ArticlePDF Available

Abstract and Figures

Cyber-physical systems (CPS) are able the collect huge amounts of data about themselves, their users, and their environment during their usage phase. By feeding these usage data back into product planning, manufacturers can optimize their engineering and decision-making processes. Despite promising potentials, most manufacturers still do not analyze usage data within product planning. Also, research on usage data-driven product planning is scarce. Therefore, this paper aims to identify the main concepts, advantages, success factors and challenges of usage data-driven product planning. To answer the corresponding research questions, a comprehensive systematic literature review is conducted. From its results, a detailed description of usage data-driven product planning consisting of six main concepts is derived. Furthermore, taxonomies for the advantages, success factors and challenges of usage data-driven product planning are presented. The six main concepts and the three taxonomies allow for a deeper understanding of the topic while highlighting necessary future actions and research needs.
Content may be subject to copyright.
Cite this article: Meyer, M., Wiederkehr, I., Koldewey, C., Dumitrescu, R. (2021) ‘Understanding Usage Data-Driven
Product Planning: A Systematic Literature Review’, in Proceedings of the International Conference on Engineering
Design (ICED21), Gothenburg, Sweden, 16-20 August 2021. DOI:10.1017/pds.2021.590
ICED21 3289
ICED21 1
Meyer, Maurice (1);
Wiederkehr, Ingrid (1);
Koldewey, Christian (1);
Dumitrescu, Roman (1,2)
1: Heinz Nixdorf Institute, University of Paderborn;
2: Fraunhofer Institute for Mechatronic Systems Design IEM
Cyber-physical systems (CPS) are able the collect huge amounts of data about themselves, their users,
and their environment during their usage phase. By feeding these usage data back into product
planning, manufacturers can optimize their engineering and decision-making processes. Despite
promising potentials, most manufacturers still do not analyze usage data within product planning.
Also, research on usage data-driven product planning is scarce. Therefore, this paper aims to identify
the main concepts, advantages, success factors and challenges of usage data-driven product planning.
To answer the corresponding research questions, a comprehensive systematic literature review is
conducted. From its results, a detailed description of usage data-driven product planning consisting of
six main concepts is derived. Furthermore, taxonomies for the advantages, success factors and
challenges of usage data-driven product planning are presented. The six main concepts and the three
taxonomies allow for a deeper understanding of the topic while highlighting necessary future actions
and research needs.
Keywords: Data-Driven Product Planning, Big data, Machine learning, Early design phases, Product
Lifecycle Management (PLM)
Meyer, Maurice
Heinz Nixdorf Institute, University of Paderborn
Advanced Systems Engineering
3290 ICED21
The megatrend digitalization has turned mostly mechanical and electrical products into complex
systems integrating hardware, sensors, data storage, microprocessors, software and connectivity
(Porter and Heppelmann, 2014). These so-called Cyber-Physical Systems (CPS) integrate computation
and physical processes (Lee, 2008) enabling the collection, analysis and sharing of huge amounts of
data about the product and its environment (Porter and Heppelmann, 2014). As a consequence, CPS
turn their users into incessant generators of data (Erevelles et al., 2016). By feeding these data from
the usage phase back into product development and especially product planning, manufacturers can
optimize their engineering and decision-making processes (Erwin et al., 2015). As a result, these data
enable the companies to turn their customers and users into a powerful source of innovation (Holler et
al., 2016a) and consequently to develop a competitive advantage (Porter and Heppelmann, 2015).
Product planning is the first activity in the product creation process. It aims at finding the success
potentials of the future to create a promising product design in the form of a principle solution
(Gausemeier et al., 2011). The results of product planning are the products to be developed by the
organization (Ulrich and Eppinger, 2016) and the corresponding requirements lists for the subsequent
product development (Pahl et al., 2007). The interconnections between product planning and product
development are strong: For Gausemeier et al., both are coupled by the conceptual design
(Gausemeier et al., 2011). Pahl et al. link product planning to the clarification of the task in product
development to underline the necessity of a content-related junction and a work-related integration of
the two (Pahl et al., 2007). Ulrich and Eppinger describe product planning as the initial phase or phase
zero of product development (Ulrich and Eppinger, 2016).
The success of product planning depends on the utilization of the available data which esp. for CPS
can be enormous. The process of accessing, aggregating and analyzing large amounts of data from
multiple sources is called data analytics (DA). It enables companies to extract knowledge from data to
understand historical and predict future events (Tyagi, 2003). Data analytics is based on mathematics,
computer science, and business analysis techniques (Porter and Heppelmann, 2015). It can be divided
into four types with increasing value and complexity: descriptive, diagnostic, predictive and
prescriptive analytics (Steenstrup et al., 2014).
As shown, product planning and data analytics are two established and independent research areas.
Together, they span the new research area of usage data-driven product planning which still needs to
be thoroughly researched. Studies show that companies increasingly make decisions based on data
analysis results, seeing data analytics as a crucial building block for creating value; yet, most
companies do not utilize usage data within product creation processes (Erwin et al., 2015). Also,
research on the topic is scarce. From these considerations, four research questions are derived:
1. Main concepts: What concepts constitute usage data-driven product planning?
2. Advantages: Why should companies pursue usage data-driven product planning?
3. Success factors: What factors contribute to the success of usage data-driven product planning?
4. Challenges: What makes the implementation of usage data-driven product planning difficult?
To answer the research questions, we conducted a systematic literature review (SLR), following the
suggestions of Webster and Watson (2002), Brocke et al. (2009) and Rowley and Slack (2004). First,
we prepared a list of relevant journals on the basis of the Financial Times Research Rank by Ormans
(2016) and an additional extensive web search for topic-related journals (Webster and Watson, 2002).
We included journals which focus on disciplines like Product Innovation, Innovation
Management, Engineering Management, R&D Management, Systems Engineering, Data and
Knowledge Engineering, Data Science, Big Data, Marketing Management, Operations
Research etc. as usage data-driven product planning is an interdisciplinary field. All in all, the final
list included 44 journals. Second, we created a concept map to identify important concepts and their
synonyms (Rowley and Slack, 2004; Brocke et al., 2009). We used the concept map to iteratively
create and test search strings as a combination of the concepts in the concept map. Through iterative
ICED21 3291
improvement, the following search string was formed: (product development OR product design
OR product engineering) AND (usage data OR operational data OR lifecycle data OR big
data OR data analytics OR closed-loop OR feedback information OR feedback data).
