Designing Community-Based Open Innovation Platforms Based on Actual User Behavior

Today’s information society is characterized by a multitude of technological advances that enable innovative products and services for end users. Many innovation projects, however, do not take actual customer needs sufficiently into account: In the innovation process, customers often play a purely recipient role without co-determination rights – products are developed for the target group, not in cooperation with it, which causes product launches on the market to fail [1]. However, innovations are essential for the competitiveness of companies, especially in a dynamic competitive environment [2]. Knowledge is the central prerequisite for the development of innovations [3]. The knowledge of customers allows conclusions to be drawn as to whether a product idea offers added value for a target group. However, such knowledge of customer needs can only be gained through dialogue with the target group [4].

The open innovation approach makes such knowledge of target groups usable for the development of products and services: the collective intelligence of (potential) customers serves as a knowledge resource in product development processes. Open innovation methods such as community-based innovation platforms motivate customers to submit, comment, rate and collaboratively develop innovative proposals online in Web 2.0 environments within a limited time frame [5]. In view of the Grand HCI Challenge “Learning and Creativity” [6], community-based innovation platforms represent a promising opportunity for users to gain knowledge and develop new or unusual ideas with others.

From a business point of view, these innovation methods have a high potential for obtaining target group-specific knowledge in high quality and quantity: From user-generated contributions, information on needs and corresponding solutions as well as trends in terms of design preferences can be derived. The successful application of the method depends significantly on how it is possible to design the platforms in a customer-centered way: The more users submit ideas and improve existing ideas according to the principle of co-creation, the higher the output of generated need and solution information. The successful implementation of open innovation platforms therefore requires a high level of platform usability on the one hand and high-quality social interaction (sociability) on the other. Research on open innovation, which is predominantly economic in nature, hardly deals with these questions.

In a study on one community-based innovation platform user interaction via commenting was analyzed. The analysis of the commentary sequences makes it possible to obtain indications of interaction problems and corresponding requirements for the usability and sociability of open innovation platforms.

The study focuses on three research questions:

1. 1.

   Quantitatively: At which points in time during the life of an innovation platform do users participate with contributions (idea submissions, comments, ratings)? How extensive are discussions between users (length of chains of related comments)?

2. 2.

   Qualitatively: For what purposes are idea comments used? Which types of comments promote co-creation and sociability, which ones hinder both aspects?

3. 3.

   Quantitatively and qualitatively: Which challenges of platform and community design can be identified from the contributions on the platforms?

The results are discussed in terms of requirements for the design quality of community-based innovation platforms – both from the perspective of (communicative) usability and sociability.

2.1 Co-creation Online

Co-creation is defined as “[…] any act of collective creativity, i.e. creativity that is shared by two or more people.” [8]. According to this approach, users jointly develop existing products incrementally: Each participant contributes to a common pool of knowledge that would not arise as a public good under a market institutionalization [9] and initiates further improvements by other participants. Co-creation thus enables a multi-perspective view of the subject matter [10] and continuous optimization of content in iterative cycles [11].

Participation in co-creation processes requires extensive domain knowledge, dedication and motivation [8]. Users invest such a high level of effort due to various incentives, which Füller [12] divides into four categories: Rewards (e.g.: material prizes), needs (e.g.: prospect of implementing an idea), curiosity (e.g.: pleasure in new challenges) and intrinsic motives (e.g.: participation in the product development process).

Collaboration between customers according to the principle of co-creation is mainly written and internet-based using web-based writing technologies [13]. Rowley et al. [14] describe how users of online product forums use web-based writing technologies to collaboratively generate content: Users publish suggestions for improvement in the form of comments on existing products. Other users, in turn, continuously publish comments in which they take up and develop the initial suggestion for improvement. Rapid feedback from users on published contributions is the basis for this type of co-creation [13]. For companies, such user-generated improvements of idea descriptions submitted by other users represent a high added value: On the one hand, idea proposals are more elaborated in this way and their marketability is increased, on the other hand, companies can forecast market acceptance based on user feedback.

