This paper presents the findings of a formal case study process, where traditional methods are combined with Artificial Intelligence (AI) techniques to develop the prompts for the researcher to develop case study for their research. Based on GenAI technology, the framework covers seven key stages of AI use to develop new knowledge by comparing AI-generated suggestions with those used in classical research to develop a rich qualitative research case study. The key contribution of the research is an explicit characterization of the seven stages, which may enable researchers to develop a robust qualitative case study. Moreover, this work represents the first attempt at using heterogeneous, AI-generated prompts to develop a case study following the seven stages of case development study to develop rigorous research. It introduces a well-rounded toolbox that details the full research workflow, from idea generation to completion. This combination will not only catalyse studies of greater veracity but will also raise the standard in terms of using an AI-assisted methodology in case studies.
The case study research strategy has come to be increasingly popular since it is seen as a well-structured approach that critically investigates complex issues and real life situations (Gerring, 2004). Yet, because of the evolution of Gen AI , it is increasingly clear that tools such as ChatGPT can be used in case study research productively. Recent developments in qualitative methodologies using AI have been discussed by Naeem et al. (2025), who introduced a Gen AI toolkit to carry out thematic analysis. They set out to explore a specific research question, which was: “how can AI be used to develop case study as foundation of systematic thematic analysis in the qualitative research?” (Naeem et al., 2025, p. 16). This paper sets out to understand how AI assisted case studies can be developed. The AI prompts that are discussed have been sourced from different authored works, and the generated outputs at each stage are compared with Naeem’s (2025) findings to illustrate the contrast between human developed case studies and those developed by AI.
The paper therefore introduces something of an instruction manual to consisting of a 7-stage framework that sheds light on the content, philosophy, scope, theory, data, case and selection of units of analysis for AI assisted case study research. Key factors and considerations at each stage are presented in tables to compare AI generated response approaches with human analysis (Naeem, 2025). The paper therefore shows how AI can be employed in the design of research, with the development of systematic procedures to incorporate various points of view in developing AI assisted case studies for qualitative research. The systematic case study development approach suggested here creates the possibility to use AI to develop robust qualitative case study.
Methodological Framework for AI-Enhanced Qualitative Case Study Development
The toolkit illustrated in Figure 1, it is designed for qualitative case study enquiry. However, it takes into account the fact that the case study is heterogeneous and can be adapted to quantitative research using the same steps. The seven-step toolkit was developed by comparing AI-generated outputs at each stage with evidence from Naeem’s (2025) study. This intersection of each step is a useful illustration of the potential of AI to reinforce what we already know about research stages, and how to incorporate traditional views into new ways of doing case studies. The model identifies a systematic, step-by-step case study process with the assistance of Gen AI.
Systematic Step-by-step Qualitative Case Study Develop Process With Assistance of Gen AI)
The first stage provides the necessary context to identify research gaps and to map the study into a relevant context. Stage 2 provides the groundwork for the articulation of the research philosophy, in which the research is grounded, and in which researchers locate themselves within an ontological and epistemological position. This position informs the research process which is relevant to the theoretical position of the study. Stage 3 involves scoping the case study, and determining its width and breadth, appropriateness to the data requirements and questions. This scope, on which the philosophical bases have an impact, informs the choice of theoretical models or underpinning in Stage 4. In this approach, the theories that offer the best explanations for the phenomenon that are the focus of study, thereby contributing to the development of the theoretical framework through limit to specific theory and specific factors of the selected theories.
During Stage 5, the type of case study is selected with full consideration given to the theoretical lines and research goals which shape approaches to data collection, analysis, and the presentation of findings. Stage 6 involves selecting specific cases that can provide rich and meaningful data. This follows on from the first five stages of case study development which involve selecting the number of cases and relating these to the steps take as part of stage five. Finally, stage 7, configures the unit of analysis in a way which limits the analysis to elements of interest to the study. This is influenced by all prior decisions in a focused analysis of the cases chosen. All stages are dynamically interconnected to increase the coherence and depth of research. AI is used as part of this process to support the use of traditional methods, potentially enabling deeper insights.
How to Use the Developed Framework: AI Assistance in Case Study Development
The above framework (Figure 1) setting out seven steps to using AI can be used by researchers to develop Gen AI assisted case study development. Based on the above process, AI prompts have been developed based on various points of view espoused by authors. These provide insight into the most important factors that need to be considered during each stage of developing a case study. Additionally, following each stage, the researcher is required to introduce their research context in order to effectively use each prompt to perform each stage of the analysis. Therefore, guidelines are provided as to what steps must be taken at each stage of the process. Once the researcher receives responses to the prompts, they should carefully review the output, taking into account the specific factors relevant to each stage. Any stage may be repeated if the response is not appropriate, in order to achieve the objective of that stage. Additionally, researchers need to carefully understand what different authors have advised at each stage to identify any missing elements. Figure 1 presents a circular diagram illustrating the steps from stages 1 to 2. It covers important factors that should be considered at each stage. The circular shape is intended to illustrate that researchers can repeat the process to ensure that they arrive at a more informed result, particularly during the second or third round.
Case Development Process to Study
Case Study
A case study is a research strategy which focuses on empirical research of a given phenomenon within its real-life context using multiple sources of evidence. It is especially useful when boundaries between phenomena and context are not clearly evident (Flyvbjerg, 2006; Yin, 2014). It involves ‘an intensive, simultaneous, multi-dimensional study of a single instance or phenomenon’, or of a small number of related phenomena (Stake, 1995). According to Harrison et al. (2017), case studies enable an in-depth analysis of complex situations while also allowing insight into the relative dynamics during each instance of research (Jarvensivu & Tornroos, 2010).
The main goal of a case study is to produce a high-quality, in-depth analysis of a specific issue on its own terms, rather than allowing for any comparisons to other contexts or settings. It has great value, especially for exploratory research, as it is can help researchers to identify variables and create hypotheses to support further examinations (Crowe et al., 2011; Kaplan & Maxwell, 1994). In descriptive and explanatory cases, studies provide an in-depth empirical account of phenomena through the illuminating of underpinning mechanisms, or the discovery of causality (Yin, 2014). The intrinsic case study sets out to understand the case itself, since it is unique because of its specific critical intention. The instrumental case study, for contrast, is used to shed light on a more general theoretical question or problem (Stake, 1995). There is a particular advantage to be gained when conducting case study research in a specific context-dependent discipline Stake (1995). The selected case has to be contextually dense, providing a chance to investigate fine-grained social, cultural, or organisational dynamics, all consistent with Stake’s (2000) desire to centre thinking on the case - and its own particularities. This contrasts with the use of fixed, predetermined methodological determinism. Stake (2000) suggested that the case should be rich in context, thus there is a need to understand the research context to develop a better case study. Figure 2 below illustrates how philosophical and theoretical underpinnings form the foundational core of case study research. Surrounding this foundation are the five distinct case study types—exploratory, explanatory, descriptive, intrinsic, and instrumental—which guide the research purpose. The next layer presents decisions around the number of cases, helping researchers determine whether a single, multiple, or cumulative case study design is most appropriate. The outermost ring emphasises the importance of contextual understanding, including scope structuring, the nature of the social phenomena, and the broader social setting. Together, these elements help researchers align their case study design with the nature of the research context, while also ensuring that design decisions are well justified and theoretically grounded. This paper focuses on how to develop a case to study and identifies the factors which must be considered in doing so. It also outlines the process that can be applied to the development of a case study using GenAI, and specifically ChatGTP.
Conceptualization of the Development of a Case Study (Author’s Own Work)
Integrating AI in Case Study Development: A Detailed Justification and Staged Approach
Stage 1: Contextual Understanding
The first phase of the process is the development of a contextual understanding, which includes an in-depth analysis of the environment, context and circumstances surrounding what we want to know (Baxter & Jack, 2008). Carter (2022, p. 2) indicates “the first step in utilising case study was a journey into knowing and understanding that there are different trains of thought about case study”. This process emphasizes a sequence of events, and how these events are influenced by contextual factors. These factors explain why actors behave as they do, and how cases can be replicated elsewhere (Yin, 2018). Moreover “the first step in any research, and particularly in case research, is to specify as clearly as possible the research question(s) to be addressed. This will usually be done in conjunction with a review of the existing research literature. The research question(s) will then shape the research design, including the research methods and even the methodology. The question(s) should be sufficiently focussed to provide a feasible research plan, given the available resources, especially time” (Humphrey & Lee, 2004, p. 265). At this point, cultural factors, social classes, social statuses, and social settings also play a major role in shaping the socio-economic terrain that emerges from the research (Simons, 2009; Stake, 1995). Additionally, political and legal environments also impact the possibility of conducting or changing certain subjects’ behaviours and actions which impact on the selected case study (Simons, 2009; Stake, 1995).
Institutional context and history are also important to understand current behaviours and responses when interpreting meaning (Simons, 2009; Stake, 1995). Additionally, the technical and geographical context of the study should be considered. These factors constitute the material, spatial, and symbolic instruments and the setting of the research, (Simons, 2009; Stake, 1995). A pivotal differentiation in social research sits between the ‘social setting’ (the context in which interactions occur under specific social circumstance) and the ‘social phenomenon’ (the action being exhibited in these settings to achieve outcomes). The latter encompasses physical and historical circumstances that enable social facts, which comprise both behaviours and social actors which exhibit human behaviours (Davis, 2017; O’Neil, 2000; Peterson, 2012; Sanders, 2010).
