3.1. Create tool: mind maps

Mind maps are graphical tools that organize and present information visually. Figure 2 shows an example, organizing the information in the Create category. Though the term ‘mind maps’ was popularized by Tony Buzan (ThinkBuzan n.d.), they have been used throughout history. The map has a central idea with ‘branches’ of related ideas. These grow branches of their own that continue to extend outward. Linked ideas on different branches can be connected.

This method of organization allows all ideas to be easily accessible and can help to draw connections between thoughts that may seem unrelated at first. Innovation Engineering recommends using mind maps for documenting individual and group association exercises. Kokotovich (Reference Kokotovich2008) found that mind maps are useful as design tools through the experimental introduction of mind mapping techniques to industrial design students. In this study, mind mapping techniques were introduced to first year industrial design students. The students used mind maps to help them generate individual project proposals, and maps and the creativity of these proposals was reviewed by judges, separately. The judges consistently rated work associated with higher-rated (high-quality, more complex) mind maps as more creative, and the results show a statistically significant correlation between the rating of the mind maps generated and the development of creative designs. Another study by Unalan (Reference Unalan2007) examined the ability of graphical organizational tools to convey information. Images of different organizational tools that displayed the same information were created and shown to subjects. The subjects then judged their clarity and ability to communicate creative content. Mind maps were generally ranked as the best graphical tool by the subjects.

3.2. Create tool: TRIZ

The Theory of Inventive Problem Solving, commonly known as TRIZ, is a systematic creativity tool. Genrich Altshuller and colleagues examined over two million patents to extract commonalities in innovation and formulate TRIZ (Wang, Chang & Kao Reference Wang, Chang and Kao2010). It provides problem-solving strategies to find creative solutions that would typically go unnoticed. The application of TRIZ relies on identifying contradictory goals or ‘elements’ in a design challenge, i.e. high strength and low weight. The component highlighted in IE Create is the TRIZ matrix, which maps the contradictions to a variety of engineering approaches (termed ‘inventive principles’ in TRIZ) to provoke creative designs. A simple, interactive TRIZ-matrix-lookup tool is available at the IE website as well as a number of unrestricted websites (TRIZ40 n.d.). In the IE program, when a challenge is not related to engineering, the elements are interpreted metaphorically.

Hernandez et al. (Reference Hernandez, Schmidt and Okudan2013) tested the effectiveness of TRIZ on engineering students at two universities separately using a test-versus-control design on within and between subject experiments. Exposure to TRIZ increased the average novelty per subject as well as the variety of ideas. Chulvi et al. (Reference Chulvi, González-Cruz, Mulet and Aguilar-Zambrano2013) compared the performances of TRIZ, SCAMPER (an intuitive idea-generation method), brainstorming, and a control. Seven teams used TRIZ, seven teams used SCAMPER, one team used brainstorming, and one team used no method. By judging the solutions that the teams provided to a design problem, Chulvi et al. (Reference Chulvi, González-Cruz, Mulet and Aguilar-Zambrano2013) found that (1) TRIZ outperformed SCAMPER and no method in terms of the novelty of the solutions, but TRIZ was not as good as brainstorming in this dimension; and (2) both TRIZ and SCAMPER had similar performances in terms of the utility of the solutions, and this utility was better than the performances of brainstorming and using no method. However, the inclusion of only one control team calls into question the robustness of these results. A study conducted by Birdi et al. (Reference Birdi, Leach and Magadley2012) helped identify the benefits of TRIZ in the workplace. Power systems engineers from a multinational firm who experienced a one-day TRIZ training workshop reported significantly higher levels of creative problem-solving skills, idea generation, and creative motivation than the engineers who did not attend the training, as measured by a follow-up survey many months (the average is 33 months) after the workshop. Over a period of three years, 41% of the trainees reported submitting one or more patent applications compared to 28% of the non-trainees. As part of the follow-up survey mentioned above, subjects were asked to generate as many original solutions as possible to a design problem. On average, trainees suggested two ideas each while non-trainees submitted only one. It should be noted that many events can take place during the time gap between the workshop and the survey to influence the outcomes of the survey. While investigating learning styles, León-Rovira et al. (Reference León-Rovira, Heredia-Escorza and Río2008) found that freshmen mechanical engineering students who were given an introduction to TRIZ were scored as more creative than those who did not learn about TRIZ. Ogot & Okudan (Reference Ogot and Okudan2006) reached a similar conclusion studying first year students in an ‘Introduction to Engineering Design’ course placed into groups. They were given a product design task in which the experimental course sections used TRIZ techniques while the control sections did not. The experimental sections generated more unique feasible ideas both per section and per group as compared to the control sections. The effect of using TRIZ in design teams was also examined by González-Cruz et al. (Reference González-Cruz, Aguilar-Zambrano, Aguilar-Zambrano and Colombel2008). Some sections of an undergraduate engineering design class were introduced to TRIZ while others used traditional brainstorming techniques. The sections using TRIZ methods generated 115% more unique design solutions than the others.