2.2 The Design of Community-Based Open Innovation Platforms

Open innovation refers to the targeted use of incoming and outflowing knowledge to promote internal innovation and to expand markets for the external exploitation of innovation [15]. Open innovation platforms are one method of achieving this goal. On such platforms, companies ask customers for structured input of topic-related suggestions for improvement and ideas [9]. Community-based open innovation platforms aim at establishing a predominantly hedonically oriented innovation community. Such communities consist of groups of end consumers who voluntarily develop improvements to existing or completely innovative products [16].

The collaborative development of innovations is enabled by community functions (comment function, news function, tagging, integration of social media such as Twitter feeds). Some of the community-based competitions offer co-development functions that allow co-authoring in the production of content [5]. In innovation communities, both need information and solution information can be collected at various stages of development, from brief descriptions to fully implemented prototypes [17].

The literature on methods of open innovation, which is mainly based on economic science, only occasionally mentions design requirements for community-based open innovation platforms. The requirements are directed towards the evaluation of ideas, the retention of users and the development of a community.

Evaluation of Ideas:

Open innovation platforms should offer conventional evaluation functions (e.g. voting or ranking) which users can use without explanations or help [18].

User Loyalty:

Users should be motivated to visit open innovation platforms frequently and to stay on the platform for a longer period of time per visit - corresponding incentives are created by functions such as ranking lists, announcements of new ideas in competition via app or the connection of social media such as Facebook or Twitter [19].

Community Development:

Requirements for the design of community-based open innovation platforms aim to give users the impression of being part of a community of innovators with expertise in a common field. Such an innovation community can be established in particular through the use of community functions, such as chat functions, user profiles or discussion forums. Submitted ideas should be able to be recommended and evaluated using community functions [19].

The requirements described are mostly limited to the naming of functions that are to be implemented in innovation platforms. With regard to the design of functions, it is assumed that users adapt their actions to the characteristics of the platforms [20]: Users are assumed to familiarize themselves with open innovation platforms if they have sufficient interest in the task to be solved and the subject area and thus qualify for participation in innovation activities (self-selection). Users are expected to use strategies such as trial and error to learn applications. According to von Hippel [20], the usability of open innovation platforms is guaranteed if users can participate even with little domain-related knowledge. Such approaches lack design requirements that consider the actual needs of users.

2.3 Communicative Usability

Usability is an essential criterion in the design, implementation and optimization of electronic applications. The field of usability research is significantly influenced by disciplines such as psychology, technical subjects and ergonomics. The definition of what is meant by usability varies in part depending on the subject. Shackel [21], for example, defines usability as the interaction of efficiency, effectiveness and user satisfaction. Nielsen [22] names six design criteria for usable websites: Hypertexts should be easy to learn, memorable, error-free as well as pleasant and efficient to use. However, such definitions often do not consider linguistic-communicative aspects of system design [23, 24]. Jakobs combines both aspects in her approach to communicative usability. Communicative usability focuses on the question how linguistic and semiotic means can contribute to making the dialogue between humans and technical artefact transparent, trouble-free and enjoyable. The approach is based on action theory and focuses on communicative modes (as the most important modality for the exchange between user and machine), superordinate action contexts (e.g., how the communicative design supports goal hierarchies of users), action contexts (e.g., does the design consider domain-specific or socio-cultural conditions) as well as the interplay of content, interface and aids.

2.4 Sociability

Sociability is a concept developed by Preece [25, 26] to evaluate the success of online communities. It records the quality of interaction between users in online communities in relation to the components purpose, users and guidelines. The purpose of a community is described as the common focus of all members on a topic, a need, information, offer or support; it embodies for each member the reason for participating in the community. Users are considered in terms of their role in the interaction, examples of roles are Leader, protagonist, entertainer or moderator. Regarding community guidelines, two types of guidelines are considered according to their implicitness: Informal rules based on a common language and common standards that guide social interaction and formal rules that are laid down in registration guidelines or codes of conduct.