Table 1 summarises various authors’ opinions (column 1) on classic studies that speak to the contextualisation of research. The first columns contain thoughts on what should be included in the context of a given study, and the second identifies the prompt which has been developed on the basis of each of these views. The third columns show the AI response to the prompt given in column 1, while column four summarises the information. The comparison reveals that most of the factors suggested by AI were addressed in the case study, with notable differences in the contextual factors outlined by AI (as shown in column 3 of Table 1). In contrast, Naeem (2025) focused on exploring technological affordance, whereas the AI-generated theoretical suggestions were limited to themes of technology adoption rather than affordance (as shown in column 4 of Table 1). This highlights a key limitation of AI, which tends to rely on the most common research patterns in the field. It reinforces the need for researchers to contribute their own contextual insight to ensure relevance and depth in the study design.
Contextualization Stage: Case Study Development Through AI-Assisted
Literature contextual factor
ChatGPT prompt
AI suggestion to select the contextual factor of scan and go related factors
Contextualization of the selected case study research (Naeem, 2025)
“I am preparing a case study for my research and need to thoroughly contextualize it based on several key environmental factors. Could you help draft a detailed contextualization considering various essential aspects? Here are the specifics needed for a comprehensive understanding of the case study’s setting • Research Rationale: [Insert the general motivation behind the research] • Research Gap: [Insert the specific gaps in the existing literature that this research aims to address] • Research Objectives: [Insert clear objectives that the research aims to achieve] Based on the above details, integrate the following contextual factors into the case study 1. Cultural Norms: How do the cultural norms influence or relate to the case? 2. Economic Stability: What is the economic context surrounding the case? 3. Political and Legal Environments: What are the political and legal factors that could affect the study? 4. Social Structures: How do social structures within the case impact the subjects or outcomes? 5. Technological Environment: What role does technology play within the case context? 6. Physical Setting: Describe the physical or geographical setting of the case 7. Historical Background: What historical factors are relevant to the case? 8. Institutional or Organizational Setting: What are the characteristics of the institution or organization involved in the case? Please draft a contextualization that integrates these factors, focusing on their implications for the case study based on the research rationale, gap, and objectives.” (ChatGPT developed prompts)
Consumer privacy concerns: Cultural attitudes towards privacy can affect acceptance of S&G apps, as users may be wary of how their data is handled and stored
In-depth discussion of consumer privacy concerns highlights that cultural attitudes toward privacy may substantially influence the adoption of S&G apps, consistent with AI literature on the impact of cultural factors on consumer behaviours towards personal data-handling technologies
Economic accessibility: Economic conditions influence the adoption of S&G technology, especially if it leads to perceived cost savings or financial management benefits during shopping
Explores the impact of economic factors on the S&G technologies adoption rate, emphasising the perceived benefits in terms of saving money and financial control when shopping
Regulatory compliance: Compliance with GDPR and other privacy regulations in the UK affects the implementation and operation of S&G systems
GDPR and general data privacy regulations are a driving factor for S&G systems implementation, which illustrates practical issues in the operationalization of data privacy in different regulatory environments
Inclusivity in technology use: The design and functionality of S&G apps must account for varying levels of technological literacy across different social groups
Inclusive use of technology: Highlights the necessity for S&G apps to meet the requirements of different social groups (characterized by different levels of technological literacy), and underlines the importance of inclusive design and use
Technology Adoption and integration: The existing technological infrastructure of supermarkets influences how effectively S&G apps can be implemented and integrated into daily operations
Reflects on the necessity of the current technological infrastructure of supermarkets in order to support the proper implementation and enhancement of S&G applications
Store Layout and design: The physical layout of stores affects how intuitively customers can use S&G apps, impacting their overall shopping experience and efficiency
Explore the impact of a physical store layout on customer interactions with S&G apps, as well as the influence these have on their shopping experience and operational efficiency
Legacy of shopping technologies: Past experiences with technology, such as self-checkouts, can influence consumer readiness and openness to adopting S&G technologies
Investigates the role of consumers’ preparedness and willingness to use S&G applications, given their previous exposure and experience of technologies such as self-checkouts
Institutional or organizational setting (Simons, 2009)
Corporate Strategy and support: Institutional support for innovation determines the speed and extent to which S&G technologies are rolled out across supermarket chains (ChatGPT response)
There is a clear indication that corporate strategy and company support play an important role in the speed and level of the S&G technology roll out, based on the level of support, adoption and integration
Stage 2: Philosophical Underpinning
Philosophical dimensions influence designing. For example, social constructivism and interpretivism are more aligned with the relativist ontological position which has an impact on the case study (Rashid et al., 2029). Moreover, “The first step in utilising a case study was a journey into knowing and understanding that there are different trains of thought about case studies. In exploring these different trains of thought, I needed to consider which one best linked to my educational field of study and my philosophical orientation” (Carter, 2020, p. 2) Therefore, it has been suggested that to achieve depth and consistency, researchers need to attune their ontological and epistemological positions with a research paradigm (Denzin & Lincoln, 1998; Wilson, 2010). Ontology is concerned with the nature of the reality that is being investigated (Denzin & Lincoln, 1998; Morehouse & Maykut, 2002). Epistemology, in contrast, generally deals with inquiry into knowledge, the nature of knowledge, acceptable knowledge, and the foundation of how we know (Guba & Lincoln, 1994; Morehouse & Maykut, 2002). The relativist ontology is frequently used in qualitative research, which suggests that reality is multiple. This can help the researcher to explore how realities are developed and interlinked, and how these realities are constructed through social experiences. Therefore, any interpretation of reality is influenced by societal conventions and cultural conditions (Crotty, 1998; Guba & Lincoln, 2005) which is why social and cultural factors should be considered in case study contextualisation.
Qualitative methods are driven by interpretivism, which focuses on how and why people experience the world differently. The former seeks to unveil social hierarchies and the possibility for transformation. It thus “…recognises the importance of the subjective human creation of meaning, but it does not reject outright some notion of objectivity” (Crabtree & Miller, 2023, p. 10). The latter is concerned with understanding something through subjective experience and consideration of the situational context (Guba & Lincoln, 2005; Kaplan & Maxwell, 1994 ; Meyers, 2004; Orlikowski & Baroudi, 1991). Informed by social constructionism, reality is constructed through cultural norms and social interactions. Merriam (1998, p. 6) indicates “the key philosophical assumption upon which all types of qualitative research are based is the view that reality is constructed by individuals interacting with their social worlds”. Therefore, researchers need to explore how the world has been experienced on the basis of their social experience and social setting (Berger & Luckmann, 1966). In this sense, there is some reliance on understanding the context and integration of multiple participant voices and points of view using different data collection tools, such as interviews and focus groups (Creswell, 2013; Hutchby, 2001; Merriam, 1998; Yin, 2013). The selection of a philosophical position, therefore, crucially frames the way researchers interact with participants and interpret the findings through the participants’ language (Naeem et al., 2023). This underscores the importance of narratives and stories in the process of understanding data (Guba & Lincoln, 1985; Stake, 1995).
Table 2 was generated by synthesizing the opinions of different authors about philosophical issues, and the importance of the philosophical underpinning of a case study. The first column provides different viewpoints as the foundation for AI prompts using ChatGPT. The second column presents different authors’ perspectives that should be considered as the philosophical underpinnings of the research, forming the basis for the second stage of case study development. The third column provides prompts from the assigned authors to consider its responses on the basis of interpretivism or social constructivism. The fourth column shows some of the recommendations generated by ChatGPT. The same research context was provided in each case, and the AI suggested what should be taken into account on the bases of the selected philosophical positions. The last column shows how Naeem (2025) handled his study without AI support, and this is provided for philosophical underpinning stage.
How ChatGPT Can Assist in Suggesting Philosophical Underpinnings to Set the Foundation of a Case Study
Philosophical factor
Rationale ChatGPT prompt
ChatGPT prompt
AI (ChatGPT) suggest how to case study factors to Be explored?