Arlitt et al. (Reference Arlitt, Nix and Stone2012) demonstrated that using a matrix implementation to present the TRIZ inventive principles did not improve their usefulness. The experiment included a control condition, in which the subjects used the correct TRIZ matrix, and an experimental condition, in which the subjects used a manipulated TRIZ matrix imbedded with randomly populated inventive principles. The subjects in both of the conditions used the given matrix to solve a design problem. Arlitt et al. found that using the correct TRIZ matrix had no significant advantage of facilitating idea generation over using randomly provided inventive principles.

3.3. Create tool: stimuli

Stimuli provide information to living organisms and directly influence their behavior by allowing the organisms to react accordingly (Merriam-Webster.com 2014c ). In the context of creativity, any input obtained from an outside source during the creative process can act to stimulate the mind in new ways, leading to new ideas and solutions. Finding stimuli is an important topic of IE under the general heading ‘stimulus mining’. Associated methods included: (1) insight mining, where the designer receives stimuli directly from the customer in the form of observations, conversations, opinions, and surveys; (2) market mining, meaning that ideas can come from current products that are available; and (3) technology mining, in which ideas are formed by examining emerging technology.

Fink et al. (Reference Fink, Grabner, Gebauer, Reishofer, Koschutnig and Ebner2010) investigated if creative cognition could be enhanced through learning the ideas of others. This study found that subjects in an alternative uses task were able to generate more unique ideas after they were presented with other subjects’ ideas than subjects who did not see others’ ideas. An experiment by Valacich et al. (Reference Valacich, Jung and Looney2006) used a computer-simulated group to provide subjects with different qualities of stimulation during an idea-generation exercise. Fifty ideas of varying quality, as judged by experts, were entered into a database. During the experiment, a computer program supplied these ideas to subjects in a manner that mimicked the way they would have been supplied by an actual group. They found that the highest performance in idea generation was associated with the highest quality stimuli from the database.

The Design Heuristic Cards created and used in a study by Daly et al. (Reference Daly, Christian, Yilmaz, Seifert and Gonzalez2012) provide stimuli for designers and improve ideation. The cards explain a design approach, identified from a large-scale study of academic and industry designers’ practices, and show two examples of the approach in use. In the experimental study, these cards were given to freshman engineering students along with a design task. The design solutions to the design task generated by students who used the cards were more original and better developed than those generated by students who did not use the cards, as rated by two judges using a Likert-scale assessment of creativity and diversity of solutions.

Some other studies, which do not involve statistical testing, also confirm the meaningfulness of using stimuli. In an industry-based study, Howard et al. (Reference Howard, Culley and Dekoninck2011) observed that stimuli were useful in maintaining idea-generation rates as brainstorming meetings approached their end. Feedback from the subjects confirmed that some stimuli could spark ideas and productive secondary conversations. In another study, Ocker (Reference Ocker2005) qualitatively analyzed 10 asynchronous virtual groups’ performances on initiating a computerized post office project. He found that the two groups that produced highly creative concepts both had stimulating colleagues and some initial creative example ideas. These colleagues and ideas either provided or served as stimuli that encouraged the whole group to generate creative ideas. Roy (Reference Roy1993), by interviewing individual designers and studying related materials, found that one designer would search for solutions from other areas or nature, which served as sources of stimuli.

Innovation Engineering emphasizes ethnographic studies as a source of customer insight, and academia widely accepts this approach as effective. Cooper et al. (Reference Cooper, Edgett and Kleinschmidt2002) found that companies conduct ethnographic research in a ‘Discovery Stage’ at the beginning of the product development process. An example they provided showed that by making site visits, Fluke Corporation received first-hand information about customers’ problems and needs, leading to the establishment of a very successful product line. Leonard & Rayport (Reference Leonard and Rayport1997) assert and support with examples that observing customers using products in their own environment offers five types of unique information: ‘Triggers of Use’, ‘Interactions with the User’s Environment’, ‘User Customization’, ‘Intangible Attributes of the Product’, and ‘Unarticulated User Needs’. This information can spark ideas for product concepts and improvements.

The concept of fixation presents a drawback to using stimuli to promote innovation. Design fixation occurs when a designer becomes familiar with an existing solution, and aspects of that solution then appear in their own designs. Linsey et al. (Reference Linsey, Tseng, Fu, Cagan, Wood and Schunn2010) investigated the topic of fixation with respect to engineering design. In their study, engineering design faculty members were given a design task and assigned to different task conditions. The experimental condition was given materials including an example design, which specifically violated one or more design constraints. Subjects in the experimental condition produced fewer unique solutions as compared to the control condition that received no example design. Moreover, the fixation caused by the example was so strong that these experienced designers submitted solutions that violated the same constraints.

