The field of design, in theory and in practice, is continuously evolving towards giving greater consideration to the needs, desires and expectations of consumers-users during product development and design. Themes that receive greater attention and weight include empathy with consumers or customers, collaborative innovation, and co-creation of value. They all drive towards firms inviting consumers to participate more actively and frequently in the development and creation of products and services — which eventually are intended to help them solve problems and perform daily tasks. It starts with listening, learning and understanding the viewpoint of consumers and taking it closely into account when devising a solution in whatever form found suitable (e.g., physical, virtual-digital). But furthermore, it involves the readiness to engage with actual and potential customers in two-way dialogue, collaboration and exchange of ideas for solutions (i.e., customers not only respond to product solutions ideated by designers but also are welcome to offer their own ideas and practical conceptions for solutions).
First of all, product design can benefit from research carried out to inform and guide the process of new product development (NPD), as design is an integral part of an NPD project. A marketing approach inherently calls for more elaborate and intensive reliance of NPD on information received from consumers-customers through research; this area has advanced and expanded especially since the 1990s, with newer and more sophisticated research methodologies being added. The role of consumers in research is relatively passive as they provide input about themselves and their responses to product or service options suggested to them but do not actively participate in a process of their creation. Consumers-customers may provide input with regard to their needs, lifestyles, preferences, expectations, feedback about their usage experiences, and more. The theory of Jobs to Be Done  offers a broader and more real-life perspective, suggesting that instead of talking about amorphic needs consumers would find it more natural to talk about the jobs or tasks they try to perform, the problems they face, and what could help them in accomplishing the goals of their jobs.
Surveys constitute an efficient and useful research mode for data collection from consumers; however, NPD should not restrict itself to types of input received through surveys since the information they can provide is incomplete. A range of methodologies highly relevant and helpful for NPD, and product design thereby, include conjoint studies (measuring preferences), experiments (lab and field), simulations (physical and virtual), and observations of consumer/customer behaviour in their natural environments and daily activities (e.g., for learning about consumer jobs, identifying problems in application of design thinking).
Furthermore, information obtained by research can and should be utilised throughout an NPD process, from early stages (e.g., exploratory study) to later stages (e.g., testing and refining semi-finished candidate product solutions). Most notable is a research programme constructed by Dahan and Hauser at Sloan School of Management (MIT), titled the ‘Virtual Customer’ . They proposed a toolbox of methods, qualified for different stages and tasks in an NPD project. For example, an application of web-based conjoint analysis with a more restrictive set of features in a fixed design is appropriate for initial examination of consumer preferences, while a more advanced conjoint model using an adaptive design would help to study preferences in greater detail among configurations of highly-specified attributes or features. Some methods are more creative or game-like, such as user design (UD) in which participants construct ‘ideal products’ or virtual concept testing (VCT) where participants “buy” from among competing, integrated product concepts, presented in rich-media format (e.g., visual model images that can also be rotated) with product profiles.
- Sawhney, Verona & Prandelli  highlight the great facility provided by the Internet (especially using Web 2.0 capabilities) for carrying out research and interacting with customers to engage them in collaboration for the advancement of innovation. They argued that in a virtual environment of the Internet a firm can engage large numbers of customers (reach) without significantly compromising the richness of the interaction. Methods can be sorted by stage of the NPD process and nature of collaboration (the authors do not strictly separate between research methodologies and non-research interactions); for illustration:
|Early Front-End Stages: Ideation & Concept||Later Back-End Stages: Product Design & Testing|
|Deep / High|
|Virtual communities; advisory panels||Toolkit for users innovation|
|Broad / High Reach||Web-based conjoint analysis||Web-based prototyping; Virtual concept testing|
In schemes of active collaboration customers contribute ideas, practical suggestions, and even designs and early models (‘mock-up’) for product solutions to a company. It means that a company listens not only to the customers’ personal concerns, preferences, experiences etc. but also listens to the suggestions, insights and models they can offer (e.g., for physical goods, packaging, software/apps & video games). There are different channels and schemes open to two-way communication and interactions between a firm and its customers (and other possible contributing stakeholders). We need to distinguish, however, between participation of customers in co-creation of products for their own use and personal benefit, and contributions made to future products for the good of many other consumers. The level of participation and contribution varies depending on the knowledge, skills and talent of the consumers.
