Identifying the right data and AI use cases – the Assessment Framework
This blog post is based on the scientific publication “Creating Value from Data the Right Way – A Framework for Assessing Data Value Creation Use Cases” (Kakuschke et al., 2025).
Suggested citation: Kakuschke, N., Legner, C., & Jung, R. (2025). Creating Value from Data the Right Way-A Framework for Assessing Data Value Creation Use Cases. Proceedings of the 58th Hawaii International Conference on System Sciences. https://hdl.handle.net/10125/109535
Everyone is talking about artificial intelligence (AI) and generative AI (GenAI) in particular. Almost daily, announcements are published in which BigTechs and other AI providers try to surpass each other with bigger, faster and better models and functions in the areas of text, image, audio and video creation. The importance of these developments for companies cannot be denied, as technological developments are capable of revolutionizing the way we work and communicate and significantly increasing the global gross domestic product (Feuerriegel et al., 2024; Goldman Sachs, 2023) . Beyond the hype, executives have agreed for some time that data and its analysis (such as through AI) will play a key role in the transformation of entire business models (Gartner, 2018) and are seen as an investment focus for companies worldwide (IBM, 2023) .
However, to date, many companies have had difficulties in successfully implementing their data and AI initiatives, for example due to a lack of scaling within the organization or the complete failure of projects (Haefner et al., 2023). This limited value creation is due to a variety of challenges from both the technological and organizational side, which can even turn the potential of using data into a burden for companies (Breidbach & Maglio, 2020). Due to the complexity of the perspectives to be considered, companies are constantly faced with the challenge of identifying the specific use cases that are best suited and promise the most value potential, whether classic AI, GenAI or data sales.
Accordingly, one research focus of the Competence Center Future Financial Services (CC FFS) is to provide companies with guidance on how to create value from data and AI. This blog post is dedicated to the decision factors that support the choice of the right data and AI use cases and presents excerpts from a scientific paper that was written as part of the CC FFS research (see Kakuschke et al., 2025) .
Decisions for the right data and AI use cases – the Assessment Framework
In order to make successful decisions on the selection of the right use cases, a basic understanding of value creation from data is first necessary. We assume that both data objects in the sense of raw materials, e.g. as bits and bytes in computer systems, and information as analyzed and contextualized data are valuable assets for companies. Both object types can generate value through their use in the company, for example for decision support, process automation or tailoring marketing campaigns to customers, as well as through their exchange, for example through sale or provision to supply chain partners (Kakuschke, 2024) . This perspective allows to compare a wide variety of use cases in the context of data value creation for companies (see this blog post).
Based on this uniform understanding, the assessment of emerging use cases can be addressed. For structuring purposes, we are guided by an existing framework from design thinking that can be used to weigh up innovation projects (Brenner et al., 2021; Chasanidou et al., 2015). This is based on the three main dimensions of desirability, feasibility and viability, which must be adressed by relevant and ultimately implementable projects. Mapped to data and AI use cases (AI is seen as a method for analyzing data. Accordingly, we also consider analytics and AI cases, in which data is analyzed for value creation, as well as “simple” data cases, in which companies exchange or sell raw data.), this structure helps to categorize the large number of assessment criteria to be considered and reduces the complexity of the decision-making process. The following graphic visualizes the decision-relevant factors for the assessment of data and AI use cases (Note: These factors correspond to bundled categories of a total of 36 identified criteria described in the original publication).
Desirability
In the context of desirability, companies must fundamentally assess at the beginning of the planning and decision-making process whether there is an overarching need for the data and AI use case under consideration. This assessment dimension ensures that the planned use case addresses a real business problem and helps, for example, to consider emerging data and AI hypes in terms of their relevance to your own company. Two decision factors should be taken into account here:
I. Strategic fit: Does the data and AI use case match the company’s strategy hierarchy (corporate, digital, IT, data and AI strategy, etc.)?
II. Demand: Is there a relevant demand for the use case or are existing challenges, both internal and external, addressed?
Feasibility
Once the use case is relevant and desirable, it must be further integrated into the circumstances of your own company. Numerous (AI) use cases present companies with comprehensive challenges for their implementation, implicating that the general technical and organizational feasibility must be included in the decision as a next step. The following factors must be considered here:
III. Technical feasibility: Is the data required for the use case available in sufficient quality and are the basic technologies such as required models and functions available?
IV. Complexity: Can the project be implemented in a reasonable amount of time and complexity?
V. Employee skills: Does the company (departments involved) have the necessary data competence, technological expertise and business understanding to realize the use case?
VI. Data governance: Are the necessary governance mechanisms such as roles and responsibilities, processes and cultural elements for implementing the use case in the company in place?