According to Preece [27], sociability can be evaluated and promoted. It distinguishes between purpose and user-related criteria. Purpose-related criteria include the number of messages sent, interactivity, reciprocity and the quality of user-generated contributions. Preece describes the criteria as follows:

• The number of messages sent reflects the commitment of the members and is an indication of the extent to which the community fulfils its purpose.

• Interactivity is measured by the number of comments posted by users in response to a topic.

• Reciprocity refers to the ratio in which users accept support from the community and make contributions themselves. An example of reciprocity is the number of questions a user asks in a community in relation to the answers he or she publishes to questions from other users.

• The quality of contributions is measured by the length and the style of wording of the contributions. The quality of contributions can be made visible by providing community members with a function for rating comments.

The number of community members is perceived as user-related success criterion for sociability. The success of policies within a community can be determined by the extent of rude and offensive comments. The trustworthiness of members is measured by the number of exclusions from the community due to inappropriate behavior.

An essential statement of the sociability approach is that the way the interactors behave is influenced by the design of the interaction environment. Sociability therefore requires the careful design of the functions and framework conditions through which users can interact with each other. Initial design decisions - especially regarding guidelines for community use - must be continuously checked for suitability and adapted to the needs of the users [27].

3.1 Data Collection

A community-based innovation platform focusing on the topic of mobile communication was selected for the study. The platform offers users the opportunity to enter ideas in a limited time frame (platform lifetime: originally 64 days, after an extension at the end of the first period 96 days in total) and comment on them using a simple web forms, rate ideas by awarding them stars. The platform provides brief descriptions of the task, but no rules of conduct. The ideas, idea comments and idea ratings were collected - the entire corpus comprises 269 idea descriptions, 655 comments and 742 ratings.

3.2 Data Preparation

The database was enriched by the publication dates of the idea descriptions, the relationships between ideas and comments, and the relationships between ideas and ratings. In addition, interrelated comments (based on direct references to the content or authors of previous comments) were grouped into comment sequences. For each comment, the position within the corresponding comment sequence was determined (depth of comment). The lengths of the comment sequences were added to the database.

3.3 Data Analysis

The data analysis examined the chronological distribution of submissions, types of comments and types of comment sequences.

Time Course of the Submissions:

The ideas were sorted chronologically. The platform duration was determined based on the start and end of the submission period and divided into 20 segments (five percent steps). For each segment, the number of ideas published during this period was determined as well as the number of associated comments and ratings.

Types of Comments:

Comment categories were inductively formed from the material in a qualitative content analysis. The comments of the database were assigned to the categories by two annotators. Differing annotations and difficulties in assigning comments were discussed and resolved. Problems that arise when using certain comment types were annotated.

Types of Comment Sequences:

Commentary sequences were qualitatively examined in terms of their composition of comment types. The annotation was carried out in the same way as the determination of the comment types. The comment sequence types were qualitatively examined regarding their potential for promoting and reducing co-creation and sociability.

4.1 Frequency of Submissions

The results of the analysis show a discontinuous publication behavior of users. Many ideas were submitted on a few individual days, and on the majority of all days no or few ideas were submitted. Figure 1 shows the distribution of publications over time: Significantly more ideas were submitted at the beginning, middle and especially towards the end of the platform lifetime.

Distribution of ideas submitted during the lifetime of the platform.

Full size image

The distribution of comments (see Fig. 2) and ratings (see Fig. 3) according to submitted ideas shows that peaks of highly commented and rated ideas are located at the beginning, the middle and the end of the platform lifetime.

Distribution of comments submitted during the lifetime of the platform.

Full size image

Distribution of ratings submitted during the lifetime of the platform.

Full size image

An increased number of idea submissions leads to an immediate increase in comments and ratings. Five measures appear to have a direct impact on user activity (submission of ideas, comments and ratings):

• Advertising: Periods of high activity occur at times when the competition is actively promoted on external websites (technology and news portals).

• Newsletter: Further influence is exerted by platform-internal calls for more participation by e-mail: Newsletters were sent to all platform users on two days during the platform period. On both days, there was a significant increase in contributions.

• Submission deadlines: In the periods shortly before the two submission deadlines, days with high user activity accumulate.