Since the number of logical connectors (OR and AND) was limited, we decided to not include the term
product planning as it is not widely used and our tests provided better results for the alternative
terms product development, product design and product engineering. Third, we searched for
articles within each journal (from 2005 to 2020) in the Google Scholar Database using the literature
research tool Publish or Perish (Harzing, 2007). The search resulted in 1870 articles which we
screened on a title and abstract basis. While hundreds of articles addressed topics like big data or
data analytics in different contexts, the abstracts of only 36 papers suggested a focus on usage data-
driven product planning. After reading these papers and conducting a backward and forward search
(Webster and Watson, 2002), we identified 12 papers which contained answers to our research
questions; the other papers were sorted out. Fourth, we examined papers from conference proceedings.
From prior research, we knew of 14 papers within the context of usage data-driven product planning.
We used these to conduct a backward and forward search, resulting in a list of 76 papers. From these,
12 papers contained answers to our research questions. Finally, following the suggestions of our
reviewers, we added 4 more articles.
In total, we identified 28 highly relevant articles. For the detailed analysis, we read each paper twice,
extracted all relevant information concerning our research questions and loosely clustered them into
thematic chunks. Thereafter, we iteratively improved the clustering by critically investigating each
cluster, splitting clusters up and building new clusters. As the result of this iterative process, we obtained
the main concepts, advantages, success factors and challenges of usage data-driven product planning.
4.1 Main Concepts
Usage data-driven product planning consists of six main concepts (see Figure 1):
1. The products sensors capture real-time readings of the product in its operating environment (Fathi
et al., 2011; Porter and Heppelmann, 2015; Hou and Jiao, 2020; Igba et al., 2015; Chowdhery et
al., 2020). The data include user-generated data (capturing user behavior), product operating data
(capturing product behavior) and environmental data (capturing environment behavior) (Hou and
Jiao, 2020, 2020). In addition, further data like manual reports can also be collected (Chowdhery
et al., 2020). The data collection approach can either be reactive and thus be driven by concrete
events (e.g. machine failures) or it can be proactive by collecting data on a large scale and
analyzing it exploratorily (Holler et al., 2016b; Holler et al., 2017).
2. Using cyber-infrastructure (van Horn et al., 2012), the captured usage data are fed back into the
product creation process (Porter and Heppelmann, 2014; Jun et al., 2007), where they are valuable
in all stages (Holler et al., 2017; Hou and Jiao, 2020; Jun et al., 2007), but offer the highest value
in the early stages like product planning as these are characterized by lots of uncertainties and the
determination of lifecycle costs (Holler et al., 2016b; Holler et al., 2017). Here, the data are used
to objectively quantify product performance and usage profiles (van Horn et al., 2012) to find
usage-centric improvements for the product under consideration (Holler et al., 2016b; Holmström
Olsson and Bosch, 2013; Jun et al., 2007; Hou and Jiao, 2020; van Horn et al., 2012).
3. To identify improvements, statistical analysis, data mining and machine learning techniques must be
applied (Hou and Jiao, 2020; Igba et al., 2015). The data analysis can (a) build upon the available
data in a bottom-up approach (less effort and faster implementation) or (b) start with a predefined
objective in a top-down approach (more effort, clear future-focus) (Wilberg et al., 2017b).
4. The results of the data analysis enable developers to make decisions based on facts instead of
assumptions only (Hou and Jiao, 2020; Chowdhery et al., 2020), thus improving the decision-
making process (Jun et al., 2007; Wu et al., 2020; Fathi et al., 2011).
5. Identified improvements can be implemented in existing and future products (Jun et al., 2007;
Wilberg et al., 2017b; Abramovici et al., 2017; Wu et al., 2020; Xu et al., 2016; van Horn et al.,
2012). For future products, especially the development of new product generations is suited for the
analysis of feedback data (Holmström Olsson and Bosch, 2013; Igba et al., 2015; Fathi et al., 2011;
Chowdhery et al., 2020).
3292 ICED21
Figure 1: The six main concepts of usage data-driven product planning
6. By feeding back usage data into product planning, usage data-driven product planning represents
an inverse design approach which will not replace, but complement the traditional, assumption-
based forward design (Hou and Jiao, 2020; Wilberg et al., 2017a). In conjunction, inverse and
forward design form a reinforcing loop for the continuous and iterative improvement of product
design (Hou and Jiao, 2020).
4.2 Advantages
Usage data-driven product planning offers a variety of advantages which are listed in a taxonomy in
Table 1 and explained below. The advantages apply on three levels: Analysis, Process and Business.
Table 1. Taxonomy of the advantages of usage data-driven product planning
Level 1: Analysis
Level 2: Process
Level 3: Business
(1.1) Finding hidden
(2.1) Improved customer- and
(3.1) Improved decision-making
(1.2) Better product
(2.2) Continuous requirements
(3.2) Usage-centric product
(1.3) Better understanding of
customer and user needs
(2.3) Reduction of hardware-
prototyping and field-testing
(3.3) Higher delivery frequency
of functionality
(1.4) Contextualize and
evaluate qualitative and
subjective data
(2.4) Faster product
(3.4) Higher innovative strength
Analysis-Level: (1.1) The utilization of analytics approaches like data mining makes it possible to
find hidden information within the data that would be impossible to find manually (Menon et al.,
2005). This is especially true when comparing many products and thousands of sensor-readings over
time (Porter and Heppelmann, 2015). (1.2) The data lead to an improved understanding of the product
in operation (Menon et al., 2005; Holmström Olsson and Bosch, 2013). Usage data-driven product
planning uses these insights to improve product design and thus eliminate failures instead of using
data to predict them (Holler et al., 2016a). (1.3) Analyzing usage data leads to a better understanding
of customer and user needs and is likely to deliver better results than traditional approaches
(Timoshenko and Hauser, 2019; Porter and Heppelmann, 2014). The data can be used to analyze
customer behavior and preferences (Hou and Jiao, 2020; Porter and Heppelmann, 2015) and derive
customer segments (Cantamessa et al., 2020). For these, individual (Hou and Jiao, 2020; Holler et al.,
2016b), future (Holmström Olsson and Bosch, 2013), unspoken (Li et al., 2015; Timoshenko and
Hauser, 2019) and latent needs (Hou and Jiao, 2020; van Horn et al., 2012) can be identified. (1.4)
Furthermore, quantitative product usage data helps to contextualize and evaluate qualitative and
subjective data (Holler et al., 2016b).
Process-Level: (2.1) By building on usage data, the customer and user involvement in the
development process is markedly improved (Hou and Jiao, 2020), leading to better collaboration
ICED21 3293
(Holler et al., 2016b). (2.2) Feedback data enable a continuous requirements analysis (Timoshenko
and Hauser, 2019) (e. g. through the continuous live elicitation of customer needs (Cantamessa et al.,
2020)) eliminating momentary biases. (2.3) As the feedback of usage data promotes a usage-centric
and fact-based decision-making process, hardware-prototyping and field-testing necessities can be
reduced significantly (Holler et al., 2016b; Holler et al., 2017). (2.4) Consequently, usage data-driven
product planning allows for a faster product development process (Menon et al., 2005).