Application to S&G technology study
Ontological considerations (subjectivism and relativism)
• Ideas and experiences of individuals: Investigate the way in which individual unique experiences shape our understanding of phenomena (Thompson et al., 1989) • Situational influence: Reflect on how one phenomenon could be experienced differently in varied situations or cultures, given the subjective nature of reality (Thompson et al., 1989)
“I am conducting a case study and need to establish a strong philosophical underpinning tailored to the unique aspects of my research. Could you assist in identifying and justifying the most suitable philosophical approaches based on the specific characteristics of my study? Here’s what needs to be included • Research rationale:[Insert the general motivation behind the research] • Research gap:[Insert the specific gaps in the existing literature that this research aims to address] • Research objectives:[Insert clear objectives that the research aims to achieve] Please analyse and integrate the following philosophical dimensions into the case study’s foundational approach 1. Ontological considerations: o individual Perceptions and experiences:How do individual experiences within the study’s context influence the understanding and outcomes of the research? o contextual influence:How might the phenomenon be perceived differently across various environments or cultures? 2. Epistemological frameworks: o constructed knowledge:How is knowledge within the case constructed through social interactions? o role of the researcher:What impact might the researcher’s perspective have on the study? 3. Phenomenological approach: o lived experiences:What are the core experiences of participants that must be understood to grasp the phenomenon fully? o depth of interaction:How can the study capture the deep, subjective impact of these experiences? 4. Reflexivity and inclusivity: o researcher Bias and inclusivity:How will the study address potential biases and ensure a diversity of perspectives? o diverse perspectives:How will the study incorporate and respect the range of viewpoints within the research environment? 5. Interpretivism and critical theory: o contextual realities:How do the social, cultural, and technological contexts influence behaviours within the study? o power dynamics:What power structures need to be understood or challenged to fully interpret the research findings? o understanding meanings:How will the study explore the meanings that participants ascribe to their experiences and circumstances? Based on these considerations, please recommend a philosophical approach or a combination of approaches that would best support the case study’s objectives, filling gaps, and aligning with the overall research rationale”
Personal Perceptions - Explore how individual beliefs and subjective perceptions of S&G apps affect the user experience - Analyse how personal expectations and previous experiences with technology shape interactions with S&G apps Contextual Variability - Assess how different shopping environments and contexts influence the use and perception of S&G technologies - Identify how situational factors like store layout, app design, and in-store promotions impact consumer behaviour (ChatGPT response)
The investigation provides an insight into how different consumer experience with S&G technologies resulting in differing frustrations and preferences based on personal and situational factors e.g. prior tech experience or store arrangement
• Knowledge-in-the-Making: Explore how knowledge is enacted in everyday activity, how shared beliefs and stories can affect understandings and practices (Berger & Luckmann, 1966; Wojnar & Swanson, 2007)
Shared Understandings - Investigate the collective norms and shared beliefs about technology within the shopping context (ChatGPT response) - Examine how cultural and social narratives around retail technology influence consumer acceptance and resistance to S&G apps Interpersonal Interactions - Study the role of social interactions in shaping consumer experiences with S&G apps, including interactions with other shoppers and store staff - Explore how communication and social influence at the point of sale affect the perception and efficacy of S&G technologies (ChatGPT response)
The research considers how communal S&G technology experiences become collectively held perceptions or myths regarding its effectiveness and usefulness, as well as the extent to which the research contexts informed the shaping and meaning making of the experiences
Phenomenological approach
• Experiences of life: Explore how people perceive and experience their interactions in context; look for detail and personal interpretation and meanings (Wojnar & Swanson, 2007) • Depth of interaction: Examination of the depth or subjectivity of influences in people’s lives to capture the essence of people’s being in the world (Wojnar & Swanson, 2007)
Lived Experiences - Capture detailed accounts of consumer journeys using S&G apps, focusing on the emotional and cognitive aspects of the shopping experience - Delve into personal narratives to understand the impact of technology on achieving shopping goals, such as convenience, savings, or empowerment Essence of Interactions - Analyse how consumers describe their interactions with S&G apps, highlighting moments of satisfaction, frustration, or confusion - Investigate the subtle nuances of how the app’s functionality aligns or conflicts with consumer expectations and shopping habits (ChatGPT response)
The research addresses the fact that the experience of customers using S&G-technologies is subjective and varied and how this affects the shopping experience and interaction with technology
Reflexivity and inclusivity
• Multiple facets: Attempt to capture the phenomenon being analysed from multiple perspectives for thorough research by presenting underrepresented perspectives of the majority of people (Creswell, 2013; Yin, 2013)
Diversity in Participation - Ensure inclusion of a broad demographic spectrum in participant selection to capture diverse consumer experiences with S&G apps - Aim to include varied socio-economic backgrounds to understand different accessibility and usage patterns of S&G technologies Researcher Bias and Reflection - Continuously reflect on and critically evaluate the researcher’s own biases and how these might influence the research process and interpretations - Encourage reflexivity in the analysis to understand how the researcher’s perspective might shape the findings and to maintain objectivity (ChatGPT response)
A diverse set of demographics are represented in the study, ensuring that the data represents a spectrum of consumer interactions with S&G technologies, while the researchers are above all conscientious in reflecting on their own prejudices to data interpretation
Interpretivism and critical theory
• Situated practices: Examine how what people do is constructed by their social, cultural, and technological contexts (Kaplan & Maxwell, 1994; Meyers, 2004) • Power dynamics: Recognize and critique power relations and imbalances in the power relationships or power differentials made evident in the research; consider how they impact the phenomenon being studied (Kaplan & Maxwell, 1994; Meyers, 2004) • Interpreting meanings: This method helps us to understand how participants make sense of their situations and experiences (and, by extension, underlines the importance of this kind of work when trying to make meaningful contributions to ‘complex problem settings’ such as consumption, education or organizational change) (Meyers, 2004)
Contextual Realities and Consumer Behaviour - Explore how contextual factors such as retail environments, technology interfaces, and social norms shape consumer behaviours and experiences with S&G apps - Study the everyday realities of consumers, focusing on how they perceive and interact with S&G technology during their shopping journeys Power Dynamics and Consumer Autonomy - Investigate how S&G apps affect consumer decision-making autonomy, particularly how technology may dictate or restrict consumer choices - Assess any emerging inequalities or power imbalances between consumers and retailers, such as privacy concerns or data control, and how these impact consumer trust and technology acceptance. (Colum is exactly the response of ChatGPT)
N/A
Based on this experience (Table 2), it is clear that AI offers precise recommendations linked to various philosophical assumptions it is oriented towards when creating case studies. For example, AI suggested exploring contextual and individual influences on the use of pre-scanners and the perception of Scan and Go (S&G) technologies, including personal and situational factors (i.e., individual beliefs, prior technology experiences, and the effects of the shopping environment). Naeem’s (2025) research also considered modelling different consumer-perceived interactions based on experience and context. Examples here include store layout and development background (i.e., past technology usage). However, Naeem did not explicitly refer to the methodological framework proposed by AI. As illustrated in Table 2 below, AI recommended exploring power dynamics to better understand meaning and context—something not mentioned in Naeem (2025). This highlights the potential of AI to enrich research by offering more comprehensive and contextually sensitive suggestions that may not have been employed in traditional research approaches.
Stage 3: Scope Structuring
Contextualisation and scoping are equally important for research, and so placing a case study within a societal, cultural or historical context adds depth and specificity. Merriam (1998, p. 27) indicates that, “I can ‘fence in’ what I am going to study”. At this stage it is necessary to be clear about what the case is, and what should be included or omitted. Some variables include geographic, temporal or thematic delimitations that are pertinent to the phenomenon under study (Baxter & Jack, 2008; Merriam, 1998). The scope of triangulation between interviews, observations and documentary reviews provides a deeper understanding of the case. Theories do not direct the collection of data per se; however, theories direct how data are interpreted, and ensure analytic rigor (Creswell, 2013; Flyvbjerg, 2006; Stake, 2000). “Take care when defining the case because the research questions reflect the boundaries and definition of the case” (Schoch, 2020, p. 247). As such, the number of theories and theoretical factors in case study research need to be limited. It is clear, however, that through the triangulation of information, the findings are more likely to converge from various sources, and a case report should provide limitations as to the type of data and case study that is being studied (Baxter & Jack, 2008; Merriam, 1998). According to Yin (1994, p. 23) “as an empirical inquiry that investigates a contemporary phenomenon within its real-life context; when the boundaries between phenomenon and context are not clearly evident; and in which multiple sources of evidence are used”. Therefore, research must establish clear boundaries around the real-life phenomenon under investigation.
As Creswell and Flyvgyerd (2006) propose, theory drives a case study, as the type of methods which are used (qualitative, quantitative, and mixed) to collect and analyse data will be determined based on theory. “What is the case? What is part of, and what is not part of, the case?” (Schoch, 2020, p. 246) therefore, at the scoping stage, researchers need to decide what is included in the study and what falls outside its scope. The researcher therefore needs to limit research to the context of the methods that are delineated. Additionally, a ‘reality check’, resources, and feasibility can all serve to limit the extent of the research, and thus its adherence to ethical principles. Such researcher reflexivity defines the boundaries of research (Denzin & Lincoln, 1998; Villiers et al., 2019). Therefore, study scope involves explicitly explaining and delimiting the characteristics of the study population, including the main focus of the investigation (theme or topic), the specific focus of the subject area, and the temporal and spatial limits of the case (Baxter & Jack, 2008; Simons, 2009; Yin, 2009). Temporal boundaries refer to the time frame in which the study is conducted and the period covered by the research. Spatial boundaries relate to the geographical scope, physical setting, environment, and the selection of the number and types of organisations, all of which help to localise the case study.
Likewise, it is important to be able to learn about the volume of research in the context of the contribution, and its focus. It also pays to ensure that the study is manageable and practical so that it meets its own goals. “Case study research offers benefits in terms of process and outcome. The case study design will help you focus your research within the confines of space and time on a specific case” (Schoch, 2020, p. 246). This scoping framework also covers the subject field, number of participants, time, space, types of participants, types of industry, number of companies, the location of the companies, and any relevant restrictions. These also enable some consideration and interpretation of the findings, and allow the researcher to maintain a degree of context specificity. . Additionally, Stake (2000, p. 435) states that “Case study is not a methodological choice but a choice of what is to be studied. We concentrate, at least for the time being, on the case”. This increases the validity and practical implications of the study (Eisenhardt & Graebner, 2007; Garcia et al., 2019). Secondly, the range must encompass practical, theoretical and descriptive aspects to arrive at a full understanding of the items. Indeed, the items must be suitable and applicable to a comparable context.