3.4. Create tool: associations

As it relates to mental processes, association is defined as ‘the process of forming mental connections or bonds between sensations, ideas, or memories’ (Merriam-Webster.com 2014a ), which is useful for generating creative ideas. Associating stimuli that are not obviously related can result in new ideas. Innovation Engineering suggests that new ideas can be combinations of old ideas. It discusses two types of associations: free and forced. Free associations are defined as those made between closely related stimuli, such as apples and grapes. When unrelated stimuli, such as a clock and an elephant, are introduced as brainstorming stimuli to find commonalities, they are referred to as forced associations. Innovation Engineering suggests that free association leads to more ideas than forced association, but the ideas generated by forced association are generally more unique.

Association warm-up exercises were found to improve creative-thinking ability during a study by Freedman (Reference Freedman1965). Subjects in the facilitation condition were given a stimulus word and then asked to speak associated words for 30 seconds, while those in the non-facilitation condition were asked to define stimulus words but not generate associated words. On a subsequent creative-thinking test, male subjects who completed the association exercises earned an average score of 18.05 out of 30, which was significantly higher than the average score of 14.15 of the non-facilitation condition. The difference in female subjects’ scores was not as large as that for male subjects, but the trend was similar. Subjects in a study by Coskun (Reference Coskun2009) generated more ideas during a brainstorming session after performing an exercise regarding closely linked word pairs. The subjects who worked with distantly linked word pairs generated fewer ideas. This finding suggests that free association may be more effective at generating larger volumes of ideas than forced association, but it does not speak to the quality or originality of these ideas. Miller et al. (Reference Miller, Russ, Gibson and Hall1970) found that free association training improved subjects’ abilities to discover additional links between stimuli on the Remote Associates Test (RAT; Mednick Reference Mednick1962). Subjects in the experimental condition were given 30 seconds to free-associate additional words for each of 10 stimulus words. To ensure that simply being exposed to the words was not enough to significantly affect scores, each subject in the control group read the words generated by a randomly selected subject in the experimental condition. Cheng et al. (Reference Cheng, Wang, Liu and Chen2010) found association instruction directly affected creativity. After a creative-thinking pretest determined that there were no significant differences between fourth grade students, both free and forced association training was given to an experimental condition over a five-week period. This training consisted of making associations between numbers, images, stories, and music, for example, the association between the number 14 and a windsurfer. After the treatment phase, all students wrote poems to be judged on creativity. The poems by students in the experimental condition were consistently judged as more creative than those from the control.

3.5. Create tool: groups

It is commonly thought that working in groups can positively affect creative productivity, as is emphasized by IE. More people working on a common task means more opinions, experiences, thoughts, and neurons influencing the final product.

By comparing the quantities of ideas generated by groups and individuals, Paulus & Yang (Reference Paulus and Yang2000) determined that groups could outperform individuals and nominal groups (pooled individual data) in idea generation. Using a brainwriting exercise, students sharing their ideas with others in a group led to the production of more unique ideas than students working individually or pooled individuals. Another finding from this study is that ideation performance is inhibited when the group members are asked to memorize the ideas generated by teammates. Different levels of group interaction during brainstorming activities were examined by McGlynn et al. (Reference McGlynn, McGurk, Effland, Johll and Harding2004). University students were assigned either high-interaction groups (working face to face) or low-interaction groups (working individually but in the same area). Low-interaction groups generated more ideas than high-interaction groups, but the ideas of the high-interaction groups were of higher quality, in terms of plausibility and correctness as rated by research assistants. The high-interaction groups’ consensus solutions were also rated as more plausible than the group members’ individual responses. The poorer performance of low-interaction groups on generating high-quality ideas was also observed by Ocker (Reference Ocker2005) in his qualitative research on virtual groups, mentioned in Section 3.3. He found that the group that had few to none discussions and provided little feedback on group members’ inputs had very few original ideas. Coskun (Reference Coskun2011) studied the effect of group size and session length on brainstorming activities. Groups of 4, 6, 8, 10, and 12 people were given an idea-generation task and allowed to work for either 15 or 20 minutes. The number of unique ideas generated by each group increased proportionally with the size of the group, but the size did not affect the performance of the individuals who comprised the group, suggesting no synergistic or combinatorial effects. Pirola-Merlo & Mann (Reference Pirola-Merlo and Mann2004) identified that even though a majority of the variance in group creativity was accounted for by the average group member creativity, a systematic variance still existed in group creativity after controlling for individual creativity. This means that even though the individual group members’ creativity accounted for most of the group’s creativity, there was evidence of improved creative thinking due to group interaction. Taggar (Reference Taggar2002) found that working in a group can help to improve creativity if the members engage in behavior that encourages the open sharing of information. However, if group members are inadequately trained or if the group is too large, group processes can stifle creativity. This suggests that ideal creativity-fostering groups are relatively small with members who understand the importance of their behavior.