Business-Level: (3.1) The integration of data analytics methods into product planning enables data-
and fact-based decisions (Wuest et al., 2014; Holler et al., 2016a) while reducing assumption- and
experience-based ones (Holler et al., 2017; Holler et al., 2016a), thus improving decision-making
processes (Xu et al., 2016; Menon et al., 2005). (3.2) Analyzing usage data promotes a usage-centric
product portfolio (Holler et al., 2016a; Holler et al., 2017; Holmström Olsson and Bosch, 2013).
(3.3) The continuous requirement analysis and the faster development process help companies to
increase their delivery frequency of functionality (Holmström Olsson and Bosch, 2013; Jun et al.,
2007) and thus react on insights not anticipated by the previous design (Cantamessa et al., 2020). (3.4)
Lastly, analyzing usage data enables companies to create high-quality innovations (Li et al., 2015;
Kiron et al., 2014) as it pushes them towards new ideas at a higher speed (Erevelles et al., 2016) and
facilitates their acceptance by keeping the companies open-minded (Kiron et al., 2014).
4.3 Success Factors
To exploit the advantages mentioned, the factors contributing significantly to the success of usage data-
driven product planning must be identified. Table 2 shows a taxonomy of these success factors within four
classes: Organization, Product, Data Analysis and Evaluation. The factors are described below.
Table 2. Taxonomy of the success factors for usage data-driven product planning
Class 1: Organization
Class 3: Data Analysis
Class 4: Evaluation
(1.1) Data strategy
(3.1) Standardized data
and systems
(4.1) Comprehensible
data analysis results
(1.2) Use cases
(3.2) Effective data col-
lection and feedback
(4.2) Decision support
(1.3) Ignorance-b.
view & induct.
(3.3) Proactive data
collection and analysis
(1.4) Cooperation of
product & data experts
(3.4) Joint analysis of
heterogeneous data
(1.5) Integration of DA
with tradition. methods
(3.5) Complex data
(1.6) Turning consu-
mers into prosumers
(3.6) Considering the
product context
Organization: (1.1) A data strategy is critical for success (Wilberg et al., 2017a; Erwin et al., 2015).
It is the result of (a) an in-depth analysis of the product under consideration, the corporate strategy etc.
and (b) a conceptional phase (Wilberg et al., 2017b) which includes the definition of use cases among
others (Wilberg et al., 2017a). (1.2) Use cases create value by e. g. defining the analysis goal, asking
open questions, outlining its potential benefits and deriving the data that need to be collected (Menon
et al., 2005; Wilberg et al., 2017a; Wilberg et al., 2017b). In order to identify opportunities and
limitations, all relevant stakeholders must be involved (Wilberg et al., 2017b). (1.3) An ignorance-
based view and inductive reasoning promote the definition of a successful use case (Erevelles et al.,
2016). The ignorance-based view triggers questions with a high potential for hidden insights while
inductive reasoning will create a greater understanding of the investigated object (Erevelles et
al., 2016). (1.4) As use cases require product knowledge, a close cooperation of product and
data experts is crucial (Shahbaz et al., 2006; Fathi et al., 2011; Li et al., 2015; Cantamessa et al.,
2020). Especially at the start of common projects, interaction points between product and data
experts are numerous, making a direct communication and a common language a necessity (Li et
al., 2019). In the evaluation phase, multiple perspectives from a heterogeneous team may capture more
3294 ICED21
valuable insights from within the data (van Horn et al., 2012). Hence, this so-called knowledge fusion is
critical when evaluating use cases and analysis results (Xu et al., 2016). (1.5) The integration of data
analytics with traditional methods promises the highest new product success and knowledge fusion (Xu
et al., 2016; Hou and Jiao, 2020). (1.6) Lastly, for data feedback, companies need to renew their
customer relationships by turning consumers into prosumers (Cantamessa et al., 2020).
Product: (2.1) Products with a high number of intelligent components are especially suited for
usage data-driven product planning as these generate the data necessary for a large scale data
analysis (Holler et al., 2016a; Jun et al., 2007). (2.2) In order to generate enough data for the
algorithms to work properly, long operating times are a prerequisite (Holler et al., 2016a). (2.3)
When comparing the data of multiple products, a high similarity between these products is essential
(Holler et al., 2016a). Due to this, usage data-driven product planning works best in product
generation planning (Holler et al., 2017) and when using the principles of design modularity and
platforms (Cantamessa et al., 2020). (2.4) Finally, data access must be achieved through providing
customers transparency concerning data usage and incentives in form of a clear value proposition
for sharing their data (Porter and Heppelmann, 2014; Wilberg et al., 2017a).
Data Analysis: (3.1) Standardized data and ecosystems facilitate data exchange and analysis (Wilberg et
al., 2017a; Holler et al., 2016a). For qualitative data, formal taxonomies and standardized information
structures improve reuse (Goh and McMahon, 2009). (3.2) An effective data collection and feedback
process is mandatory for usage data-driven product planning to work. It can be achieved through (a)
smart sensing and advanced data-collection methods (Hou and Jiao, 2020) and through (b) clearly
defined points of data capture and reuse as well as incentives within the peoples work processes (Goh
and McMahon, 2009; Menon et al., 2005). The data must be mapped to concrete product instances as
well as the product type (Abramovici et al., 2017). (3.3) In general, a proactive data collection and
analysis promises to deliver the most value (Holler et al., 2016b). (3.4) In order to achieve valuable
analysis results, heterogeneous data (e. g. structured and unstructured data) need to be jointly analysed
(Porter and Heppelmann, 2015; Igba et al., 2015; Abramovici et al., 2017) in (3.5) sophisticated data
analyses as these promise to achieve deeper insights and more satisfied users (Porter and Heppelmann,
2015; Erwin et al., 2015; Cantamessa et al., 2020). (3.6) Before analyzing data, it is important to
consider the context of the product and its data: These contextual information need to be evaluated when
deciding about the data analysis approach (Shahbaz et al., 2006). Furthermore, they help to understand
the analysis results, potentially leading to deeper insights (Wilberg et al., 2017a).
Evaluation: (4.1) The results of the data analysis must be easily comprehensible as decision-makers are
not involved in all analysis steps (Li et al., 2015). Managers and product developers need to be presented
aggregated and accurate knowledge instead of vast amounts of data (Abramovici et al., 2017).
(4.2) Furthermore, they need decision support systems which give them suitable advice (Jun et al., 2007)
and help them make deliberate decisions based on the information available.