In the more advanced type of collaboration, customers-consumers with general knowledge are likely to contribute mainly feedback based on their personal usage experiences and more abstract ideas, yet more knowledgeable and skilled customers, such as serious hobbyists, may offer more practical suggestions and take part in creation of product design plans and models. Schemes may range from discussions in company-supported forums (e.g., social media, dedicated & protected website) to toolkit applications that assist motivated users to construct model designs based on their ideas and creative talent. These advanced forms of collaboration are usually undertaken as implementation of a company-led initiative of open innovation, that is being open to practical and working design ideas from external contributors (e.g., serious hobbyist customers, free-lance professionals in design and relevant product domains). Schemes of mass customisation and self-design provide a different kind of opportunity for consumers-customers to take part in design activities and co-creating value of their own products. The design options applied by customers are often simpler, focused on visual design. But companies can open-up options for functional and more technical design choices for the more sophisticated customers. In some cases the more impressive and promising designs, in visual form and appearance as well as functional, may be adopted by a company and get incorporated in new product versions for larger customer audiences. Consumers who participate is such ventures seek the opportunity to take part in shaping the product they intend to use, and enjoy a feeling of gratification and accomplishment from engaging in designing the product item (an effect known as “I designed it myself”).
The principle of user-centered design is rooted in the approach of Design Thinking. A ‘suitable solution’ is apparently interpreted by the leading global design firm IDEO as a solution that achieves a good balance between desirability (by consumers), feasibility (technical) and viability (economic) (IDEO Design Thinking). Models of the design process, such as the five stages of design thinking (Empathise, Define, Ideate, Prototype, & Test) or the phases of design proposed by IDEO, emphasise that the process should be non-linear, iterative, and allow design teams to return to earlier stages, refine problem definitions or solutions, update prototypes and re-test. Principles that seem clear to us now were not so obvious several decades ago (when a product/engineering-driven approach was more dominant). Greater reliance on information from consumer research and collaboration with customers in different forms adds dimensions of knowledge and insight to the design process. Incorporating the information or contributions from customers may subsequently modify and even change course of a design process to achieve better solutions, and deliver increased value to customers.
Design processes have long been human-intensive, wherein the talent, ingenuity and involvement of skilled designers have been considered essential. Nevertheless, the multitude of design tasks arising all the time, the dynamic creation of personalised product and service versions, and the shorter product lifecycles, increase awareness to the limitations of current practices and thereon acceptance of assistance from advanced computer-based and data-rich methods and tools of artificial intelligence (AI). The shift may be accelerated by the change in nature of product development processes from a stand-alone NPD project carried out for a given goal and mission (with explicit start and end points) to an on-going development and design process where updates and adaptations are continuous and incremental. Hence, improving and better fitting products or services to customers’ needs and revealed preferences never end. Verganti, Vendraminelli and Iansiti  sketch outlines to this transformation in computer-assisted design to actually computer-executed autonomous design.
Verganti and his colleagues suggest dividing tasks of design, following principles of design thinking, between humans and algorithms of AI: commissioning human designers with identifying problems, selecting the most relevant ones, and devising a strategy and approach for problem solving (configuring the algorithms of ‘problem solving loop’), but then delegating the implementation of creative problem solving in design to AI algorithms. Humans would be in charge of ‘sense making’ of the design mission and process (i.e., “to understand which problems make sense to be addressed”). They argue that applying AI does not sacrifice or undermine key principles of Design Thinking: people-centered (innovation driven by empathy with users), abductive (making hypotheses about what things might be), and iterative (learning, creating, and testing); yet, the practice of design — how design is executed — may change considerably. Actually, they add, AI can enhance the application of those principles more successfully in dynamic environments. Artificial intelligence can enable a different process of design: instead of performing separate repeating sequences of design -> make -> use (with insights transferred between sequences), design (a human-led creative phase) leads to a ‘problem solving loop’ of make -> use empowered by sensors, data and AI. It means that after setting a creative course for a product solution, the detailed development process with many decisions that have to be made (e.g., shape, features, functions, interface) can be undertaken by AI algorithms, kept continuously and highly informed about user behaviour (AI learns, makes predictions on subsequent behaviour, and updates designs). Hence, the authors propose that “solutions are continuously improved and innovated”.