Table 3 outlines the embedding of AI tools, e.g., ChatGPT, within the scope of the planning stage. It also frames the case and underscores its relevance to the research aims and methods. The first column provides a justification for the development of prompts on the basis of different author viewpoints. The second column presents different authors’ perspectives that should be considered as the scope of case study, forming the basis for the third stage of case study development. The third column provides prompts from the listed assigned authors which task AI to think about what has to be taken into account given the case study scope position. The fourth provides some recommendations from ChatGPT. Again, the research context was disclosed to the AI along with the selected philosophical position. The last column shows how Naeem (2025) handled his study without AI support, and this is provided for scope stage. The suggestions generated by AI (column 4 in Table 3) included a concrete and well-defined study scope, whereas the scope in Naeem’s (2025) study was scattered (column 5 in Table 3). Therefore, the use of AI could help in developing a more structured and specific study scope.
Enhancing Case Study Design With AI: Scope Structuring Guidance
Scope structuring factors
Rationale ChatGPT prompt
ChatGPT prompt
AI (ChatGPT) guidelines for scan & go app research
Scope factor consideration in the study
Research objectives
Alignment of the objectives: The objectives need to be consistent, and align with the overall focus of the study, so that there is a clear course of inquiry (Yin, 2009)
“Based on the details of a planned case study, I need to rigorously define the scope of the research to ensure it is well-structured and focused. Could you assist in delineating the scope across various parameters based on the following requirements? Please integrate and tailor your response to the unique aspects of this research 1. .Research Rationale: [Insert the underlying reasons or needs that prompted the research.] 2. Research Objectives: [Insert specific, measurable, and achievable goals of the research.] Please include the following scope structuring factors in your analysis • Objective Alignment: Ensure the study’s objectives are aligned with its broader aims • SMART Criteria: Define objectives that are Specific, Measurable, Achievable, Relevant, and Time-bound • Demographic Detailing: Specify the demographic characteristics of the target population, including inclusion and exclusion criteria • Sample Diversity: Aim to capture a wide range of perspectives to enhance the generalizability of findings • Thematic Focus: Provide clarity on the thematic focus and balance the scope to maintain research effectiveness • Spatial Boundaries: Define the geographical limits and consider environmental influences on the study • Qualitative Methods: Select methods that align well with the research questions and objectives • Data Collection Techniques: Specify techniques like interviews, observations, and focus groups for rich data gathering • Research Outcomes: Clearly specify expected outcomes and their potential impact on the field • Subject Area: Define the field of study and align the research with current debates and gaps • Research Design and Methodology: Detail the research design and methodologies to be employed • Theoretical Framework: Choose and justify a theoretical framework to guide data analysis and interpretation • Boundaries: Establish geographical, technological, sector, temporal, user perspective, conceptual, and methodological boundaries as appropriate Based on these criteria, could you recommend a structured approach for defining the scope of this case study, ensuring it is tightly aligned with the established objectives and the broader research agenda?” (ChatGPT prompt)
“Define clear objectives:Focus on specific aspects of user interaction with Scan & Go technology, such as the impact on shopping speed and consumer decision-making Measure user satisfaction:Develop measurable goals to assess changes in consumer behaviour due to app usage Diverse user groups:Include a variety of users from different demographic backgrounds to understand different user experiences with Scan & Go apps Include Frequent and infrequent users:Study both frequent and occasional users to explore diverse experiences and satisfaction levels” ” (ChatGPT response)
One of the objectives was to explore the negative aspects of customer experiences of S&G apps, which resonates well with the overall purpose of achieving better customer experience management in the digital retail setting. This was specific, linked, and easily demonstrable by the research design
Study population
Population characteristics: Define the target population such as the demographic details of the intended audience, with precise inclusion and exclusion criteria in mind (Jones et al., 2013). Diversity of sample: Strive for a diverse sample in order to include a variety of viewpoints and increase the general visibility and implication of the research findings
The sample was comprised of S&G app users but there was no detailed information about the demographic dimensions of the app users; the study used a variety of data collection approaches, online reviews, interview and focus groups, and presented a mixed sample in terms of population
Thematic focus
Topic clarity: Have a clear sense of the topic in mind to mitigate confusion, guide the research pursuit, and achieve the expected outcomes (Merriam, 1998)
“Focus on technological challenges:Concentrate on issues such as privacy concerns, app reliability, and user interface design User experience journey:Explore how Scan & Go technology affects the shopping journey from entry to checkout” ” (ChatGPT response)
Concentrate on ‘dark experiences’ (the negaffordances) of S&G apps to preserve the overall coherence of the theme, and explore the possibility of proceeding with a complex, subtle interplay of technology and user experience
Spatial boundaries
Scope and relevance: Achieve an optimal level in terms of the depth and breadth of the topic to allow complex interconnections around the topic to be traced Geographical localization: Describe the geographical or physical scale of investigation of the study to locate it in a particular area (Baxter & Jack, 2008; Simons, 2009)
“Select specific retail settings:Focus on urban versus rural supermarkets or explore regional differences in technology adoption within the UK Environmental impact on usage:Study how different store layouts and sizes influence the effectiveness and user satisfaction with Scan & Go apps” ” (ChatGPT response)
Scope of the geographic coverage: The exact geographic scope was not well defined, but retail environments, which employ S&G technology were considered, with no specific geographic area
Qualitative methods
Choice of method: Select specific qualitative methods appropriate to the researcher’s questions and purpose (Creswell, 2013; Stake, 1995). Includes the particular qualitative techniques and procedures employed in the study (e.g., interviews, observation, thematic analysis)
“In-depth Interviews and observations:Conduct structured interviews with users immediately after shopping and observe interactions with the app in-store Focus groups:Organize focus groups to discuss common issues and improvements with diverse user groups”
Phenomenological orientation, thematic analysis, semi structured interviews and focus groups, were selected as suitable for the type of study needed to explore subjective experiences, and to obtain detailed insights in connection with the subject under consideration The aim of this research was to get nearer to the users’ experience; therefore a phenomenological approach was followed. Method: Thematic analysis as described in the study and the chosen methodological approach
Research outcomes
Outcome definition: Define the expected outcome a research aims from the research, either theoretical, practical, or descriptive (Yin, 2018)
“Practical recommendations:Provide actionable suggestions for enhancing user experience based on research findings Theoretical contributions:Develop a model of consumer behaviour specific to interactive retail technologies like Scan & Go” (ChatGPT response)
Results: Implications for theory and practice by extending understanding of S&G app effects on customer experience, and provide recommendation for digital retail technology To explain how S&G apps enable or inhibit shopping purposes, influence the purchase intention or the occurrence of NEUs when using the apps. For this purpose, affordance theory was adopted
Subject area
Scope: Establish the field, domain or discipline to which the research contributes (Adams &others, 2014) and demonstrate the application of theoretical and practical relevance of the selected field Impact and contribution: Explain how the results indicate the potential impact of the findings
“Retail technology focus:Situate the study within current discussions on retail innovation and technology’s impact on consumer behaviour Link to consumer psychology:Explore psychological factors influencing the adoption and resistance to new shopping technologies” ” (ChatGPT response)
This study covered the subject area in the context of customer experience and the affordance of the selected technology, but exploration of consumer psychology was missing
Research design and methodology
Design choice: Specify the research design (exploratory, descriptive, explanatory) and the rationale behind the choice (Creswell & Miller, 2000) Methods and analysis: Provide a description of qualitative methods including research approach, research design and data analysis. Capturing and analysing and how you will apply them in order to interpret the data effectively