Contrary to the above findings, some previous literature (Diehl & Stroebe Reference Diehl and Stroebe1987; Mullen et al. Reference Mullen, Johnson and Salas1991; Thornburg Reference Thornburg1991; Larey & Paulus Reference Larey and Paulus1999) has shown evidence that brainstorming groups have poorer creativity performance than nominal groups. Diehl & Stroebe (Reference Diehl and Stroebe1987) examined a number of possible ways that group brainstorming could negatively affect the productivity of the members. The data they collected suggest the existence of blocking, or the inability to share ideas because another member was sharing, as a possible hindrance to group idea generation. A blocking delay between generating and sharing an idea may prevent new ideas from fully forming.

3.6. Create tool: dual brain

Innovation Engineering discusses that creative and impulsive behavior originates in the brain’s right hemisphere while logical and analytical behavior stems from the left. Although this hemispheric distinction does not actually exist, the terms ‘right brain’ and ‘left brain’ have become figures of speech pertaining to creative and analytical personality types, respectively. Innovation Engineering refers to the right brain and the left brain as descriptions of personality with remarks such as ‘logical left brain’ and ‘radical right brain’. ‘Whole brain’ thinking is also discussed, with the ability to combine the strengths of both thinking types and obtain the best results.

Harkins & Macrosson (Reference Harkins and Macrosson1990) tested if exercising the right hemisphere with a 10-week drawing course could enhance creativity. The experimental condition performed significantly better than the control condition in two dimensions as measured by the Torrance Test of Creative Thinking. Mihov et al. (Reference Mihov, Denzler and Förster2010) conducted a meta-analytic review of the literature that addressed relationships between creativity and hemispheric dominance. They provided statistical results showing that creativity was more likely to be associated with right-hemispheric than left-hemispheric activation.

Kowatari et al. (Reference Kowatari, Lee, Yamamura, Nagamori, Levy, Yamane and Yamamoto2009) conducted functional magnetic resonance imaging (fMRI) scans on both expert and novice designers during a design task to find that expert designers showed a bias toward right-hemisphere activation while the novices experienced a more bilateral response. Many experts showed a negative change in blood oxygenated level dependent signals in the left hemisphere. Both fMRI and electroencephalogram (EEG) measurement methods were used by Fink et al. (Reference Fink, Grabner, Benedek, Reishofer, Hauswirth, Fally, Neuper, Ebner and Neubauer2009) to record brain activity during creative-thinking tasks. Subjects working on tasks such as generating alternative uses, describing object characteristics, inventing names, and completing words experienced whole brain activation rather than a bias to either side. Abraham et al. (Reference Abraham, Pieritz, Thybusch, Rutter, Kroeger, Schweckendiek, Stark, Windmann and Hermann2012) examined brain activity associated with conceptual expansion and found that the regions involved were those responsible for the retention, retrieval, and integration of conceptual knowledge, which are strongly lateralized to the left hemisphere. Howard-Jones et al. (Reference Howard-Jones, Blakemore, Samuel, Summers and Claxton2005) identified the brain area involved in approaching a creative story generation task as the right prefrontal cortex. In general, the literature suggests that the notion of left/right brain thinking is not accurate. While some creative processes showed hemisphere bias in brain activity, others caused whole brain activation. This activity is influenced by the experience subjects have with the creative activity being studied. Also, while certain creative processes may elicit the activation of a particular area of the brain, other areas of the brain are necessary to utilize these processes.

3.7. Create tool: humor

Humor is one of the many ways that people can interact with one another, and it is important in maintaining healthy interpersonal relationships (De Koning & Weiss Reference De Koning and Weiss2002; Graham Reference Graham1995). Humor also helps in enhancing a person’s self-confidence, and self-confidence can benefit creativity (Kelley & Kelley Reference Kelley and Kelley2013). Self-confidence is associated with self-efficacy, and the willingness to take risks in design. Highly creative design solutions often involve an element of risk. Innovation Engineering uses humor and satire to demonstrate points. Beyond applying to innovative idea generation, it was noted that humor at work will increase the productivity and mood of employees because it leads to a more enjoyable experience.

To identify the relationship between humor and creativity, Humke & Schaefer (Reference Humke and Schaefer1996) had subjects complete humor and creative-thinking measures. They found a Pearson correlation coefficient of 0.77 between scores on the Multidimensional Sense of Humor Scale and the Franck Drawing Completion Test. Rouff (Reference Rouff1975) found a positive correlation between a subject’s abilities to identify humorous incongruities and RAT scores. Treadwell (Reference Treadwell1970) found significant positive correlations between ‘humor creation scores’, measured by a Cartoon Test, and three creativity scores: RAT scores, Gestalt Transformations test scores, and Novelty of Productions scores. Ziv (Reference Ziv1976) investigated the relationship between laughter and creative thinking. A humorous recording, selected due to its ability to produce high amplitude and long duration laughter in subjects listening to it, was played for subjects in an experimental condition prior to a Torrance Test of Creative Thinking. The subjects who listened to the recording performed significantly better on the test than the control condition that did not. In Ocker’s (Reference Ocker2005) observation of virtual groups’ performances, he noticed that witty but task-related humor was involved in the two most creative groups’ internal interactions, which suggests positive effects of humor on creativity.

