnographic insights
Market landscape overviewsDevelopment personas
Stage 3: Product definitionand development
Product value propositionsTechnical and marketing requirements
Stage 2: Usage model andexperience definitionConsumer experience
Stage 4: Collaborative development with multipletechnical and experience validation cycles
Revised product documentation
Reflect and iterateReflection and iteration occurs throughout development process
Figure 1. An overview of the Consumer Experience Architecture framework. As part of the consumer-experience development process, the development team asks who are the customers, what experience are those customers looking for, and how can the company enable that experience.
JUlySEptEmbEr 2008 PERVASIVE computing 83
STandaRdS & EmERging TEchnologiES
the sum total of multiple inputs related to the consumers understanding of a product. All these inputs provide a mental model to help the consumer better understand and use the prod-uct. Again, I point out Williss insight: how a consumer understands and uses a map should influence how a device or service presents a map. Consequently, a mental model of the consumer can help the development team overcome barriers that might otherwise lower a consumers adoption ofand satisfac-tion witha product.
From the technical developers to the marketing team, this knowledge about the consumer proves invaluable during the development cycles. It informs the design process and becomes a shared understanding between all members of the development team. It gives them a shared goal to which they can return when addressing wider techni-cal problems.
Stage 3: Early product definitionAfter the development team identi-fies the experience opportunities and maps the consumers experience, they must deconstruct these opportunities into usage models and value proposi-tions. Usage models are an industry-accepted standard format for devel-oping technology specifications and prototypes.2
From the experience opportunities and usage models, the development team and research and domain experts then develop the products value propo-sitions. These value propositions act as an expression of the product to the con-sumer, using their own language. Doc-umenting these value propositions in consumer-specific language is an essen-tial part of the framework. This stage also serves as a point of reflection and iteration. It lets the team make minor adjustments to the experience and prod-uct theyre developing. This articulation of the product experience can serve as a way to discuss the products attributes and value to people both inside and out-side the development team.
Stage 4: Production and validationThis is the longest stageand its the most complex to execute. Dur-ing product development and valida-tion, the development team applies a user experience (UX) validation pro-cess. This process employs a variety of systematic methods to evaluate and understand peoples perceptions of and experiences with the product. UXs targeted methods examine the user experience with concepts, pro-totypes, and functional product. UX isnt market research or focus-group testing; rather, its an assessment of peoples actual interactions with a prototype or product.
At each key milestone in the devel-opment process (such as the develop-ment of a prototype or release of an alpha or beta version), the develop-ment team uses UX to validate that the product addresses the original consumer experience goals. The test protocols for the UX validation are based on the core documents of the CEA. The experience specification describes the test environments and how the product should present itself to the consumer. The team can even test the value propositions to see if they do indeed offer value to the con-sumer and if the product is meeting the promise of these propositions. Likewise, the team can develop and test possible solutions to overcoming adoption and usage barriers.
The UX validation process pro-vides iterative feedback directly from the consumer regarding the products successes and failures. By perform-ing this validation process using early versions of the product, multiple times throughout development and basing
all stages on a consistent framework, UX lets the development team refine the product multiple times to meet the original experience opportunities out-lined for the product.
The results of the UX validation process are valuable to the develop-ment team. However, coupled with the experience documents from pre-vious stages, they also help provide a clear and compelling picture of the product, even before it has shipped. The results of the UX validation can provide clarity to upper management, possible partners, and the investment community.
T here is no greater tragedy than a brilliant idea doomed to obscu-rity. The CEA provides a framework for industry development teams to consider and apply the work and innovations of researchers and sci-entists. The frameworks practices, workflows, and documents provide a language for industry to collaborate with academia and research institu-tions, thereby creating a process for applying innovation to everyday prod-ucts and services.
1. K.S. Willis, C. Hoelscher, and G. Wil-bertz, Understanding Mobile Spatial Interaction in Urban Environments, Proc. 3rd IET Intl Conf. Intelligent Environments (IE 07), IEEE Press, 2007, pp. 6168.
2. T. Salvador, G. Bell, and K. Anderson, Design Ethnography, Design Manage-ment J., vol. 10, no. 4, 1999, pp. 912.
brian David Johnson is
the Consumer Experience
Architect for Intels Digital
Group. Contact him at
how a consumer understands and uses a map should influence how a device or service
presents a map.