• Deadline extensions: After the announcement of the extension of the first submission deadline, there was a brief increase in user activity.

• Mutually reinforcing user activity: In the middle of the platform period, 5 individual users (possibly motivated by the platform’s advertising measures) began to comment and evaluate ideas more intensively. These activities resulted in reactions from other rather passive users.

Regarding the sequences of related comments, the frequency with which users comment on the comments of other users was investigated (the frequency is measured in the length of the comment sequences). The data analysis identified 118 comment sequences consisting of 324 comments. The majority of all sequences consist of two to three comments (see Table 1).

4.2 Contents of Comments

Users of the platform use comments to clarify difficulties in understanding ideas, to negotiate the innovative content of ideas and to develop ideas further. Five main types of comments were identified in the content analysis: Questions about ideas, endorsement of ideas, criticism of ideas, suggestions for improvement of ideas, and author reactions. The distribution of the comment types is shown in Table 2, the individual types are described below.

Questions about ideas: Users of the innovation platform address questions to the authors of idea descriptions in order to obtain additional information for a better understanding of the described idea. Questions refer to the target group (10.81% of all questions), the functionality (71.62% of all questions) or design aspects (17.57% of all questions) of an idea.

The question about the target group of an idea aims at two difficulties of understanding: It is either unclear whether the idea is suitable for the target group as a whole. Other questions relate to the usability of the idea for a subset of the target group with special restrictions such as wearers of pacemakers.

Questions about the functionality of an idea result from inaccurate or incomplete idea descriptions and aim at increasing the elaboration of an idea.

Questions about the design implementation are asked in order to be able to understand design decisions directed at the size, material, arrangement, color or shape of design elements of an idea.

• Endorsement: This category records positive comments on the ideas of others. One third of all endorsements (31.15%) consist of a short praise (“Great idea!”) to which no justification is added. The rest (68.85% of all ideas) are justified by

  • listing advantages of the idea (19.13% of all endorsements),

  • emphasizing individual aspects of the idea (15.3% of all endorsements),

  • applying the idea to a practical situation (14.75% of all endorsements),

  • positively highlighting the elaboration of the idea (13.66% of all endorsements) and by

  • comparing the idea with existing products (6% of all endorsements).

Justified endorsements highlight the innovative potential of an idea and provide other users with arguments why an idea should be positively evaluated.

• Listing of advantages: Some ideas are evaluated by simply listing advantages such as ease of use, proximity to the target group or protection of the user’s health.

• Highlighting of aspects: The commentator cites a single aspect or object from the idea description as the reason for his or her positive perception.

• Application to a practical situation: The commentator describes a recurring everyday problem that could be solved by the idea and thus emphasizes its value.

• Highlighting the elaboration: In several cases, the positive perception of ideas refers to the idea description and added files such as images.

• Comparison with existing products: The idea is compared to existing solutions and thus exemplified as solution to issues of such products.

• Criticism: Rejections are used to criticize ideas posted on the platform. A small minority of rejections are not justified (2.14% of all rejections). The majority of rejections (97.86% of all rejections) are justified by

  • questioning the feasibility of the idea (2.53% of all criticisms),

  • criticizing the quality of the drafting of the idea description (5.06% of all criticisms)

  • criticizing the idea as being too complex (5.39% of all criticisms),

  • questioning the usefulness of the idea for the target group (8.23% of all criticisms),

  • criticizing the usability of the idea (10.78% of all criticisms)

  • pointing out that the idea already exists (65.87% of all criticisms)

Reasoned refusals reduce the perceived innovation potential of submitted ideas. In the case of references to already existing ideas, the criticized idea submission is denied any innovative potential - and thus the right to participate in the competition. Reasoned rejections are explained below.

• Lack of feasibility of ideas: Commentators doubt whether the idea can be implemented as a product.

• Poor quality of drafting: Commentators criticize the quality of idea descriptions.

• Lack of usefulness of ideas: Comments in this subcategory question whether an idea is suitable for the target group.

• Lack of usability of idea implementations: With comments in this subcategory users criticize weaknesses in the usability of an idea implemented as a product.