4.1. Communicate tool: benefit promise

Benefit promise is the IE term for the specific benefit that an innovation will provide to the audience. The program stresses that it is important to tell people why they should care as opposed to simply listing features of a product or service. A study by Cooper & Kleinschmidt (Reference Cooper and Kleinschmidt1993) showed the relative importance of communicating product benefits to product success. After examining 109 different products from chemical companies, they identified the traits that lead to successful products. Featuring a specific benefit of importance to the customer ranked as the fourth most important trait of success, behind traits that involved the product’s relative quality and price as compared to the competition.

Table 2. Literature related to the IE Communicate category

4.2. Communicate tool: simplicity

In communication, simplicity refers to targeting the amount and type of information being exchanged. More information communicated means more to remember. Innovation Engineering recommends that all communications be as simple as possible to allow the most important aspects of the message to be received and remembered.

It is documented in Miller’s review (Reference Miller1994) that people can only discriminate about seven levels of single-dimensional stimuli (for example, seven numbers), indicating the limited information processing capabilities of humans. While the general recommendation of simple communication is featured in general-audience scientific books (Heath & Heath Reference Heath and Heath2007; Maeda Reference Maeda2006) there was limited academic research identified by the authors. One reason for this could be that simplicity in communication is assumed to be common sense.

Some sources were located that refute the benefit of simple communications. Shedler & Manis (Reference Shedler and Manis1986) tested the effect additional but unrelated details have on the credibility of arguments. Subjects in two conditions heard equal numbers of opposing arguments about the fitness of a mother and were asked whether they thought the mother should retain custody of her child. One condition heard additional details for the positive arguments while the other condition heard additional details for the negative arguments. Even though the details were unrelated to overall paternal fitness, the additional information significantly affected the subjects’ judgments in both conditions. Another study by Macklin et al. (Reference Macklin, Bruvold and Shea1985) examined how the readability of a print advertisement affected a reader’s ability to retain important information. Different versions of a printed advertisement were designed and statistically verified to contain the same information and vividness but have different levels of readability. These were shown to subjects whose recall was subsequently tested. The experiment revealed that the readability did not affect the ability to recall ad claims, attitude toward the brand or ad, or purchase intent.

4.3. Communicate tool: tell the truth

Innovation Engineering advises that honesty in communications imparts trust between a company and its customers. The Edelman Trust Barometer (Edelman 2013) collected information from 26 000 public subjects in 26 countries to identify attributes that encourage trust in companies. Honest communication from companies was identified as one of the most important attributes. Trifts & Haubl (Reference Trifts and Haubl2003), Corbitt et al. (Reference Corbitt, Thanasankit and Yi2003) and Jarvenpaa et al. (Reference Jarvenpaa, Tractinsky and Vitale2000) all show that the perceived trustworthiness of an Internet vendor is positively correlated with the probability that customers choose to purchase from those vendors. A mail survey conducted by Lau & Lee (Reference Lau and Lee1999) identified that brand performance is a strong predictor of trust in a brand, and that trust in a brand contributes to brand loyalty. Subjects in a study by Settle & Golden (Reference Settle and Golden1974) were shown advertisements where a new product was compared to the current best-selling alternative. One study condition was shown a version that claimed the new product was superior in five of five areas while another condition was shown a version that only claimed superiority in three of five areas. The subjects shown the latter reported a higher believability for the advertisement. Anderson (Reference Anderson1973) investigated the relationship between expectations and product performance; when subjects were given realistic expectations of a pen’s performance, they rated its performance higher than when given overly positive expectations.

One study offers insight into how much honest information should be supplied. Liberali et al. (Reference Liberali, Urban and Hauser2013) found that proactively supplying information that made brands vulnerable to criticism did not increase subject trust in the brand. Honesty should be paired with simplicity of communications in a strategic manner.

4.4. Communicate tool: repetition

Innovation Engineering advises that repeating a product message increases memorability. D’Souza & Rao (Reference D’Souza and Rao1995) tested the effect of repeating a specific advertisement relative to that of competitors. They found that higher repetition rates resulted in significant increases in top-of-mind awareness, brand preference, and brand choice. Ray & Sawyer (Reference Ray and Sawyer1971) exposed female shoppers to slides of advertisements with varying repetition rates, and then the shoppers took a product survey. Repetition rate had the most significant effect on advertisement recall, with an almost 40% increase when comparing a single exposure to four exposures. Purchase intention for the products shown also increased significantly. The significant effect of repetition on information recall is also supported by the literature in psychology (Cacioppo & Petty Reference Cacioppo and Petty1979; Mayer Reference Mayer1983). Mayer (Reference Mayer1983) found that repetition not only increased the amount of recalled information that was related to the provided recordings, but also had effects on what was remembered. The recall of conceptual principles benefited from repetition more than the recall of details such as concrete analogies and facts.

Cacioppo & Petty (Reference Cacioppo and Petty1979) found that message repetition increased the amount of information that was recalled and that as the number of repetitions increased, subject agreement with the repeated statement first increased and then decreased. This suggests that there may be an oversaturation point in message repetition that could have a negative effect on innovation communication.