“Exploratory and descriptive design:Use an exploratory approach to uncover new insights about user interactions, followed by a descriptive analysis to detail the processes and experiences Thematic analysis:Utilize thematic analysis to identify common themes and patterns in user feedback and observational data”
The research is limited to contemporary developments in S&G technology and does not cover earlier or historical S&G technologies, unless it has relevance to the presentation of current issues. (Specifically mentioned)
Theoretical framework
Selection of framework: Select a theoretical framework that will maintain the analysis and interpretation in a lens Interpretive utility: The extent to which the framework can help answer the questions in terms of data collection, analysis and interpreting the data that we have come up with Theoretical and methods fit: Ensuring that the research methods and study design in the work is compatible with established theoretical frameworks study (Yin, 2013)
“Adopt behavioural theories:Use theories of technology adoption and behavioural change to interpret how consumers adapt to and integrate Scan & Go technologies into their shopping habits Framework for technology acceptance:Apply the Technology Acceptance Model (TAM) to evaluate factors influencing the acceptance or rejection of Scan & Go apps” ” (ChatGPT response)
The study emphasises the perceptions and thoughts of consumers with respect to S&G technology. The views of other stakeholders are not considered in the study insofar as it indirectly affects consumer perceptions. (Specifically mentioned) Conceptual boundaries the article introduces an abstract theoretical level known as negaffordance to identify the negative aspects of technology that inhibit, rather than assist in pursuing consumer goals. (Specifically mentioned)
Boundaries
Geographical boundary: As discussed above “Technological focus”: Refers to the type of technologies or systems being examined, not looking at irrelevant tools or platforms unless they are linked to that particular aspect “Sector boundary”: Identifies the exact industry or other area of interest for the study; other sectors are not included, despite whether they have similar characteristics or practices (Yin, 2018) “Temporal bound”: Sets the time frame to be investigated, which usually concentrates on the past or present, but would not include historic background except when necessary (Flyvbjerg, 2006) “User perspective boundary”: What is user/participant specific perspective you would like to collect (Stake, 1995) “Conceptual Boundary”-including the central theoretical or analytic focus of the investigation, focusing on particular dimensions being considered in study “Methodological boundary”: be specific to research method, design, approach, data type and analysis (Yin, 2014)
“Geographical boundary: This study is confined to the United Kingdom, where the retail environment, consumer behaviour, and technological infrastructure are relatively uniform. It focuses on supermarket chains operating within this national context to maintain consistency in cultural and regulatory factors Technological boundary: The research limits itself to a single form of digital retail technology, deliberately excluding other types such as self-checkout kiosks or frictionless stores, unless they are directly integrated with the technology under study. This ensures a focused analysis of user interaction with one specific system Sector boundary: The investigation is restricted to the supermarket retail sector. Other retail domains such as fashion, electronics, or general merchandise are excluded, as their customer engagement dynamics and technological applications differ significantly from the grocery sector Temporal boundary: The timeframe centres on developments and user experiences from 2020 onwards, capturing the impact of technological adoption during and after the COVID-19 pandemic. This period is relevant due to heightened consumer reliance on contactless and digital shopping solutions ” (ChatGPT response) User perspective boundary: The study adopts the viewpoint of everyday consumers, exploring their subjective experiences, frustrations, and behaviours. Perspectives from store staff, developers, or management are not included unless indirectly mentioned by users Conceptual boundary: Conceptually, the study focuses on the negative or problematic aspects of technology use—such as frustration, confusion, or abandonment—rather than positive experiences. This “dark side” lens allows for a deeper understanding of barriers to user satisfaction and adoption (this column is ChatGPT response” (ChatGPT response)
Geographical limitation: Not mentioned in the study Tech border: They are really hammering in-store scan and go technology. (Specifically mentioned) Sector boundaries: The sector is unambiguously retail. (Specifically mentioned) Boundary in time: The analysis covers up-to-date matters in S&G technologies and the horizon of explanations does not drift into the past, unless it is helpful for explaining the present. (Not specifically mentioned) User perspective boundary: The research focuses on consumer experiences and perspectives of S&G technology, not considering any other factors (specifically mentioned) Conceptual boundaries: The paper employs the conceptual frame of ‘negaffordance’ that refers to the negative dimensions of technology that prevent, as opposed to enable, customer outcomes (specifically mentioned) Methodological boundaries: Applies a qualitative phenomenological methodology using thematic analysis, excluding quantitative or other qualitative methods that are not appropriate to investigate this deep, explorative inquiry. (Specifically mentioned) A qualitative phenomenological approach employing thematic content analysis, not quantitative or alternative qualitative traditional ones not pertinent here to such a deep, explorative enquiry. (Specifically mentioned)
Stage 4: Theoretical Underpinnings
Theory is the foundation of qualitative research, and it provides a structure that influences research, from conceptualization, research questions and data collection through to analysis. Theories provide new perspectives or sets of concepts that guide the entire research process, so that data are collected which are not random (Bogdan & Biklen, 1982; Collins & Stockton, 2018). Maxwell (2009, p. 227) raised the importance of “useful theory (which) illuminates what you are seeing in your research” while the seeing of research is also based on the theoretical underpinning of the research (Naeem et al., 2023). There are conflicting views on the use of pre-existing theories that may introduce bias in a study or compromise particular findings (Eisenhardt, 1989; Flinders, 1993). Other scholars argue that they provide a solid position from which to interpret complex events (Baxter & Chua, 2003; Gaffikin, 2007; Jorgensen, 1989).
Theoretical lenses play a more important role in the research process, and also allow the researcher to make inferences from specific observations to more important concepts (Humphrey & Scapens, 1996; Mitchell, 1983; Yin, 2009). Additionally, Christenson (1976, p. 641) indicates “theory is like the bucket, rather than like the water in the bucket. A person has to have a bucket before he can collect the water and he has to have a theory – a framework for measurement – before he can collect “the facts”. Theory also pays a role in relation to the research approach. For example, inductive theorising should be allowed to emerge from the data (Creswell, 1994, 2009). However, Humphrey and Scapens (1996) warn of applying an overly rigid framework, because it may overlook important organisational dynamics. Maxwell (2009, p. 222) indicates that theoretical underpinings “is a formulation of what you think is going on with the phenomena you are studying – a tentative theory of what is happening and why”. Therefore, the theoretical framework is the central lens through which to the view the world, but it does not strictly limit the research to only theoretical factors.
Theoretical underpinning also deepens the researcher’s insight and interpretation of social phenomena (Christenson, 1976). The theoretical underpinning of research provides a space in which to hold and interpret facts (Laughlin et al., 1989; Llewelyn, 2003). The theoretical underpinning also influences the theoretical contribution of the research (Naeem et al., 2023). Moreover, the theoretical underpinning enables depth and structure (Eisenhardt, 1989; Stake, 1995; Yin, 2018) within research to deliver meaningful, organic contributions to wider academic debate (Flyvbjerg, 2006). “Whether or to what extent you pre-structure your study, but in what ways you do this, and why” Maxwell (2008, p. 234). Therefore, researchers need to develop a pre-structured approach through consideration of the theoretical underpinnings of the study, which provide a structured foundation for case study design. Moreover, Maxwell (2009, p. 227) states that a “useful theory illuminates what you are seeing in your research”. Additionally, Naeem et al. (2024) suggest that the use of a specific theory also influences the contribution of research (Naeem et al., 2023), and the theoretical contribution should align with the selected theory in a real context (Naeem et al., 2023, p. 2023).
Column 1, and 2 of Table 4 were produced by synthesizing author viewpoints on theoretical frameworks in the context of case studies. This study borrows from these beliefs about the importance of the theoretical underpinning when creating prompts for ChatGPT. Column 3 summarises the prompts provided to ChatGPT based on these viewpoints. The 4th column charts the response of AI in terms of what it suggests should be considered in the case study. The fifth column lists the theoretical approach Naeem (2025) took in terms of utilising AI assistance. Column three therefore summarises what AI suggests is important on the bases of the prompts it was given.