• Idea theft: The most frequent criticism of ideas is the accusation of adopting ideas on the platform of already submitted ideas (59.42%) or existing products (40.58%). The majority of the accusations (56.52%) are substantiated by stating a hyperlink leading to similar ideas on the innovation platform or to articles of existing products.

• Improvement suggestions for ideas: Suggestions for improvement help users discover the potential for optimization of submitted ideas and support idea authors in improving their submissions. Suggestions for improvement refer to

  • the design of an idea (3.88% of all suggestions for improvement),

  • alternative areas of application for an idea (4.3% of all suggestions for improvement),

  • improving the elaboration of an idea (12.3% of all suggestions for improvement),

  • the technical implementation of an idea (32.79% of all improvement suggestions) and

  • the extension of an idea (46.72% of all improvement suggestions).

Suggestions for improvement are explained below.

• Design suggestions: Design Suggestions focus on the visual appearance of the idea implemented as a product - users use this sub-category of suggestions for improvement to show possibilities for optimizing design aspects.

• Suggestions for alternative areas of application: This subcategory of suggestions for improvement indicates areas of application in which the idea could be used beneficially.

• Elaboration suggestions: Comments in this subcategory show possibilities for optimizing an idea. They refer to the scope of the idea description, the creation of visualizations, structuring as well as formulation hints. Notes on scope recommend a more detailed and precise description of the entire idea or individual aspects. Recommendations for the creation of visualizations guide authors as to the form (e.g. as a sketch) and tools (e.g. image editing software) in which an idea can be visualized.

• Proposals for technical implementation: Proposals for technical implementation refer to the question of which technologies or mobile phone components can be used to implement an idea as a product.

• Extension suggestions: Users state possibilities how the benefit of ideas can be increased by additional content, features or functions. Users often suggest combining the commented idea with an idea they have submitted themselves.

• Reactions: Some of the comments are written by authors who explicitly address the content of comments on their ideas. The majority of reactions can be assigned to one of the four comment types described above (87.41%) - reactions mainly refer to rejection of ideas (36.42% of all reactions) and questions (23.18% of all reactions), less frequently to suggestions for improvement (15.89% of all reactions) and endorsements (11.92% of all reactions). In addition, there is a sub-category of reactions criticizing the behavior of users of the platform (12.58% of all reactions). Behavioral criticism refers to both ratings of ideas and commentary content. Criticism of ratings is published when low ratings are not justified with a comment or when rating and justification in the comment do not match. Criticism of comment content refers to comments that are perceived as inappropriate.

4.3 Comment Sequences

The distribution of the 118 commentary sequences among the four categories mentioned can be seen in Table 3. There is a clear majority of commentary episodes that have a positive impact on user interaction (77.97% of all episodes). A minority of the consequences have a negative impact (22.03% of all consequences). The categories are explained below.

• Sociability-promoting comment sequences: Sociability-enhancing discussions are discussions in which users express and negotiate positions on an idea. Such an exchange of views enables the formation of a community on the innovation platform. There are three reasons for this exchange of views: To clarify questions about an idea, to negotiate criticism of an idea and to praise an idea. The majority of these commentary sequences consist of only two comments (76.79% of the sociability-promoting commentary sequences), which are usually published on the platform at short intervals one after the other. Three-part chains supplement the above-mentioned two-part chains by thanking for the reaction to the first comment (63.64%), by a specification of the question asked at the beginning (42.86%) or by an objection to the rejection of criticism of an idea (14.29%). The only sequence comprising four comments begins with a question that is answered by the author of the idea. The questioner finds the answer insufficient and specifies his question, whereupon he receives a more precise answer from the author of the idea.