VII. Risks: How high are the risks associated with the use case, such as project failure or data security aspects?
Viability
Once the desirability and feasibility of a use case have been ensured, it must finally be considered whether it can be designed as a viable business case. A cost-benefit analysis must be carried out:
VIII. Benefits: What advantages can be realized through the use case? It is important to take a holistic perspective here, which includes not only monetary benefits but also other aspects such as risk reduction or the resulting external impact (signaling).
IX. Costs: What costs are incurred as part of the use case? As with the benefit analysis, costs should be included comprehensively along the entire value creation process.
Framing parameters
In addition to these three central dimensions, which primarily affect the company itself, the planned introduction of data and AI use cases must also take into account external influencing factors that can significantly limit their realization or user adoption:
X. External influence: Can the project be implemented in compliance with legal and internal company compliance requirements and are the planned applications ethically justifiable and socially acceptable?
Practical application of the Assessment Framework
The Assessment Framework offers companies a comprehensive overview of the factors relevant to decision-making for the adoption of specific data and AI use cases. Using the model, decisions can be made on a more informed basis as part of an utility analysis, for example in workshop settings, and can be compared with a wide range of use cases from data analytics or (Gen)AI. The model also helps to communicate decisions to internal stakeholders, for example from management, as well as external stakeholders such as company partners in a comprehensible manner. Finally, companies can set priorities for data and AI use cases and deploy their resources in a more targeted manner.
In order to adapt the complexity of the assessment for each company to its own needs, the model is structured hierarchically so that initial decisions can be made based on the dimensions of desirability, feasibility, viability and the framing parameters. For more complex cases with high potential and serious effects on the company, the assessment can also be carried out at the level of the factors presented here or even on the basis of all 30+ criteria. For easy handling the model, the CC FFS has created an applicable assessment tool that can be applied for a wide range of use cases to identify their suitability for your own company.
If you are interested in this or other topics in the Data & AI Value Creation research area, you can contact me at nick.kakuschke@bei-sg.ch.
References
Breidbach, C. F., & Maglio, P. (2020). Accountable algorithms? The ethical implications of data-driven business models. Journal of Service Management,31 (2), 163-185. https://doi.org/10.1108/JOSM-03-2019-0073
Brenner, W., van Giffen, B., & Koehler, J. (2021). Management of Artificial Intelligence: Feasibility, Desirability and Viability. In S. Aier, P. Rohner, & J. Schelp (Eds.), Engineering the Transformation of the Enterprise: A Design Science Research Perspective (pp. 15-36). Springer International Publishing. https://doi.org/10.1007/978-3-030-84655-8_2
Chasanidou, D., Gasparini, A. A., & Lee, E. (2015). Design Thinking Methods and Tools for Innovation. In Design, User Experience, and Usability: Design Discourse. Lecture Notes in Computer Science. Springer. https://doi.org/10.1007/978-3-319-20886-2_2
Feuerriegel, S., Hartmann, J., Janiesch, C., & Zschech, P. (2024). Generative AI. Business & Information Systems Engineering,66 (1), 111-126. https://doi.org/10.1007/s12599-023-00834-7
Gartner. (2018, October 16). Gartner Survey of More Than 3,000 CIOs Reveals That Enterprises Are Entering the Third Era of IT. Gartner. https://www.gartner.com/en/newsroom/press-releases/2018-10-16-gartner-survey-of-more-than-3000-cios-reveals-that-enterprises-are-entering-the-third-era-of-it
Goldman Sachs. (2023, April 5). Generative AI could raise global GDP by 7%. Generative AI Could Raise Global GDP by 7%. https://www.goldmansachs.com/insights/articles/generative-ai-could-raise-global-gdp-by-7-percent
Haefner, N., Parida, V., Gassmann, O., & Wincent, J. (2023). Implementing and scaling artificial intelligence: A review, framework, and research agenda. Technological Forecasting and Social Change,197 , 122878. https://doi.org/10.1016/j.techfore.2023.122878
IBM. (2023, September). IBM Global AI Adoption Index-Enterprise Report. IBM Newsroom. https://newsroom.ibm.com/2024-01-10-Data-Suggests-Growth-in-Enterprise-Adoption-of-AI-is-Due-to-Widespread-Deployment-by-Early-Adopters
Kakuschke, N. (2024). Data Value Creation Matrix-Options for Organizations to Create Value from Data. ECIS 2024 Proceedings,9 . https://aisel.aisnet.org/ecis2024/track07_busanalytics/track07_busanalytics/9
Kakuschke, N., Legner, C., & Jung, R. (2025). Creating Value from Data the Right Way-A Framework for Assessing Data Value Creation Use Cases. Proceedings of the 58th Hawaii International Conference on System Sciences. https://hdl.handle.net/10125/109535