Theoretical Framework Prompts and Manual Selection of Theoretical Frameworks
Factors to consider for selecting theoretical underpinnings
Rationale ChatGPT prompt
ChatGPT prompt
ChatGPT response
Theoretical factors of affordance theory were used in Naeem (2025) study
Relevance to research objectives
The best theoretical underpinnings provide clear endpoints, ensuring that data are collected, analysed, and interpreted in relation to the chosen theoretical lenses (Maxwell, 2009; Naeem et al., 2023, 2024)
“Based on the detailed considerations provided below regarding the selection of a theoretical framework for a qualitative research study, I am seeking a recommendation for a theory or a combination of theories. Please consider all aspects of my research including the rationale, research gap, and objectives. The details of these elements are as follows • Research rationale:[Insert a brief description of the research rationale here] • Research gap:[Insert a brief description of the identified research gap here] • Research objectives:[Insert a list of research objectives here] Given these details, along with the factors listed in the table, suggest a theoretical framework or a combination of theories that would be most appropriate for my study. The factors to consider are 1. Relevance to research objectives:Theories should guide the study aligned with the research objectives 2. Ability to provide Structure and focus:Theories should organize the study, focus on significant phenomena, and identify patterns 3. Enhancement of analytical rigor:Theories should allow for deep analysis and interpretation of complex social phenomena 4. Support for generalization of findings:Theories should extend the applicability of findings beyond the immediate context 5. Mitigation of research bias:Theories should mitigate biases by providing a systematic approach to data analysis 6. Facilitation of Inductive and deductive approaches:Theories should support both emerging from data and being applied deductively 7. Guidance for Longitudinal and comparative studies:Theories should facilitate tracing developments over time or across different cases Please provide a recommended theoretical framework that aligns with these considerations.” (ChatGPT generated prompts on the bases of column 1&2)
“Technology anxiety theoryhelps explain how negative emotions—such as stress, confusion, or fear—emerge from complex or malfunctioning technology during the customer journey, leading to avoidance or app abandonment. This directly supports your focus on customer dissatisfaction TAM (technology acceptance model)complements this by addressing how perceived usefulness and ease of use influence customer adoption or rejection of S&G apps. It also accommodates the moderating effect of anxiety on perceived ease of use Together, these theories • Align closely with your objectives by explaining both adoption and abandonment • Offer structure for examining experiences across shopping journey stages • Enhance analytical depth by integrating emotion (anxiety) with cognitive evaluations (usefulness, ease) • Support generalization through established models applicable across settings • Allow mixed inductive-deductive reasoning for coding and interpretation This combination ensures both psychological and technological dimensions are covered rigorously” (ChatGPT response)
• This study, based on affordance theoretical lenses, offers a suitable theoretical approach with which to achieve its purpose, which is to investigate the dark side of S&G technology in shopping journeys • Affordance theory is employed to develop a conceptual framework that serves the foundation of research; to identify themes and codes, and analyse negative affordances (‘negaffordances’) of the consumer interaction with the S&G apps • Affordance theory contributes to the analysis of the interplay between the features and characteristics of S&G technology and consumer behaviours and expectations • This study has broader applicability, which is that the findings about “negaffordances” in S&G apps can apply to other digital retail technologies • By grounding the data collection, analysis and interpretation inductively in affordance theory, it is possible to reduce the potential for any subjective interpretation of negative customer experiences • This study uses the affordance theory lens to develop the “negaffordances” concept on the bases of empirical evidence. The study thus advances both the theory and its development into a final conceptual framework • Although not strictly speaking a longitudinal or comparative study, affordance theory may be of use for future research to ensure a robust analytical framework to compare different retail technologies in the future. This takes the study in the direction of “negaffordances”
Ability to provide structure and focus
Selected theories in research help organize the research process and focus the researcher on those phenomena of interest (Collins & Stockton, 2018). They also provide a structured way to look into matters
Enhancement of analytical rigor
Conceptual frames are aids to understand which cases to look at as complex social realities using a structured approach (Stake, 1995). They also help to justify the theoretical contrition (Naeem 2023)
Support for generalization of findings
The credibility and trustworthiness of qualitative studies is often questioned, but with a good theoretical foundation one can argue that the findings have relevance to current theory and context and contribute to the general development of theory (cf Mitchell, 1983; Humphrey & Scapens, 1996; Naeem et al., 2025)
Mitigation of research bias
Furthermore, adopting available theories can also help to reduce potential bias by providing a structured approach to data collection, data analysis, interpretation and conceptualization (Naeem et al., 2023). This will guard against conclusions that are simply subjective in the first place, and position the results around acknowledged theoretical relevance of the research (Eisenhardt, 1989)
Facilitation of inductive and deductive approaches
Based on this evidence-based pathway, in qualitative research, theories can grow in fertile, grounded and authentic ways. Conversely, exploring how previously developed theories can be used in an inductive manner in various settings may provide a means to find relevant argument from the data (Creswell, 1994, 2009). This would also be useful to support inductive or deductive coding at the analytical stage (Naeem et al., 2023)
Guidance for longitudinal and comparative studies
For complex studies, theoretical frameworks provide the opportunity to follow what happened over time or between cases (Eisenhardt, 1989) resulting in a higher subjective theoretical yield of selected case studies
As illustrated in Table 4 (Column 4), AI initially recommended the Technology Acceptance Model (TAM) and Technology Anxiety Theory. However, Naeem (2025) selected Affordance Theory, which was more appropriate than TAM for the context of the study. In response, AI was asked to reframe its theoretical underpinning using Affordance Theory. Consequently, AI proposed several affordance factors (see Table 5) that could serve as theoretical underpinnings. This demonstrates that human involvement is essential in selecting the most suitable theory to guide a case study. It implies the need for a more tailored application of Affordance Theory to fully grasp the demands and complexities of the study (see Table 5). The level of granularity provided by AI illustrates its potential not only to identify relevant elements of Affordance Theory but also to break them down into more specific components that might otherwise be overlooked by human researchers. This synergy between AI-generated insights and human judgment can lead to a richer and more integrated theoretical foundation, enhancing both the analytical depth and practical relevance of the findings. Table 5 outlines the theoretical factors suggested by AI as possible affordance dimensions. Clearly, the theoretical lenses proposed by AI appear more robust and detailed than those initially employed by Naeem (2025), as reflected in Column 5 of Table 4.
AI-Enhanced Theoretical Framework Selection
Affordance theory factors and brief description
Justification for relevance
Technology materiality: The natural capabilities that enable particular action abilities to use technology (Zammuto et al., 2007)
Key for gaining insights into the effects of the design and capabilities of S&G apps on use and experiences
Organizational context: Leveraging of technology in the wider organization environment to achieve organizational objectives (Zammuto et al., 2007)
Assists in determining whether S&G technology is compatible with organizational habits and culture
Reconfiguration of boundaries: Technology has eroded to create boundaries with the use of technology (Barrett et al., 2012)
Useful for investigating how S&G apps revolutionize both shopping experience and customer-retailer relationships
Inferred action potentials: What people perceive is achievable using technology
Important for understanding users’ experience and usage of S&G technology, including in cases where users encounter failures of technology
Practice-based interpretation: Users interpret technology according to their previous experiences (Leonardi, 2011)
Critical to resistance or accommodation of user behaviour to S&G technology in everyday life
Richness of media: Affordances offer multiple services or facilities offered by technology which affect decisions (Jung & Lyytinen, 2013)
Investigates the motives behind why shoppers might choose to use S&G apps over conventional checkout methods, and how S&G apps influence consumers’ shopping behaviour
Sociocultural context: Culture and society in which technology are used (Fayard & Weeks, 2007)
Appropriate to examine how varied demographic groups use S&G technology for adaptability, as well as how it shape them
Challenges: Affordance barriers associated with the technology, for example interface and features (Fayard & Metiu, 2014)
Important for understanding obstacles that cause negative user experiences, for example, privacy issues or usability problems
Presentational modes: How organizations present technology to be congruent with practices to achieve goals (Goffman, 1974)
Important for examining retailers’ promotion of S&G apps and the impact of this on consumer expectations and adoption
Data content: The type and form of data captured by technology shapes the use and output of the user (Essén & Värlander, 2018)
Also useful for investigating privacy and personal data usage issues in S&G apps
User rights: The levels of access and restrictions of use of technology for users (Essén & Värlander, 2018)
Helpful to answer how the control distributed between users and system can influence the adoption and effectiveness of S&G apps
System integration: The extent to which technology seamlessly integrates into the current technology or systems (Essén & Värlander, 2018)
Important for understanding the operational and performance consequences of S&G apps for the retail industry
Collective-action frames: Agreed upon beliefs about technology among users and organizations (Essén & Värlander, 2018)
It is a useful tool to analyse how a collective image about S&G technology can stimulate or prevent its larger take-up
Perceived features: Users’ perceptions of technology features and claimed usefulness (Leonardi, 2011)
Helping to explore the gap between expected and experienced usefulness in S&G apps, thus supporting the negaffordance debate
Practice-based: How current practices or use shape the use of technology apps by users (Orlikowski, 1992)
Crucial to understanding the influence of S&G apps on conventional shopping behaviour and consumer adoption
Material constraints: The physical and design constraints of technology that influence user action (Essén & Värlander, 2018)
Also relevant when considering the impact of a change in attitude (due to negative acts) on any further use of S&G apps
Framing shifts: Alterations in the way users and organizations think about, and justify their usage of technology (Benford & Snow, 2000)
Essential for investigating the way in which end-users and organizations discover and harness the potential offered by S&G technology
Perception: Recognition of technological affordances and potential for action using technology for both individuals and organizations (Greeno, 1994)
Contributes to the measurement of the performance of S&G applications in improving shopping efficiency and customer satisfaction
Affordance realisation: Operationalising and the actual use of technology to achieve goals in specific contexts (Strong et al., 2014)
Critical for measuring the global influence of S&G apps over the retail industry and customer relationships
Affordance effect: Impact of technology used for operational and customer experience (Volkoff & Strong, 2013)
Essential for evaluating the actual use of features and offered features of the apps
Stage 5: Types of Case Study Selection
Writing a case study requires a disciplined and flexible approach to writing aligned with both quantitative and qualitative methods (Merriam, 1998). Yet the most important process is building the narrative as a structured case study type. Moreover, Yin (1994, p. 39) indicates that “the trap of trying to select a ‘representative’ case or set of cases”. Therefore, Stake (1995) suggests outlining a first vignette to create connection and involvement with the narrative situation. “The researcher must consider what type of case study will best answer the research question identified. The selection of a specific type of case study design is guided by the overall study purpose” (Patnaik and Pandey, 2019, p. 166). Therefore, the purpose of the research is key to selecting the appropriate type of case study. Moreover, Yin (2013) provides an explicit structure for case study inquiry that is helpful when it comes to setting out research objectives and directly relating the case to a specific case. This makes the case more meaningful, and focuses on its boundaries, which are identified during the scoping stage.
The selection of an appropriate case study contributes to methodological rigour by linking selected case study types with data collection, analysis and synthesis and the theoretical underpinning of the research (Simons, 2009; Yin, 2018). Researchers can choose from exploratory, explanatory, descriptive, intrinsic, instrumental, or cumulative case studies (Hathaway & Norton, 2018; Stake, 1995; Yin, 2002, 2009), and each supports unique types of research objectives and methods. Yet all involve a detailed context-sensitive examination of the suitability of the case study given the context of the research objectives, and research methodology. Moreover, Yin (2009) classifies case studies into exploratory, explanatory, and descriptive types to address various research questions. Exploratory case studies reveal critical themes, whilst explanatory case studies explain causal relationships. Descriptive case studies provide in-depth contextual information.