• Sociability-hindering comment sequences: Sequences in which users accuse each other of misconduct are an obstacle to the development of sociability. There are two consequences: First, sociability-reducing consequences inhibit the perceived quality of social interaction and second, they distract users from activities that generate profit for the platform operator (evaluation and further development of ideas). Comment sequences in which users accuse each other of misconduct result from criticism of ideas as well as from the accusation that the idea has been adopted. Such sequences are among the longer consequences on the platform: 42.86% consist of three comments, 14.29% of four comments, 42.86% of six comments. The accusation of misconduct represents a serious damage to the reputation of the commentator – the need for clarification of the accusation causes longer consequences of comments which, in contrast to other commentary sequences, have an aggressive to offensive style of wording. Sociability-reducing commentary sequences make it difficult to build trust between users.

• Co-creation-promoting comment sequences: Comment sequences foster co-creation on the innovation platform if users further develop collaboratively submitted ideas via comments. The results show that improvements in ideas can result from endorsements (5.56% of co-creation-supporting comment sequences), questions (8.33% of co-creation-supporting comment sequences) or the accusation of idea theft (22.22% of co-creation-supporting comment sequences). In the third case, the original idea is expanded in response to the accusation of idea theft in order to distinguish it from the existing idea or product. In the majority of all cases, an improvement proposal is submitted directly without prior discussion (63.89% co-creation-promoting commentary sequences). Direct suggestions for improvement lead to two different types of commentary sequences: either one or more endorsements of the improvement suggestion (27.78% of co-creation-promoting comment sequences), or iterating suggestions for improvement (36.11% of co-creation-promoting comment sequences).

• Co-creation-related are among the longest commentary sequences on the innovation platform: 50% of the episodes contain between three and fourteen comments. Especially relevant for co-creation on the innovation platform are iterative suggestions for improvement: Optimizations of an idea are mutually improved by different users. In this way, ideas are further developed by the community instead of by individual representatives of the community. However, suggestions for improvement are seldom added to the idea descriptions: Only in 5.2% of all idea descriptions text passages were identified which indicate revisions based on suggestions for improvement (usually starting with a note (“To clarify:”, “Update:”, “Explained in more detail:”).

• Co-creation-hindering comment sequences: Comment sequences impede or prevent co-creation if suggestions for improvement of ideas are rejected by the author, if it is pointed out to users that suggestions for improvement have already been submitted as ideas or if accusations of adopting ideas are rejected. In the first two cases, no further suggestions for improvement were submitted after the rejection of a suggestion for improvement – the rejection of suggestions for improvement ends all interaction abruptly. Desired effects such as iterating suggestions for improvement are thus prevented.

• In the third case, the focus of user interaction shifts: Users no longer work together on the further development of ideas, but act against each other to clarify the accusation of idea theft. Discussions about the similarity of ideas or ideas and existing products are not made usable for co-creation, e.g. by proposing extensions of the discussed idea that would result in a significant difference to the objects of comparison. Instead of promoting co-creation, discussions end in accusations of idea adoption.

The discussion-interrupting effect of the three cases also indicates that commentary sequences in this category are very short: 73.64% consist of only two comments, 10.53% of three comments and 15.8% of four comments. Co-creation-reducing commentary sequences make it difficult to build trust between users.

5.1 Implications for the Design of Innovation Platforms Regarding Communicative Usability

Platforms represent a hypertext genre for the acquisition of need and solution information from target groups. The main goal for users to offer such information is the prospect of winning a prize by submitting an idea. Achieving this goal is impeded on the one hand by competing users (and their strategies for gaining an advantage, see Sect. 5.2), and on the other hand by the platform design. The results indicate several problems for users to screen, submit, comment on and rate ideas. These problems are related to the quality of the communicative platform design regarding (the interplay of) content, interface and help features (see Sect. 2.3).

Screening ideas: Users interested in submitting an idea are confronted with the high volume of user-generated information, which poses two major challenges:

• Users cannot invest the necessary time to check whether an idea to be submitted already exists as a contribution on the platform – there are multiple submissions of similar idea descriptions which lead to discussions about idea theft.

• Users who wish to participate in the development of solutions to specific topics or questions can only identify suitable contributions with high efforts.

For the challenges described, approaches are needed that automatically check contributions for topic affiliation and arguments contained in them. A promising approach could be to offer search and proposal functions which apply text mining methods. Such interface elements allow to access the platform contents using topic tracking and argumentation mining on the basis of automated part of speech annotation. Categorization of contributions offers users the possibility to check short descriptions of their ideas for similarity to existing contributions before publication or identify ideas in their own field of expertise for co-development activities.