Advantages of the methodology described by Verganti et al. would primarily be in satisfying the needs and preferences of huge numbers of users in very short time intervals, as near to real-time as possible. In digital and online environments the AI-empowered system seems to make mostly refined and intricate modifications of design in adaptation to patterns of users’ behaviour it tracks and learns (e.g., composition of interfaces, information presented, image content and visual appearance; examples given from Airbnb and Netflix). With physical products the implementation can be more difficult: The approach is to use information collected from cameras and other types of sensors on performance and user behaviour for devising design improvements; sensors may be installed for a while before the information is practically utilised in real-time to adjust features and performance of the product to best fit preferences and behaviour of the user (e.g., Tesla cars). The plan set by Verganti, Vendraminelli and Iansiti for application of AI in design raises some critical questions: (1) As practices change so extensively, might it not change the theory and principles of design thinking, and what it means to design? (2) How justified is it to delegate the design of solutions from human designers to AI while assigning the designers with designing problem solving loops? Consider what impact it could have on creativity, art, and genuine originality of design.
We empahsised above physical products as objects of design. However, methods of design, and approaches of design thinking in particular, may also be applied to non-tangible commercial services, covering processes of service, procedures for handling different situations during interactions with customers, and configuration and visual design of interfaces for self-service (as seen with regard to application of AI). Moreover, Tim Brown, president and CEO of IDEO, suggested in an interview to Fortune magazine (2019) that design thinking can and should be employed to social services and systems and to institutions (e.g., school education, voting). He argued that “design thinkers have the responsibility to understand the outcomes they are designing for”, and that includes making intentional choices about how technology shall be used in the service of humanity. Browns sees a necessary shift for designers “learning to think not only in terms of stand-alone product but also of systems, the complex networks of meaning, behavior, and power within which products are embedded” .
There is a lot to gain for companies from familiarising with and more deeply understanding their customers, what solutions (products or services) could help them achieve their goals, and how they use them to accomplish their tasks. Information can be learned from research methods and from schemes of collaboration, which can help to support and guide the NPD or design process in different stages and conditions. Furthermore, in the context of AI, knowledge about customers is obtained through on-going tracking of their usage behaviour. The information utilised and practices applied can influence and modify the design process on different levels, from its outcomes to objectives and principles, and to what it means for the profession of designers. Design teams need to find a sensible balance between insights obtained, technology deployed, and their own judgement and talent in the field to be innovative while creating most approving solutions for their customers-end users.
Ron Ventura, Ph.D. (Marketing)
 Competing Against Luck: The Story of Innovation and Customer Choice; Clayton M. Christensen (with Taddy Hall, Karen Dillon, & David S. Duncan), 2016; Harper Business.
 The Virtual Customer; Ely Dahan and John R. Hauser, 2002; The Journal of Product Innovation Management, 19, pp. 332-353
 Collaborating to Create: The Internet as a Platform for Customer Engagement in Product Innovation; Mohanbir Sawhney, Gianmario Verona, & Emanuela Prandelli, 2005; Journal of Interactive Marketing, 19 (4), pp. 1-14
 Design in the Age of Artificial Intelligence; Roberto Verganti, Luca Vendraminelli, & Marco Iansiti, 2020; Harvard Business School: Working Paper 20-091
 The New Blueprint; Tim Brown with Barry Katz, Fortune (Europe Edition), March 2019, pp, 41-43 (Brown is currently chair of IDEO).