Developing a case study involves determining the scope of the research and then choosing the right kind of case study (Yin, 2009). Case studies apply qualitative data collection techniques, such as discussion, interviews and observations to explore phenomena in natural settings, (Eisenhardt & Graebner, 2007; Simons, 2009; Stake, 1995; Yin, 2009). As such, researchers have to work with individuals or organisations to generate a rich and detailed understanding of the selected cases. Cumulative case studies compile knowledge across cases to create generalised observations and analysis (Hathaway & Norton, 2018). Exploratory studies are formative and assist with hypothesis development. Explanatory case studies establish and explain causal links and provide a detailed depiction of data without inferring causality (Jones et al., 2013; Smith, 2022). Therefore, it is necessary to choose an appropriate type of case study at this stage of the development of any case study.
The Table 6 below summarises how AI can help to establish an appropriate type of case study. The third column of Table 6 was generated by integrating views from various authors in terms of how to apply exploratory research in a traditional way. These viewpoints (column 1 and 2 in Table 6) were distilled into prompts for ChatGPT, which was asked to consider a range of factors to decide upon a particular type of case study. The fourth column delineates how and what an exploratory case study should comprise of, as proposed by the AI. The AI was only asked about exploratory case studies because Naeem (2025) used an exploratory case study. The purpose of the analysis below was to compare the results of what AI suggested with Naeem’s (2025) choices. The fifth column represents Naeem’s (2025) considerations without the support of AI. Hence, this table summarises the selection criteria that were used to support an exploratory case study design. It maps each of these process stages, ranging from identifying the problem to replicating and building theory with illustrative examples and suggestions.
Strategic Guidelines to Select Specific Type of Case Study: Exploratory Case Study Selection
Major characteristic/factor of selection of exploratory case study
Rationale ChatGPT prompt
ChatGPT prompt
Guidelines for exploratory case study on scan & go technologies
Case study consideration
1. Problem clarification
Purpose: Clear research problem (Yin, 2018). Using research questions like “why” and “how” to allow depth and flexibility (Baxter & Jack, 2008)
“Given the need to strategically select the type of case study that aligns with the comprehensive goals of an ongoing research project, could you assist in formulating a detailed selection strategy? The strategy should consider all pertinent research details, including rationale, existing gaps, and objectives. Here is the information you need to include 1.Research Rationale: [Insert the purpose and need behind initiating the research project.] 2. Research Gap: [Detail the specific knowledge gaps that the research intends to address.] 3. Research Objectives: [List the concrete objectives the research aims to achieve.] Please develop a selection strategy that integrates the following elements • Problem Clarification: Emphasize understanding complex problems within a broad area lacking structured knowledge. Utilize ‘Why’ and ‘How’ questions to deepen the exploration • Flexibility in Research Design: Recommend maintaining adaptability in the research design, allowing for methodological revisions as insights evolve • Multiple Data Sources: Advise on using a triangulation method to collect data from diverse sources such as interviews, observations, and documents to ensure a comprehensive narrative • Contextual Depth and Richness: Focus on real-world settings to capture in-depth, descriptive data and understand the interactions within the context • Theory Building: Guide the development of new theoretical insights without the constraints of initial hypotheses, letting patterns emerge from the data • Open-Ended and Inductive Inquiry: Encourage inductive reasoning to allow themes and patterns to develop organically from the data • Case Selection Based on Relevance: Propose strategies for selecting cases that are particularly meaningful and rich in information, considering various types of cases like critical, extreme, or typical • Early Stage of Research Process: Discuss the use of exploratory case studies in early research stages to develop further inquiries • Ethical and Reflexive Considerations: Highlight the importance of reflexivity in recognizing biases and maintaining ethical sensitivity throughout the research • Replication and Theory-Building Logic: Explain how to structure the analysis for internal replication and theory alignment Based on these components, can you recommend a structured approach for selecting the type of case study that best suits the described research scenario, ensuring it is methodologically sound and aligned with the overall research framework?” (ChatGPT prompts)
Identify specific challenges: Focus on specific operational, privacy, or user experience challenges faced by S&G technology. Develop research questions that probe why these issues persist and how they impact customer satisfaction and operational efficiency
This study characterised the interrelated issues of S&G apps faced by customers in a retail context
2. Flexibility in research design
Approach: Maintain adaptability through the research design in relation to research questions to bring new insights to the surface (Stake, 1995). Remain open to collect in depth data (Merriam, 1998)
Adapt research based on findings: As new data on S&G app usage and customer feedback emerge, be prepared to adjust the research design to explore these emerging trends or unexpected results in different retail settings
Flexibility was apparent in the use of a phenomenological methodology
3. Use of multiple data sources
Triangulation of multi data (Yin, 2018). Combining different types of data (Tellis, 1997)
Diverse data collection: Gather data from multiple supermarkets using S&G technologies. Include customer surveys, operational data, and employee feedback to understand the varied impacts across different organizational contexts
Use of data from online reviews, interviews, and focus groups increased credibility of the study
In-depth contextual analysis: Conduct site visits and user observations in supermarkets to gain insights into the real-world application and challenges of S&G technologies. Explore how environmental factors like store layout and customer flow influence the technology’s effectiveness
The research was conducted in situ in a real-world digital retail context
5. Theory building
Outcome (contribution): New theoretical insights and a model, framework or concept. (Eisenhardt, 1989). Need to focus on new patterns that emerge from the data (Yin, 2018)
Develop preliminary models: Use the collected data to propose new frameworks or models that explain the interplay between S&G technology use and customer behaviour, potentially leading to enhanced strategies for digital retail innovation
The findings provided new perspectives on negaffordances in e-tail, extending theoretical frameworks
6. Open-ended and inductive inquiry
Methodology: Inductive reasoning (Thomas, 2006). Avoid structured categories, build patterns and themes from the bottom-up (Merriam, 1998)
Inductive analysis approach: Analyse qualitative data gathered from customer interviews and feedback to identify emerging themes around usability, satisfaction, or resistance without preconceived categories
The phenomenological method was a method of induction in which themes were allowed to appear from the data
7. Case selection based on relevance
Sampling: Choose purposefully meaningful and information-rich cases (Patton, 2014). Consider using critical, extreme, or typical cases to uncover deep understanding (Flyvbjerg, 2006)
Select representative samples: Choose a variety of supermarkets—from large chains to smaller independents—that implement S&G technology to examine how different scales of operation affect the technology’s adoption and impact
The case of S&G technology was chosen because it is illustrative of current challenges in retail
8. Early stage of research process
Timing: Use exploratory design of in-depth field of inquiry (Stebbins, 2001). Need to develop the hypothesis for future research (Yin, 2018)
Preliminary exploration: Given the evolving nature of retail technology, use this exploratory study to lay the groundwork for more detailed, hypothesis-driven research on the efficacy and consumer reception of S&G apps
The research served [to inform] future research questions related to technology’s effect on retail
9. Ethical and reflexive considerations
Reflexivity: Researchers need to be aware of biases and subjectivity of the research (Stake, 1995) Ethical sensitivity: Ethical practices should be practices (Creswell, 2013)
Maintain ethical rigor: Ensure all participant data from user interactions with S&G technology is collected with informed consent and remain critically aware of bias in analysing data from different customer demographics
Reflexivity and ethical issues were not explicitly discussed, most specifically reflection on the role and bias of the researcher at all stages
11. Replication and theory-building logic
Replication strategy: Should be well structured to compare sub-units or temporal episodes for consistency (Yin, 2013). Theory alignment: Illuminate theoretical constructs (Eisenhardt & Graebner, 2007)
Comparative analysis: Use data from different types of supermarkets to compare the implementation and effects of S&G technology, aiding in theory development about technological adoption and consumer interaction in retail settings (this column is ChatGPT response)
The internal structure of the analysis was built by comparing patterns of customer interaction through different steps in the S&G towards applications
Naeem (2025) implemented the characteristics of exploratory case study research, and addressed important challenges associated with S&G technology. These are listed in column 6 of Table 6. Column 5 presents the factors that AI suggested should be considered in designing an exploratory case study. These include the use of diverse data sources and nuanced contextual analysis to produce a comprehensive understanding of the impact of technology in different settings. Additionally, the AI called for the formulation of new theory on the basis of rich data relating to student experiences. In this sense, it encouraged an inductive research approach using an open-ended form of inquiry.
Stage 6: Case Selection
In terms of the development of theory from case studies, one of the most important steps is to #develop something with excellent theoretical qualities. It is important to ensure that an appropriate number of cases are selected. “Researchers can adopt either a single-case or multiple-case design depending on the issue in question. In cases where there are no other cases available for replication, the researcher can adopt the single-case design” (Zainal, 2007, p. 2). So, case selection relies on rationalising sampling, research objectives, and research questions to shed light on the theoretical and practical contribution of the research (Eisenhardt & Graebner, 2007).