4.5. Communicate tool: provide proof

Providing proof presents potential customers with concrete evidence to support a marketing claim. Innovation Engineering examples of proof include test results, company pedigree, testimonials, and guarantees. Innovation Engineering advises that more proof is better as long as the proof is meaningful.

Liberali et al. (Reference Liberali, Urban and Hauser2013) tested if access to direct product experience, print and online resources, word of mouth, or recommendations from trusted advisors would alter customer brand preferences. The study found that providing direct product experience and testimonials significantly improved brand consideration and trust in a brand. East et al. (Reference East, Hammond and Lomax2008) used mail-in surveys to measure the change in purchase intent after exposure to positive and negative word of mouth. The results showed that positive word of mouth improved the subjects’ probability of purchase by an average of around 50% while the negative word of mouth decreased the average purchase probability by 25%. Multiple references from Section 4.3 (Corbitt et al. Reference Corbitt, Thanasankit and Yi2003; Jarvenpaa et al. Reference Jarvenpaa, Tractinsky and Vitale2000; Trifts & Haubl Reference Trifts and Haubl2003) explain that customer trust improves brand consideration; this trust can be affected by the proof that is available.

Bearden & Shimp (Reference Bearden and Shimp1982) considered product warranty as an extrinsic cue that would influence customers’ perceptions of innovative products. By constructing models of customers’ risk perceptions and overall affect toward two new product concepts, they proved that a warranty mediated customers’ affective responses to innovative products via perceptions of risk. Thorelli et al. (Reference Thorelli, Lim and Ye1989) found a significant effect of warranties on three measures of product evaluations: perceived quality, overall attitude, and purchase intention of the product. Boulding & Kirmani (Reference Boulding and Kirmani1993) found that a better warranty helped a high-credibility firm to achieve better customer perceptions of product performance and overall quality, as well as to improve purchase intent. But this was not the case for a low-credibility firm, which suggests an interaction with trust. Fu & Chen (Reference Fu and Chen2011) present mixed findings on offering guarantees. In a study examining online auctions, they identified that while offering a refund guarantee did increase the price premium of the items up for bid, it did not affect the time until the first bid, number of bidders, or number of bids. This means that the customers were willing to pay more knowing they were guaranteed a refund, but the guarantee did not attract more interest or additional customers. In the psychology literature, Maddux & Rogers (Reference Maddux and Rogers1980) found that people who provided supporting evidence while arguing a position were significantly more likely to persuade a subject, independent of the arguer’s expertise and physical attractiveness.

4.6. Communicate tool: include a number

Innovation Engineering suggests that including numbers makes for concrete and understandable communications; for example, ‘this toothbrush design removes three times as much plaque as the leading brand’. The number three in this marketing message gives the audience a value to place on the product.

Holbrook (Reference Holbrook1978) had subjects read and rate advertisements that were either objective (i.e., ‘27 miles per gallon’) or subjective (‘truly excellent gas mileage’). Objective advertisements were statistically associated with higher brand attitude and claim belief than subjective ads. Rossiter & Percy (Reference Rossiter and Percy1978) had customers rate a product after seeing its advertisement. Subjects who saw ads that included concrete information, such as ‘Bavaria’s number one selling beer for the last ten years’, responded more favorably to the product and the brand than those who saw less concrete statements like ‘Bavaria’s finest beer’. Davis (Reference Davis1993), in his literature review, strongly suggested the use of numerical data and specific statements in environmental claims.

It is also important to include numbers and be specific when communicating internal product development goals. Six Sigma is a strategy used in industry to ‘improve profitability, to drive out waste in business processes and to improve the efficiency of all operations that meet or exceed customers’ needs and expectations’ (Antony & Banuelas Coronado Reference Antony and Banuelas Coronado2002, p. 24). Design for Six Sigma requires the identification of numerical customer requirements and engineering specifications (Mader Reference Mader2002). These specific goals can improve the performance of the project as proposed by Linderman et al. (Reference Linderman, Schroeder, Zaheer and Choo2003) and can also facilitate communication within a design team. Similarly, House of Quality, which is a ‘basic design tool of the management approach’ (Hauser & Clausing Reference Hauser and Clausing1988, p. 63) contains steps for identifying specific customer requirements and engineering specifications. These steps help benchmark development activities and set target design values (Hauser & Clausing Reference Hauser and Clausing1988), and are particularly useful for interdisciplinary product design teams.

4.7. Communicate tool: tell a story

A good story captivates audiences both intellectually and emotionally, and IE advises their use in communicating marketing messages. Green & Brock (Reference Green and Brock2000) gave subjects a short story to read and told them that it was either fictitious or real. A post-survey identified that subjects who reported higher transportation by the story (deeper mental processing/emotional involvement) rated the story’s situation more realistic and the characters more favorably, whether or not they were told the story was real. The positive effect of arranging facts into a story was examined by Pennington & Hastie (Reference Pennington and Hastie1988). Subjects acted as jury members for a mock trial, and after being presented evidence, were asked to determine if the defendant was innocent or guilty. The study found that when evidence supporting one verdict was presented in a story-like order, the subjects were significantly more likely to pass that verdict and reported significantly higher confidence in their decisions. This shows that stories not only influence judgments and perceptions, but also improve the persuasiveness of an argument.