• Creation of ideas: Ideas submitted to the innovation platform show fluctuating quality, especially regarding the linguistic implementation: Contributions contain many orthographic errors and deficiencies in coherent text structure. Such issues should be addressed with platform functions: Authors should be offered powerful writing technologies (instead of simple web forms) that support idea structuring and formulation as well as the creation of supplementary files such as visualizations. Help features should guide the user in every step of the process, e.g. by suggesting technical terminology. The most important feature to support users in creating ideas is content provided by the platform operators: Such content needs to state clearly what the operators expect from the users, how submissions will be used by the operators and which criteria submissions need to meet in order to be considered as adequate to win a prize.

• Commenting: Idea comments are often brief and can lead to difficulties of understanding or undesired reactions on the part of the authors, if they are not carefully formulated (see Sects. 4.2, 4.3). Since comments are submitted in the same way as idea submissions via web forms, it is likely that commentators face the same problems as idea authors. Such problems can be solved with writing technologies as described above. Another problem is the use of comments to link ideas: This strategy is used by users to promote their own ideas (see Sect. 4.3). From the point of view of platform operators, such proposals are particularly promising as they show synergies between submissions. From the users’ point of view, such suggestions for combining ideas can only be indicated via comments, where they remain largely invisible. Desirable would be platform functions that allow different ideas to be linked and emphasize the added value.

• Rating: The sociability of a community-based innovation platform is decreased if users rate ideas with a low score and do not justify the evaluation via comment. Authors suspect that other users try to gain competitive advantages in this way. To prevent such strategies, the platform design should connect commenting and rating in a way that only justified ratings are allowed. The examined platform allows to rate ideas with a score between one and five stars – in such cases explanations are needed as to what the individual scores mean from the point of view of the platform operator. It should be explained in which cases which score is suitable for an idea of which level of elaboration.

5.2 Implications for the Design of Innovation Platforms Regarding Sociability

The results indicate that user interaction on the examined innovation platform takes place only sporadically during the runtime as well as purposefully in short commentary sequences (see Sect. 4.1). However, findings like mutually reinforcing user activity (see Sect. 4.1) and co-creation-promoting comment sequences (see Sect. 4.3) show the potential of user interaction for innovation platforms. Such positive interaction can, however, be disturbed by negative interaction, which is manifests, for example, in the accusation of idea theft (see Sect. 4.2). According to the measures of sociability (see Sect. 2.4), the examined platform shows a relatively low quality of social interaction. Interaction problems arise from particular comment types and types of commentary sequences. Identified problems are described below. For each problem, rules are proposed which are intended to promote confidence building, sociability and co-creation as a code of conduct on community-based innovation platforms. The rules should be monitored by moderators, multiple violations should be punished by blocking user accounts or withdrawing access to the comment function. To encourage community interaction and co-creation, the reward and incentive system should be expanded: Community awards should be awarded for frequent constructive comments and co-developer awards for improvements.

The results indicate that some of the interaction problems are caused by a lack of quality in the submission of ideas: Qualitatively inferior idea descriptions and appendices are criticized, inferior idea descriptions result in inferior comments. A first step is to encourage users to submit ideas in the best possible quality.

• Rule 1: Make sure to submit ideas in the best possible quality.

• General interaction problems arise from unobjective and insulting formulations in comments. Users should be obliged to follow the rules of netiquette. The netiquette should be made available on the platform additionally.

• Rule 2: Please comply with the netiquette. Violations will be punished by blocking user accounts or the comment function.

• It is often the case that users only comment on descriptions of ideas, but do not take up comments from other users. Reactive comments neither promote confidence building nor co-creation. Users should be encouraged to comment discursively.

• Rule 3: In your comments, always consider the comments that were previously published on the idea in question.

• Interaction problems arise from all comment types, they are described below per type. Questions are often formulated in a less concrete way so that it not obvious which aspect of the idea is referenced. A corresponding guideline should encourage users to formulate questions concisely.