4.4 Challenges
While the success factors help companies to exploit the advantages of usage data-driven product planning,
they will still face numerous challenges. Table 3 shows a taxonomy of challenges using the four classes
Organization, Product, Data Analysis and Evaluation. The challenges are described below.
Table 3: Taxonomy of the challenges of usage data-driven product planning
Class 1: Organization
Class 2: Product
Class 3: Data Analysis
Class 4: Evaluation
(1.1) Definition of a
(2.1) Highly individual
(3.1) Selection of the
data to be collected
(4.1) Validity-check of
the analysis results
(1.2) Definition and
selection of use cases
(2.2) Short-cyclical
product improvements
(3.2) Overviewing data
availability and usage
(4.2) Interpretation of
data analysis results
(1.3) Positive cost-
(2.3) Capturing of the
product context
(3.3) Data pre-
(4.3) Creation of new
(1.4) Integr. of DA into
traditional processes
(2.4) Missing sensors
(3.4) Analysis of large,
multimodal data
(2.5) Prohibition of data
(3.5) Choice of analysis
ICED21 3295
Organization: (1.1) Defining a data-strategy is a major challenge many companies struggle with (Erwin
et al., 2015; Wilberg et al., 2017b). (1.2) This includes the definition and selection of promising use
cases as initially it is often difficult to define use cases at all and later there may be too many to
implement (Wilberg et al., 2017a; Wilberg et al., 2017b). (1.3) Use cases must offer a positive cost-
benefit-ratio (Holler et al., 2016a; Wilberg et al., 2017b), remain within a companys capacities (Wilberg
et al., 2017a) and show their usefulness (Holmström Olsson and Bosch, 2013; van Horn et al., 2012).
(1.4) Another challenge is the integration of data analytics into traditional product creation and
decision-making processes (Hou and Jiao, 2020; Wilberg et al., 2017b), both technically and
organizationally (Wilberg et al., 2017a).
Product: (2.1) Highly individual products represent an obstacle as data analysis insights may not
apply to other instances in the product class and thus may not be transferable (Holler et al., 2016a;
Holler et al., 2017; Cantamessa et al., 2020). (2.2) Furthermore, short-cyclical product improvements
contradict the necessity of long operating times within a given configuration (Holler et al., 2016a).
(2.3) Facing these challenges and the reduced comparability of products, capturing the product usage
context becomes even more important for data reuse (Igba et al., 2015), but it is a major challenge
(Hou and Jiao, 2020; Menon et al., 2005). (2.4) Due to the added costs, products may also miss
sensors and thus be unable to collect the data necessary (Wilberg et al., 2017a). When retrofitting
sensors, functional and economic aspects must be considered (van Horn et al., 2012). (2.5) Moreover,
customers may prohibit data access (Wilberg et al., 2017a), especially to sensitive data (van Horn et
al., 2012), exposing companies to additional complexity and costs to obtain rights to the data (Porter
and Heppelmann, 2014).
Data Analysis: (3.1) While sensors generate lots of data, user-generated data are also extensive for
complex products (Timoshenko and Hauser, 2019). But as more collected data do not necessarily lead
to better results (Hou and Jiao, 2020), the selection of the data to be collected must be consistent with
the strategy (Porter and Heppelmann, 2014) and carefully planned (Menon et al., 2005; Hou and Jiao,
2020). (3.2) Here, another challenge arises: Companies often lack an overview of data availability and
usage (Holmström Olsson and Bosch, 2013; Wilberg et al., 2017b) which prevents information reuse
(Goh and McMahon, 2009). (3.3) Once the data are collected, pre-processing confronts companies
with multiple obstacles. Data validity must be checked (Wilberg et al., 2017a) and multimodal data,
structured and unstructured, must be transformed and integrated (Porter and Heppelmann, 2015; Hou
and Jiao, 2020; Shahbaz et al., 2006; Abramovici et al., 2017) which can be overwhelming for
companies (Kiron et al., 2014). Here, especially textual data are difficult to work with because of their
heterogeneity (Menon et al., 2005). (3.4) In complex industries, analyzing large, multimodal data is
necessary to investigate the product from diverse perspectives (Xu et al., 2016; Hou and Jiao, 2020).
However, companies struggle with a substantial lack of experience (Wilberg et al., 2017b) and seldom
use complex analyses (Erwin et al., 2015). (3.5) Lastly, the choice and appropriateness of data
analysis methods largely affect the quality of the solutions (Cantamessa et al., 2020).
Evaluation: (4.1) Analyzed data forms the basis for product improvement. But before impactful decisions
are made, the validity of the data analysis results must be checked (van Horn et al., 2012) as the data
samples analyzed are typically limited (e. g. due to small sample sizes) (Hou and Jiao, 2020). Here, even
small errors could lead to major mistakes in the solution (Hou and Jiao, 2020). Therefore, it is necessary
that product experts verify data quality and analysis results (Li et al., 2019). (4.2) The interpretation of the
data analysis results shall turn data into valuable insights. However, as inverse problems do not have a
unique solution (Hou and Jiao, 2020), the correct interpretation represents a major challenge (Hou and Jiao,
2020; Xu et al., 2016). To extract valuable insights, knowledge in both engineering and data analytics are
required. Also, contextual information needs to be considered to avoid wrong interpretations (Wilberg et
al., 2017a). (4.3) Lastly, the creation of new ideas based on the analysis and interpretation of existing data
is not trivial (Wu et al., 2020). Also, the derivation of improvements for a whole product class from
(partially) subjective information about one instance is difficult (Abramovici et al., 2017).
5.1 Key Insights
1. Usage data-driven product planning consists of six main concepts (see section 4.1). (1) The
captured user-generated, product operating and environmental data (2) are fed back into product
3296 ICED21
planning and (3) analyzed with statistical analysis, data mining and machine learning techniques.
(4) The results enable fact-based decisions (5) to improve existing and future products. (6) The
feedback process represents an inverse design approach and forms a reinforcing loop with
traditional forward design.
2. Usage data-driven product planning offers advantages on three levels (see section 4.2). On the
analysis-level, the analysis of products, customers and users is improved. On the process-level,
planning and development processes are enhanced while reducing time and costs. On the business-
level, fact-based decisions enable a usage-centric product portfolio and a higher innovative strength.
3. Success factors and challenges can be structured using taxonomies with the classes organization,
product, data analysis and evaluation (see sections 4.3 and 4.4). While some success factors and
challenges are independent of each other, a significant number of success factors correspond
directly with challenges: data strategy, use cases, integration of data analytics with traditional
methods, long operating times, high similarity with other products, data access, effective data
collection and feedback, joint analysis of heterogeneous data and complex data analyses. Their
realization is decisive for the successful implementation of usage data-driven product planning.