5.1. Commercialize tool: do one thing great

Innovation Engineering advises companies to identify a core benefit that they can offer customers, and use this benefit to drive the mission and focus of the company. This can mean, for example, eliminating product offerings that are auxiliary to this core benefit, or changing the manner in which product benefits are presented. Supporting this, Boatwright & Nunes (Reference Boatwright and Nunes2001) found that removing low-sales items from grocery store shelves increased average category sales. Although the reduction in options caused some customers to leave the retailer, the majority of customers whose preferred brand was cut adopted a different brand. Iyengar & Lepper (Reference Iyengar and Lepper2000) examined the effect of limiting the number of choices available. Subjects in a grocery market were offered free samples from a large selection of jams (24 options) or a reduced selection of 6 options. The large selection set attracted more subjects to investigate the offer, but produced fewer final sales than the limited set. Redelmeier & Shafir (Reference Redelmeier and Shafir1995) investigated how decisions were altered by the addition of new options. When presented a hypothetical situation in which subjects had plans to complete one activity but were then offered a single more attractive alternative, only 21% chose the original activity. Interestingly, when subjects were offered two better alternatives, 40% chose to follow through with the original activity. When given multiple alternatives, a sizeable number of subjects avoided making a switching decision.

Table 3. Literature related to the IE Commercialize category

In a report of an interview with Dell’s founder, Magretta (Reference Magretta1998) emphasized the importance of focus and specialization in Dell’s success. Dell did not decide to make every piece that was needed for a computer, instead purchasing well-made components from other companies. They focused on ‘delivering solutions and systems to customers’ (Magretta Reference Magretta1998, p. 74), creating unique value for its customers and developing the company.

One study located does potentially refute the idea of limiting offerings. Oppewal & Koelemeijer (Reference Oppewal and Koelemeijer2005) mailed surveys to customers that asked for opinions about a variety of flower assortments. When subjects were given more assortment options, the average evaluation across assortments significantly improved, in a linear manner. However, this study did not investigate how many subjects purchased one of the assortments.

5.2. Commercialize tool: predicting sales and selecting ideas

After refining innovative ideas, it is necessary to select ideas to focus on developing them, while putting other less-promising ideas aside. This selection process can be partially guided by advice, such as Do One Thing Great: if the idea strays too far from the core company mission, it may not be right for development. But, ultimately, the success of an idea must be linked to metrics such as predicted revenues. As Herstatt, Verworn & Nagahira (Reference Herstatt, Verworn and Nagahira2004) observed, companies use economic criteria to assess their innovative ideas. The selection can be done by experts with simple selection criteria (Ferioli et al. Reference Ferioli, Dekoninck, Culley, Roussel and Renaud2010), both objective and subjective, but involving explicit tools can better support the selection. Innovation Engineering recommends the Fourt–Woodlock equation (Fourt & Woodlock Reference Fourt and Woodlock1960) as a tool to aid in this evaluation. The equation, as it is presented by the IE program, is

$$\begin{eqnarray}AS=FDM\times TR\times FPR+FDM\times TR\times RR\times RPR\times NRP,\end{eqnarray}$$

where AS is the annual sales forecast, FDM is the number of final decision makers (potential purchasers), TR is the trial rate, FPR is the first purchase revenue, RR is the repeat rate, RPR is the repeat purchase revenue, and NRP is the number of repeat purchases. It should be noted that these variables are not the same in all representations of the equation, but the meaning of the variables used does not change markedly from one version to another.

To create this equation, Fourt and Woodlock used trial rate and repeat ratios based on observed test markets and real purchase data, and made a number of assumptions about the consistency of the environment in which the products would be sold. The assumptions were that the distribution, promotional expenditures, prices, competitive activity, product, and packaging would not change substantially during the time of the estimate. In situations where these assumptions held and the rates could be determined by an initial test sample, this method predicted volume changes within 5% of the actual values. If the prediction shows that a product will bring substantial profit, a company can justify mass production. Conversely, if the forecast implies a poor return, the company can edit the offering in an attempt to improve the prediction, or scrap the offering altogether. As presented by IE, the equation did not include rates determined by an initial test market or stress on any of the stated assumptions. However, in situations observed by Fourt and Woodlock where one or more of the assumptions were violated, the estimate became much less accurate. Innovation Engineering also suggested estimating the trial and repeat rates, which will likely lead to further loss of accuracy.