• Rule 4: If you have a question about an idea, always indicate to which point of the idea description and which aspect mentioned there you refer to. Add all necessary information to your question so that other users of the platform can understand your question.

• A common problem is unfounded endorsement and criticism – the author of the idea cannot understand which aspect of the idea is evaluated; other users cannot use such comments as basis for a discussion. Commentators should be required to provide mandatory justifications for idea evaluations.

• Rule 5: When evaluating an idea, always give a reason so that other users can participate in the discussion.

• Endorsement and criticism that are given by applying an idea to a practical situation from the everyday life of the commentator offer advantages for users and for the platform operator: From the user’s point of view, praise can be better understood; from the platform operator’s point of view, such descriptions provide need information that can be used for product development.

• Rule 6: When evaluating an idea, please always address situations in everyday life where the implementation of the idea would have an impact.

• Criticism of ideas is often the starting point for unobjective, insulting reactions. Furthermore, the results indicate that rejections are used to reduce the chances of ideas submitted by other users. Abuse of the commentary function must be prevented. For this reason, each rejection should be accompanied by a precise description of what the comment refers to. Unsubstantiated criticism does not enable authors to improve their ideas. Justifications should be mandatory in the context of rejections.

• Rule 7: When you criticize an idea, always indicate to which point of the idea description and which aspect mentioned there you refer to. Criticism of ideas must always be justified with arguments and contain constructive suggestions for improvement.

• The most frequently discussed topic is the accusation of idea theft. Such comments reduce the quality of social interaction, building trust and co-creation. In addition, the accusation of idea theft ends other discussions abruptly. For this reason, accusations of adoption should not be discussed publicly between users. Instead, users should report suspected idea theft to the platform operator, who will investigate the case and take action. In addition, users should be informed about how the competition jury assesses ideas that have been submitted several times.

• Rule 8: If you find that an idea has been submitted to the platform before or exists as a product, please report this to the platform operator. Such matters should not be discussed between users of the platform. If a similar idea is submitted by several users, the jury will evaluate the version that was submitted first.

• As the high amount comments that are not part of comment sequences as well as the low amount of the comment type ‘reaction’ indicate, reactions to comments by authors of ideas are often omitted, thus preventing continuous interaction on the platform. Authors should be required to respond to each comment. Violations of the guidelines should be taken into account in the evaluation by the competition jury.

• Rule 9: If one of your ideas is commented on, always write a comment in which you react to the corresponding contribution. Missing reactions to comments from the community will result in a devaluation of the idea by the jury.

• The results indicate that authors often do not take up criticism or suggestions for improvement to revise their idea descriptions. They should be motivated to improve their own ideas by means of appropriate guidelines.

• Rule 10: If suggestions for improvement are made or weaknesses of your idea are revealed in the comments, please revise your idea accordingly.

• Authors often respond to criticism by merely denying counterarguments. To encourage interaction and co-creation, authors should always be required to justify such statements with pro-arguments.

• Rule 11: If you want to refute criticism of your ideas, this is always done objectively and by providing arguments.

User comments offer promising opportunities to improve ideas on innovation platforms. From the user’s point of view, suggestions for improvement promote the quality of social interaction and confidence building between users. From the operator’s point of view, they increase the quality of solution information and increase the market potential of product developments. Ideally, users apply collaboration-oriented usage strategies such as iterative co-creation in order to continuously optimize ideas from each other.

If the platform operator does not communicate desired and undesired behavior in the interaction between users in a concise and comprehensible way, the quality-increasing potential of user comments is reversed: platform participants apply informal competition-oriented rules in order to gain competitive advantages for themselves and corresponding disadvantages for competitors. Clear formal behavioral guidelines as well as their monitoring and enforcement by moderators are an essential requirement for the design of community-based innovation platforms: Only with such measures can continuous, constructive, collaboration-oriented community interaction and the successful application of the open innovation method be guaranteed. Such measures must be supported by high usability as well as functions that enable users to focus on the collaborative development of ideas.