5.2 Limitations
The limitations of our study are related to our research design. First, our journal selection process may
have excluded further journals with relevant articles. Yet, we believe that by conducting a thorough
backward and forward search, we found the most relevant articles. Second, due to high number of
related concepts, the construction of our search string may not be optimal. We tested multiple options
on a sample basis, but we did not investigate all papers for every option. Search strings including
different concepts (like product planning, data mining) may yield results not found with our search
string. Third, we searched only within the Google Scholar Database, setting up our dataset based on
titles, abstracts, and keywords. Fourth, the selection of relevant articles as well as the clustering
process were subjective. Other scholars might have included different articles and clustered the
elements in another way.
5.3 Implications for Managers and Future Research
Managers can use the six main concepts to create a better understanding of the topic within their
companies. Likewise, the listed advantages can be utilized to achieve a high level of acceptance and
promote future projects. Using the success factors taxonomy, managers can assess the current and
needed maturity of their companies for usage data-driven product planning (e. g. with a capability
maturity model). Lastly, managers should analyze the challenges and create plans to overcome them.
As usage data-driven product planning is a new research area, various future research is needed.
From our results, three research needs stand out: (1) The creation of a data strategy with promising use
cases is crucial as it guides all following steps. Here, the data strategy as well as the use cases must
align with the companies corporate strategy and its current situation. (2) The integration of data
analytics into traditional forward design processes is also critical as usage data-driven product
planning as a form of inverse design will not replace but complement traditional forward design. (3)
The utilization of data analytics must be made accessible to small and medium-sized enterprises. In
contrast to large companies, they mostly lack resources to build up their own sophisticated solutions.
Finally, as our research focuses on product usage data only, the integration of non-usage data should
be analysed, e.g., how can data from sources like customer reviews complement the usage phase data?
Especially for the identification of customer needs, there is already a high number of approaches using
text mining techniques on social media and online reviews (Bertoni, 2020).
The research and development project DizRuPt is funded by the German Federal Ministry of
Education and Research (BMBF) within the Innovations for Tomorrows Production, Services and
Work Program. The author is responsible for the content of this publication. Thanks to the reviewers
for their detailed feedback and their valuable literature suggestions.
ICED21 3297
Abramovici, M., Gebus, P., Göbel, J.C. and Savarino, P. (2017), Utilizing Unstructured Feedback Data from
MRO Reports for the continuous improvement of standard products, 21st International Conference on
Engineering Design (ICED17), Vancouver, Aug. 21-25, 2017, The Design Society, Glasgow, pp. 327336.
Bertoni, A. (2020), Data-Driven Design in Concept Development: Systematic Review and Missed
Opportunities, Proceedings of the Design Society: DESIGN Conference, Vol. 1, pp. 101110.
Brocke, J., Simons, A., Niehaves, B., Niehaves, B., Reimer, K., Plattfaut, R. and Cleven, A. (2009),
Reconstructing the Giant: On the Importance of Rigour in Documenting the Literature Search Process,
European Conference on Information Systems (ECIS), Verona, Italy, June 8-10, 2009, pp. 22062217.
Cantamessa, M., Montagna, F., Altavilla, S. and Casagrande-Seretti, A. (2020), Data-driven design: the new
challenges of digitalization on product design and development, Design Science, Vol. 6.
Chowdhery, S.A., Bertoni, M., Wall, J. and Larsson, T. (2020), A data-driven design framework for early stage
PSS design exploration, Design Science.
Erevelles, S., Fukawa, N. and Swayne, L. (2016), Big Data consumer analytics and the transformation of
marketing, Journal of Business Research, Vol. 69, No. 2, pp. 897904.
Erwin, T., Heidkamp, P. and Pols, A. (2015), Creating Value With Data: Report 2015. KPMG and Bitkom
Research. Available at:
Studie-MDWS-final-2.pdf (March 3, 2020).
Fathi, M., Abramovici, M., Holland, A., Lindner, A. and Dienst, S. (2011), Usage scenarios of a knowledge-
based assistance system for decision support in product improvement, 6th Conference on Professional
Knowledge Management - From Knowledge to Action, Innsbruck, Austria, February 21-23, 2011,
Gesellschaft für Informatik e.V., Bonn, Germany, pp. 295304.
Gausemeier, J., Dumitrescu, R., Kahl, S. and Nordsiek, D. (2011), Integrative development of product and
production system for mechatronic products, Robotics and Computer-Integrated Manufacturing, Vol. 27
No. 4, pp. 772778.
Goh, Y.M. and McMahon, C. (2009), Improving reuse of in-service information capture and feedback, Journal
of Manufacturing Technology Management, Vol. 20 No. 5, pp. 626639.
Harzing, A.W. (2007), Publish or Perish. Available at:
(November 12, 2020).
Holler, M., Neiditsch, G., Uebernickel, F. and Brenner, W. (2017), Digital Product Innovation in Manufacturing
Industries - Towards a Taxonomy for Feedback-driven Product Development Scenarios, 50th Hawaii
International Conference on System Sciences (HICSS), Waikoloa Village, Hawaii, USA, January 4-7, 2017,
pp. 4726-4735.
Holler, M., Stoeckli, E., Uebernickel, F. and Brenner, W. (2016a), Towards Understanding closed-loop PLM:
The Role of Product Usage Data for Product Development enabled by intelligent Properties, 29th Bled
eConference on Digital Economy (Bled), Bled, Slovenia, June 19-22, 2016, Association for Information
Systems, AIS Electronic Library, pp. 479491.
Holler, M., Uebernickel, F. and Brenner, W. (2016b), Understanding the Business Value of Intelligent Products
for Product Development in Manufacturing Industries, 8th International Conference on Information
Management and Engineering (ICIME), Istanbul, Turkey, November 2-5, 2016, Association for Computing
Machinery, New York, USA, pp. 1824.
Holmström Olsson, H. and Bosch, J. (2013), Towards Data-Driven Product Development: A Multiple Case
Study on Post-deployment Data Usage in Software-Intensive Embedded Systems, 4th International
Conference on Lean Enterprise Software and Systems (LESS), Galway, Ireland, December 1-4, 2013,
Springer, Berlin, Heidelberg, Germany, pp. 152164.
Hou, L. and Jiao, R.J. (2020), Data-informed inverse design by product usage information: a review, framework
and outlook, Journal of Intelligent Manufacturing, Vol. 31 No. 3, pp. 529552.
Igba, J., Alemzadeh, K., Gibbons, P.M. and Henningsen, K. (2015), A framework for optimising product
performance through feedback and reuse of in-service experience, Robotics and Computer-Integrated
Manufacturing, Vol. 36, pp. 212.