Test markets have been used for many years, and forecasting models have been generated to predict new product performance; see Parfitt & Collins (Reference Parfitt and Collins1968), Eskin (Reference Eskin1973), Nakanishi (Reference Nakanishi1973), Jain, Mahajan & Muller (Reference Jain, Mahajan and Muller1995) and Decker & Gnibba-Yukawa (Reference Decker and Gnibba-Yukawa2010). These models, however, are more complex than the Fourt–Woodlock equation, and thus more difficult for a general audience to adopt. The Bass diffusion model (Bass Reference Bass1969) for adoption rate prediction has been studied substantially and modified to accurately predict specific adoption rates for specific markets (Bass, Krishnan & Jain Reference Bass, Krishnan and Jain1994; Satoh Reference Satoh2001; Li & Yang Reference Li and Yang2006; Diao, Zhang & Jia Reference Diao, Zhang and Jia2009; Tseng & Hu Reference Tseng and Hu2009). However, this model assumes that the innovation will end with complete adoption across a population, like personal computers or television.

The authors offer alternative approaches for selecting innovative ideas. Multi-Attribute Utility Theory (MAUT), based on Utility Theory in economics, enables idea selection by comparing overall utility of each idea in decision makers’ perspectives (Yates Reference Yates2003). The ideas are first evaluated in terms of multiple criteria respectively with corresponding utility functions; then the overall utility of an idea is usually obtained in an addictive manner for further comparisons. Applications of the MAUT include comparing an innovative parabolic solar cooker against existing cooking devices in India (Pohekar & Ramachandran Reference Pohekar and Ramachandran2006) and optimally positioning a product innovation in an existing market (Gavish, Horsky & Srikanth Reference Gavish, Horsky and Srikanth1983). Choice share (amongst potential customers) of innovations can also be estimated using customer-survey-based methods such as discrete choice analysis (Train Reference Train1986, Reference Train2009; Wassenaar & Chen Reference Wassenaar and Chen2003; Michalek et al. Reference Michalek, Feinberg and Papalambros2005; Kumar et al. Reference Kumar, Hoyle, Chen, Wang, Gomez-Levi and Koppelman2009; Zhang, Gensler & Garcia Reference Zhang, Gensler and Garcia2011).

Simple and effective, Pugh matrixes are widely used by engineers for idea selection (Honkala, Hämäläinen & Salonen Reference Honkala, Hämäläinen and Salonen2007; Ullman Reference Ullman2009). Design alternatives are evaluated in terms of a set of criteria against a base design concept. A scoring system (1, 0, $-$ 1) indicates that a design alternative performs better, equal to, or worse than the base concept for each criterion. A total score for each alternative is tallied, the best aspects of different ideas are combined, and the process continues iteratively until one idea is clearly the best alternative. Frey et al. (Reference Frey, Herder, Wijnia, Subrahmanian, Katsikopoulos and Clausing2009) evaluated this method by quantitatively modeling its implementation process. They confirmed the effectiveness of this method during the early design stage, and found that it facilitated the decision convergence with objectivity.

Idea selection can also be guided by formal engineering optimization approaches combined with market share predictions. Interdisciplinary optimization mathematically finds the ‘best’ alternative as evaluated by an objective function subject to conditions including customer preferences and design constraints (Papalambros & Wilde Reference Papalambros and Wilde2000), and has identified optimal product designs with predicted market performance (Michalek et al. Reference Michalek, Feinberg and Papalambros2005; MacDonald et al. Reference MacDonald, Lubensky, Sohns and Papalambros2009; Kelly et al. Reference Kelly, Maheut, Petiot and Papalambros2011).

Decision-Based Design (DBD) (Hazelrigg Reference Hazelrigg1998; Li & Azarm Reference Li and Azarm2000; Gupta & Samuel Reference Gupta and Samuel2001; Wassenaar & Chen Reference Wassenaar and Chen2003; Wassenaar et al. Reference Wassenaar, Chen, Cheng and Sudjianto2005) helps select a product design alternative through systematic steps. Hazelrigg (Reference Hazelrigg1998) offers a DBD framework that considers factors like design variables controlled by designers, product attributes valuable to customers, demand, price, and time. It allows for the incorporation of methods like discrete choice analysis and profit optimization, and can be adapted to a variety of design problems (Li & Azarm Reference Li and Azarm2000; Gupta & Samuel Reference Gupta and Samuel2001; Wassenaar & Chen Reference Wassenaar and Chen2003; Wassenaar et al. Reference Wassenaar, Chen, Cheng and Sudjianto2005). Wassenaar & Chen (Reference Wassenaar and Chen2003) use DBD to identify the design alternative with the largest net revenue, demonstrated with the design of an electric universal motor.

Day (Reference Day2007) suggested using both the risk matrix and the R–W–W (‘Real, Win, Worth it’) screen to determine whether an innovation should be implemented or not. Positioning an innovation on the risk matrix, which was defined by two familiarity dimensions (the company’s familiarity with the new product and its familiarity with the intended market), allowed an estimation of the innovation’s probability of failure. The R–W–W screen provided a three-level question set that the company should answer for its innovations, which helped the company assess the innovations for screening.