Jun, H.-B., Shin, J.-H., Kiritsis, D. and Xirouchakis, P. (2007), System architecture for closed-loop PLM,
International Journal of Computer Integrated Manufacturing, Vol. 20 No. 7, pp. 684698.
Kiron, D., Kirk Prentice, P. and Boucher Ferguson, R. (2014), The Analytics Mandate, MIT Sloan
Management Review, Vol. 55 No. 4, pp. 125.
3298 ICED21
Lee, E.A. (2008), Cyber Physical Systems: Design Challenges, 11th IEEE International Symposium on Object
and Component-Oriented Real-Time Distributed Computing (ISORC), Orlando, Florida, USA, May 5-7,
2008, IEEE Computer Society, Los Alamitos, California, USA, pp. 363369.
Li, J., Tao, F., Cheng, Y. and Zhao, L. (2015), Big Data in product lifecycle management, The International
Journal of Advanced Manufacturing Technology, Vol. 81, pp. 667684.
Li, Y., Roy, U. and Saltz, J.S. (2019), Towards an integrated process model for new product development with
data-driven features (NPD3), Research in Engineering Design, Vol. 30 No. 2, pp. 271289.
Menon, R., Tong, L.H. and Sathiyakeerthi, S. (2005), Analyzing Textual Databases using Data Mining to
Enable Fast Product Development Processes, Reliability Engineering & System Safety, Vol. 88 No. 2,
pp. 171180.
Ormans, L. (2016), 50 Journals used in FT Research Rank. Financial Times. Available at: (November 12, 2020).
Pahl, G., Beitz, W., Feldhusen, J. and Grote, K.-H. (2007), Engineering design: A systematic approach, 3rd ed.,
Springer, London, UK.
Porter, M.E. and Heppelmann, J.E. (2014), How Smart, Connected Products Are Transforming Competition,
Harvard Business Review, Vol. 92 No. 11, pp. 6488.
Porter, M.E. and Heppelmann, J.E. (2015), How Smart, Connected Products Are Transforming Companies,
Harvard Business Review, Vol. 93 No. 10, pp. 97114.
Rowley, J. and Slack, F. (2004), Conducting a literature review, Management Research News, Vol. 27 No. 6,
pp. 3139.
Shahbaz, M., Srinivas, M., Harding, J.A. and Turner, M. (2006), Product Design and Manufacturing Process
Improvement Using Association Rules, Proceedings of the Institution of Mechanical Engineers, Part B:
Journal of Engineering Manufacture, Vol. 220 No. 2, pp. 243254.
Steenstrup, K., Sallam, R., Eriksen, L. and Jacobson, S. (2014), Industrial Analytics Revolutionizes Big Data in
the Digital Business. Gartner Research G00264728.
Timoshenko, A. and Hauser, J.R. (2019), Identifying Customer Needs from User-Generated Content,
Marketing Science, Vol. 38 No. 1, pp. 120.
Tyagi, S. (2003), Using data analytics for greater profits, Journal of Business Strategy, Vol. 24 No. 3,
pp. 1214.
Ulrich, K.T. and Eppinger, S.D. (2016), Product design and development, 6th ed., McGraw-Hill, New York.
van Horn, D., Olewnik, A. and Lewis, K. (2012), Design Analytics: Capturing, Understanding, and Meeting
Customer Needs Using Big Data, Proceedings of the ASME 2012 International Design Engineering
Technical Conferences & Computers and Information in Engineering Conference, Chicago, Illinois, USA,
August 12-15, 2012, American Society of Mechanical Engineers, pp. 863875.
Webster, J. and Watson, R. (2002), Analyzing the past to prepare for the future: Writing a literature review,
Management Information Systems Quarterly, Vol. 26 No. 2, pp. xiiixxiii.
Wilberg, J., Schäfer, F., Kandlbinder, P., Hollauer, C., Omer, M. and Lindemann, U. (2017a), Data Analytics in
Product Development: Implications from Expert Interviews, 2017 IEEE International Conference on
Industrial Engineering and Engineering Management (IEEM), Singapore, December 10-13, 2017, IEEE,
pp. 818822.
Wilberg, J., Triep, I., Hollauer, C. and Omer, M. (2017b), Big Data in product development: Need for a data
strategy, 2017 Portland International Conference on Management of Engineering and Technology
(PICMET), Portland, Oregon, USA, July 9-13, 2017, IEEE, pp. 110.
Wu, L., Hitt, L. and Lou, B. (2020), Data Analytics, Innovation, and Firm Productivity, Management Science,
Vol. 66 No. 5, pp. 20172039.
Wuest, T., Hribernik, K. and Thoben, K.-D. (2014), Capturing, Managing and Sharing Product Information
along the Lifecycle for Design Improvement, 10th International Workshop on Integrated Design
Engineering, Gommern, Germany, September 10-12, 2014, Chair of Information Technologies in
Mechanical Engineering, Otto-von-Guericke-University, Magdeburg, Germany, pp. 107115.
Xu, Z., Frankwick, G.L. and Ramirez, E. (2016), Effects of big data analytics and traditional marketing
analytics on new product success: A knowledge fusion perspective, Journal of Business Research, Vol. 69
No. 5, pp. 15621566.
... Due to the increasing integration of intelligent components and connectivity in products, more and more data from their use phase is available [1]. A careful investigation of these data with adequate analytics approaches promises new insights into the product itself, the product context as well as the users [2]. In product planning, insights gained from use phase data can help reduce the characteristic uncertainties of this early phase and improve products based on facts instead of assumptions [3]. ...
... For product managers, a particular challenge is the planning of the use phase data analyses [2]. Since the data contains valuable insights, product managers must define promising use cases for analyzing them. ...
Full-text available
Product planning is transforming. For decades, product managers searched for potentials for improvement of their products using methods such as interviews and workshops with customers. Since the information gained with these methods was predominantly qualitative and often also incomplete, product managers also had to rely on their own experience as well as on assumptions about potentials for improvement. However, as a result of the transformation from mechatronic products to cyber-physical systems, a product’s use phase can now be investigated in detail utilizing extensive use phase data. For example, the strengths and weaknesses of the product, as well as the behavior of customers and users, can be observed. The analysis of these data enables product planning based on facts. Currently, however, product managers struggle to identify potentials for improvement resulting from use phase data analyses. In addition to a lack of methods, they especially lack useful examples and use cases for the analysis of use phase data in product planning, which provide them with orientation in the sense of references to plan their analyses. To identify such use cases, we conducted two workshops with 17 product planning and data science experts. This paper presents the results of these workshops: 17 use cases for analyzing use phase data in product planning. Each use case includes exemplary questions which could be answered through data analytics and suggestions on the data required. These suggestions are based on five categories of use phase data that are also derived from the results of the two workshops. Furthermore, each use case is connected to specific elements of value to demonstrate its usefulness and its intended utilization. With these results, we present the first comprehensive overview of use cases for analyzing use phase data in product planning.
... By integrating analytical insights into decision-making processes, existing and future products can be optimized. These concepts form the research area of data-driven product planning (Meyer et al., 2021). ...
Full-text available
Comprehensive data understanding is a key success driver for data analytics projects. Knowing the characteristics of the data helps a lot in selecting the appropriate data analysis techniques. Especially in data-driven product planning, knowledge about the data is a necessary prerequisite because data of the use phase is very heterogeneous. However, companies often do not have the necessary know-how or time to build up solid data understanding in connection with data analysis. In this paper, we develop a methodology to organize and categorize and thus understand use phase data in a way that makes it accessible to general data analytics workflows, following a design science research approach. We first present a knowledge base that lists typical use phase data from a product planning view. Second, we develop a taxonomy based on standard literature and real data objects, which covers the diversity of the data considered. The taxonomy provides 8 dimensions that support classification of use phase data and allows to capture data characteristics from a data analytics view. Finally, we combine both views by clustering the objects of the knowledge base according to the taxonomy. Each of the resulting clusters covers a typical combination of analytics relevant characteristics occurring in practice. By abstracting from the diversity of use phase data into artifacts with manageable complexity, our approach provides guidance to choose appropriate data analysis and AI techniques.
... The products in use can provide designers with valuable information regarding the different usage scenarios, unidentified or misinterpreted user needs, product quality, and defects. Recently, several approaches and frameworks have been proposed by researchers that investigate data-driven engineering design, focusing particularly on the use phase [38]. According to [39] the term data-driven design tends to initiate discussions regarding the use of data to optimise and refine products in relation to easily quantified performance metrics. ...
Full-text available
In the last two decades, data regarding engineering design and product development has increased rapidly. Big data exploration and mining offer numerous opportunities for engineering design; however, owing to the multitude of data sources and formats coupled with the high complexity of the design process, these techniques are yet to be utilised to the best of their full potential. In this study, a comprehensive assessment of the state-of-the-art data-driven engineering design (DDED) in the last 20 years was conducted. A scientometric approach was employed wherein first, a systematic article acquisition procedure was performed, where a dataset of 3339 articles related to engineering design and big data analytics applications were extracted from Web of Science (WoS) and Scopus. Thereafter, this dataset was reduced to a dataset of 366 articles based on concise data screening. The resulting articles were used to analyse the dynamics of research in DDED throughout the last 20 years and to detect the primary research topics related to DDED, the most influential authors, and the papers with the highest impact in the DDED domain. Furthermore, the co-occurrence network of keywords/keyphrases and co-authorship networks were constructed and analysed to reveal the interconnection of the research topics and the collaboration between the most prolific authors. Finally, an insight how big data analytics is being applied through product development activities to support decision-making in engineering design was presented.
... The insights generated from data analytics can reveal potentials for product improvements and help manufacturers of those CPS to optimize product performance and better adapt to actual customer needs [1]. This is an important task in the early phases of product development such as product planning [2]. As a general framework, standard models like CRISP-DM show the relevant process steps of data analytics projects: business understanding, data understanding, data preparation, modeling, evaluation, and deployment [3]. ...
Full-text available
Cyber-physical systems (CPS) generate huge amounts of data during the usage phase. By analysing these data, CPS providers can systematically uncover hidden product improvement potentials for future product generations. The successful implementation of such analytics use cases depends to a large extent on whether the stakeholders involved succeed in coordinating their goals and procedures. In particular, product managers and data scientists must come to a common understanding in the context of defining and concretizing the use cases. A common vocabulary is necessary so that the data scientists or those responsible for analysis can determine target-oriented, analysis-capable use cases with which the processing of the data can start quickly and successfully. The research question that arises at this point is: How can business goals or use cases be translated into realizable analytics use cases or tasks? In this paper we present the Busines-to-Analytics Canvas as a result of an action design research approach. It supports the translation of business use cases and goals into concrete data analytics tasks for product planning. By providing various information elements and guiding questions, the canvas helps data scientists translate the business goal into a data analytics approach, i.e., an algorithm class, and gather the necessary information to start processing data.
... With Manovich (2001), "we are facing the shift of all cultures to computer-mediated forms of production, distribution, and communication". Digitization, therefore, has a deep impact on design culture as it addresses production, organization, and transmission, returning a scenario more relevant than ever for design which is challenged by a profoundly transformed design process (Meyer et al., 2021), operating in an increasingly fluid environment with blurred boundaries between physical, digital, and biological spheres. Yet nowadays it seems that technology is dictating the rules of change: the extent of technological advancement will entail a complex underlying cultural maneuver, since industrial products, both in appearance and in performance, will be placed in contexts in which technology will offer new social, environmental and cultural values. ...
Full-text available
Design processes managed by algorithms provide solutions and improvements in terms of efficiency, performance, choice of materials, and cost optimization. It is a whole new approach to industrial design in which artificial intelligence participates directly in the design processes. The paper aims to investigate the way we design through algorithms, and consequent changes in thoughts, approaches, and generation of ideas that are rising determining new ways of defining things and their relations.
... These data can be investigated with statistical analysis, data mining, and machine learning methods (Hou and Jiao, 2020;Igba et al., 2015). The results of the data analysis lead to new insights about the product and its users (Meyer and Wiederkehr et al., 2021). For the planning of a future, improved product generation, they enable decisions based on facts instead of assumptions (Holler et al., 2017;Wuest, Hribernik and Thoben, 2014;Xu, Frankwick and Ramirez, 2016). ...
Full-text available
The ongoing digitalization of products offers product managers new potentials to plan future product generations based on data from the use phase instead of assumptions. However, product managers often face difficulties in identifying promising opportunities for analyzing use phase data. In this paper, we propose a method for planning the analysis of use phase data in product planning. It leads product managers from the identification of promising investigation needs to the derivation of specific use cases. The application of the method is shown using the example of a manufacturing company.
Full-text available
The successful planning of future product generations requires reliable insights into the actual products’ problems and potentials for improvement. A valuable source for these insights is the product use phase. In practice, product planners are often forced to work with assumptions and speculations as insights from the use phase are insufficiently identified and documented. A new opportunity to address this problem arises from the ongoing digitalization that enables products to generate and collect data during their utilization. Analyzing these data could enable their manufacturers to generate and exploit insights concerning product performance and user behavior, revealing problems and potentials for improvement. However, research on analyzing use phase data in product planning of manufacturing companies is scarce. Therefore, we conducted an exploratory interview study with decision-makers of eight manufacturing companies. The result of this paper is a detailed description of the potentials and challenges that the interviewees associated with analyzing use phase data in product planning. The potentials explain the intended purpose and generic application examples. The challenges concern the products, the data, the customers, the implementation, and the employees. By gathering the potentials and challenges through expert interviews, our study structures the topic from the perspective of the potential users and shows the needs for future research.
Full-text available
Digitalization and the momentous role being assumed by data are commonly viewed as pervasive phenomena whose impact is felt in all aspects of society and the economy. Design activity is by no means immune from this trend, and the relationship between digitalization and design is decades old. However, what is the current impact of this ‘data revolution’ on design? How will the design activity change? What are the resulting research questions of interest to academics? What are the main challenges for firms and for educational institutions having to cope with this change? The paper provides a comprehensive conceptual framework, based on recent literature and anecdotal evidence from the industry. It identifies three main streams: namely the consequences on designers, the consequences on design processes and the role of methods for data analytics. In turn, these three streams lead to implications at individual, organizational and managerial level, and several questions arise worthy of defining future research agendas. Moreover, the paper introduces relational diagrams depicting the interactions between the objects and the actors involved in the design process and suggests that what is occurring is by no means a simple evolution but a paradigmatic shift in the way artefacts are designed.
Full-text available
Ubiquitous and pervasive computing holds great potential in the domain of Product-Service Systems to introduce a model-driven paradigm for decision support. Data-driven design is often discussed as a critical enabler for developing simulation models that comprehensively explore the PSS design space for complex systems, linking of performances to customer and provider value. Emerging from the findings of two empirical studies conducted in collaboration with multinational manufacturing companies in the business-to-business market, this paper defines a data-driven framework to support engineering teams in exploring, early in the design process, the available design space for Product-Service Systems from a value perspective. Verification activities show that the framework and modeling approach is considered to fill a gap when it comes to stimulating value discussions across functions and organizational roles, as well as to grow a clearer picture of how different disciplines contribute to the creation of value for new solutions.
Full-text available
The paper presents a systematic literature review investigating definitions, uses, and application of data-driven design in the concept development process. The analysis shows a predominance of the use of text mining techniques on social media and online reviews to identify customers’ needs, not exploiting the opportunity granted by the increased accessibility of IoT in cyber-physical systems. The paper argues that such a gap limits the potential of capturing tacit customers’ needs and highlights the need to proactively plan and design for a transition toward data-driven design.
Full-text available
New information and communication technologies are changing the way products are developed, manufactured, serviced and managed over the product’s lifecycle. Today’s smart products not only consist of their physical components, but are also endowed with intelligence. Data and the capabilities to process data into knowledge and eventually decisions have become critical components of the product itself and of the process to develop/operate the product. This paper investigates how engineers and a new functional role, data scientists, can effectively collaborate in a mixed team for new product development with data-driven features (NPD³). We focus on the concept development stage, typically the fuzziest phase of product development. In this paper, an integrated process model is explored by revisiting the traditional new product development (NPD) process model as well as the knowledge discovery and data mining (KDDM) process model. Then a case study of the development of an application-specific unmanned aircraft system (UAS) is used to examine the proposed model.
Full-text available
A significant body of knowledge exists on inverse problems and extensive research has been conducted on data-driven design in the past decade. This paper provides a comprehensive review of the state-of-the-art methods and practice reported in the literature dealing with many different aspects of data-informed inverse design. By reviewing the origins and common practice of inverse problems in engineering design, the paper presents a closed-loop decision framework of product usage data-informed inverse design. Specifically reviewed areas of focus include data-informed inverse requirement analysis by user generated content, data-informed inverse conceptual design for product innovation, data-informed inverse embodiment design for product families and product platforming, data-informed inverse analysis and optimization in detailed design, along with prevailing techniques for product usage data collection and analytics. The paper also discusses the challenges of data-informed inverse design and the prospects for future research.
We examine the relationship between data analytics capabilities and innovation using detailed firm-level data. To measure innovation, we first utilize a survey to capture two types of firm practices, process improvement and new technology development for 331 firms. We then use patent data to further analyze new technology development for a broader sample of more than 2,000 publicly traded firms. We find that data analytics capabilities are more likely to be present and are more valuable in firms that are oriented around process improvement and that create new technologies by combining a diverse set of existing technologies than they are in firms that are focused on generating entirely new technologies. These results are consistent with the theory that data analytics are complementary to certain types of innovation because they enable firms to expand the search space of existing knowledge to combine into new technologies, as well as the theoretical arguments that data analytics support incremental process improvements. Data analytics appears less effective for developing entirely new technologies or creating combinations involving a few areas of knowledge, innovative approaches where there is either limited data or limited value in integrating diverse knowledge. Overall, our results suggest that firms that have historically focused on specific types of innovation—process innovation and innovation by diverse recombination—may receive the most benefits from using data analytics. This paper was accepted by Chris Forman, information systems.
Firms traditionally rely on interviews and focus groups to identify customer needs for marketing strategy and product development. User-generated content (UGC) is a promising alternative source for identifying customer needs. However, established methods are neither efficient nor effective for large UGC corpora because much content is noninformative or repetitive. We propose a machine-learning approach to facilitate qualitative analysis by selecting content for efficient review. We use a convolutional neural network to filter out noninformative content and cluster dense sentence embeddings to avoid sampling repetitive content. We further address two key questions: Are UGC-based customer needs comparable to interview-based customer needs? Do the machine-learning methods improve customer-need identification? These comparisons are enabled by a custom data set of customer needs for oral care products identified by professional analysts using industry-standard experiential interviews. The analysts also coded 12,000 UGC sentences to identify which previously identified customer needs and/or new customer needs were articulated in each sentence. We show that (1) UGC is at least as valuable as a source of customer needs for product development, likely more valuable, compared with conventional methods, and (2) machine-learning methods improve efficiency of identifying customer needs from UGC (unique customer needs per unit of professional services cost).
Conference Paper
Product lifecycle management (PLM) is a strategy of managing a company’s products all the way across their lifecycles. Empowered by new capabilities, intelligent products enable seamless information flow and thus enable closed-loop PLM. Hence, one phenomenon of particular interest is the appreciation of beginning of life activities through middle of life information. Grounded on empirical data from a multiple-case study in three distinct manufacturing industries, we explore this emergent role of product usage data for product development. In detail, we address rationales, opportunities, conditions, and obstacles. Findings indicate that (1) heterogeneous motives drive the exploitation, (2) a positive impact on every product development stage is perceivable, (3) some products and industry ecosystems are more suitable than others, and (4) technical, economic, and social obstacles challenge the exploitation. With the limitation of an interpretive, qualitative research design, our work represents a first step to understand the role of closed-loop PLM.