The type of case study (single or multiple cases) is also central to the research. Descriptive, exploratory, explanatory, intrinsic, instrumental or collective case studies define the methodological approach and the type of analysis undertaken (Stake, 1995; Yin, 2009). For example, intrinsic case studies focus on cases so the researcher needs to select more than one case. In contrast, collective case studies exist to produce insights from multiple cases to generalize theory (Stake, 1995). Such an approach is key to selecting more than one case. Therefore, the selection of cases, whether single, multiple, or collective, shapes the depth and transferability of the conclusions of the research. Case studies examine phenomena in some depth to generate thick description, and this is a particularly useful approach in rare or extreme conditions (Stake, 1995; Yin, 2009). Following replication logic, case studies can be replicated and compared in several contexts, which increases the strength of the findings (Eisenhardt & Graebner, 2007; Yin, 2013). Cross-case analysis integrates findings across multiple cases to construct a picture of the phenomenon in its entirety (Stake, 1995). Yet the purpose of the research should be identified to justify the number of cases selected and whether one or more case studies is required. In the case of exploratory case studies, one case can be selected, but this depends on whether enough data can be collected from a single case study.
Another debateable element here is that research can also take different organizations as a single case study on the basis of the nature of the problem, because if the problem is collective in nature, then different organizations can be considered as a single case study. “However, the drawback of a single-case design is its inability to provide a generalising conclusion, in particular when the events are rare. One way of overcoming this is by triangulating the study with other methods in order to confirm the validity of the process. The multiple-case design, on the other hand, can be adopted with real-life events that show numerous sources of evidence through replication rather than sampling logic” (Zainal, 2007, p. 2). Yet when a researcher treats a multi organisation set up as a single case study, they must isolate these organizations from the same industry. If the research setting involves multiple organizations form the same sector, a multi case study should be considered. The rationale for selecting multiple organisations as a single case study is that where the problem is the same for each organisation, this helps the researcher to explore the case study in depth. Furthermore, the researcher can take multiple organisation form the same industry as a multi case study to explore any differences between the organisation and the research context. If the research setting involves multiple organisations from the same sector but aims to compare them, then a multiple case study approach should be considered instead.
The rationale for selecting multiple organisations as a single case lies in the uniformity of the issue under investigation—if the same problem exists across these organisations, it enables the researcher to explore that issue in greater depth. However, “in cases where there are no other cases available for replication, the researcher can adopt the single-case design” (Zainal, 2007, p. 2). Conversely, if the aim is to identify differences in how similar organisations respond to the same problem, then a multiple case study is more appropriate. This approach can reveal variations in challenges, responses, and contextual dynamics. Similarly, when the researcher selects organisations from different industries, the objective may still be to explore how a common issue is experienced or addressed differently due to varying sectoral contexts, service structures, or organisational models. In such cases, a multiple case study design is essential to justify cross-industry comparisons. Even if organisations belong to the same industry—for example, public versus private providers—a multiple case study may be required if the goal is to explore contrasts. However, if the organisations are from the same sector and provide similar services, the researcher has the option to treat them as a single case study. In this scenario, clear justification must be provided for why the single case study approach is appropriate and how it aligns with the research objectives.
Columns 1 and 2 of Table 7 below summarise the opinions of various authors regarding the selection of a single case study. This pool of views was then used to generate prompts for ChatGPT to suggest guiding principles (see Column 3). Column 4 outlines the AI-generated tasks for the research (the case study) and the aspirational goals or ‘dream’, while Column 5 presents the components considered by Naeem (2025). The AI reflected on the Scan and Go (S&G) technology case study by proposing a comprehensive research scope. It offered suggestions related to data type selection, theoretical orientation, and methodological approach, and provided a rationale for using a single case study—similar to the approach taken by Naeem (2025). Naeem’s decision to adopt a single case study indicates its utility in addressing a specific problem within a real-life retail context. Nevertheless, the AI-generated suggestions provide a broad and robust methodological framework for studying that context through a single case study design.
Strategic Framework for Exploratory Case Study
Focus/factor of single case study
Rationale ChatGPT prompt
ChatGPT prompt
Guidelines for S&G app case study research
Explanation and study details
1. Research purpose and questions
Nature of inquiry Need to focus on “how” and “why” types of questions (Yin, 2018)
“Considering the complexity and specific needs of my current research project, I need to select the most appropriate case for an in-depth case study analysis. The selection must meticulously align with the research’s purpose, questions, and theoretical underpinnings. Here’s the essential information you’ll need to formulate your recommendations 1.Research Rationale: [Provide a detailed explanation of why the research is necessary and what it seeks to address in the broader field.] 2. Research Gap: [Describe the specific areas of knowledge that remain unexplored within the current research landscape.] 3. Research Objectives: [List the goals the research aims to achieve through the investigation.] Based on this information, please develop a comprehensive strategy for selecting a case study that includes • Nature of Inquiry: Recommendations on how to frame ‘how’ and ‘why’ questions that probe deeply into the phenomenon of interest • Case Selection Logic: Guidelines for choosing a case based on its potential to refine, replicate, or extend theoretical constructs. Suggestions for identifying unique or critical cases that provide valuable insights • Data Collection Strategies: Advice on using multiple sources and triangulating data to enhance the validity and depth of the findings • Contextual Understanding: How to ensure that the study acknowledges and integrates the impact of the local context on the case dynamics • Exploratory Nature: Strategies to utilize the case study’s exploratory potential to uncover emerging themes and relationships in early-stage research • Theoretical Contribution: How to approach theory generation and provide deep conceptual insights without the intent of testing predefined hypotheses • Research Design Flexibility: Recommendations on maintaining adaptability in research design to accommodate new insights and emergent themes • Ethical and Reflexive Considerations: How to maintain reflexivity and ethical sensitivity throughout the research process • Replication and Theory-Building Logic: Approaches to structure the analysis for effective internal replication and theoretical alignment • Multiple Organizations as a Single Case: If applicable, strategies for treating multiple organizations as a single case to explore cross-organizational patterns under a unified theme Can you provide a structured approach that outlines the case selection process, ensuring it aligns with the stated research rationale, gaps, and objectives, and enhances the overall impact and relevance of the study?” (ChatGPT prompt)
“Define research scope: Focus on how S&G technology impacts consumer behaviour and retail operations in specific contexts such as supermarkets “Formulate key questions: How do S&G apps influence shopping efficiency and customer satisfaction? Why do certain demographic groups prefer or avoid using S&G apps?
The rationale for selecting a single case study is that the research problem is common and universal across many organizations in an industry. When services that are transferred by these organizations are similar, and encounter the same problems, they can be approximated as a single unit of analysis. With this in mind, a detailed and focused analysis of the problem can be made in a single consistent setting. The grounds for this are the homogeneity of the organisations and the stability of the type of technological experience under study in such a naturalistic setting as retail. Yet there has been little discussion about the use of ‘how’ and ‘why’ research questions, and the theoretical background must align with the selection of case study
2. Case selection logic
Theoretical sampling: Selection based on theoretical constructs (Eisenhardt & Graebner, 2007) Unique or critical cases: Single cases may represent extreme and rich data (Yin, 2009)
“Select pertinent cases: Choose retail environments that have shown significant adoption or unique implementations of S&G technology. Look for unique cases where S&G implementation has led to notable changes in shopping behaviour or operational efficiency
4. Data collection strategies
Multiple data sources need to be considered (Tellis, 1997; Yin, 2018) Data triangulation: Integrates sources to improve depth of understanding on the base of data (Yin, 2018)
Employ multiple data sources: Gather data through customer surveys, employee interviews, and analysis of transaction logs. Use video observations to understand interactions with S&G technology
“Contextual analysis: Examine how the layout of stores, regional consumer preferences, and local economic conditions influence the use and effectiveness of S&G apps
6. Exploratory nature
Initial understanding: Useful to explore new solution of new problem (Stebbins, 2001)
“Explore emerging patterns: Identify patterns in consumer acceptance and resistance to S&G technology. Assess the evolving trends and potential operational challenges
7. Theoretical contribution
Theory generation: Aims to build initial frameworks or propositions to guide future research (Eisenhardt, 1989) Conceptual insights: Provides deep conceptual understanding rather than testing existing models (Yin, 2018)
“Develop theoretical insights: Construct a framework that explains the interaction between consumer behaviour and S&G technology adoption. Explore how this impacts retail strategy and customer experience
“Adapt research design: Be flexible in adapting the research approach as new insights emerge from initial findings. Modify research questions or data collection methods as needed
11. Replication and theory-building logic
Replication strategy: Even within a single case, analysis may be more focused on collective inquiry (Yin, 2013) Theory alignment: Select the case that can best align with the theory (Eisenhardt & Graebner, 2007)
“Facilitate theory building: Use findings from the S&G case study to propose or refine theories related to technology adoption and consumer behaviour. Consider temporal comparisons to understand trends over time
Multiple organizations as a single case
In studies that use multiple organizations to form one case, the investigation serves as the theoretical focus in which to compare and contrast cross-organizational patterns and distinctions (Yin, 2018). The focus should be on how different companies respond to similar problems or seize opportunities in the market.
“Select multiple supermarket apps as cases:Choose Scan & Go apps from various supermarkets such as Tesco, Sainsbury’s, and Asda to analyse how each organization implements this technology. Focus on how these different implementations influence customer experiences, operational efficiencies, and business outcomes. This selection enables a comprehensive understanding of the sector-wide impacts of Scan & Go technologies, allowing for a detailed comparison and richer insights into sectoral trends and practices” (ChatGPT response)
Stage 7: Unit of Analysis
In case study research, the unit of analysis is the most basic and debateable element which determines the core of the case; the ‘what’ or ‘whom’ in question (Yin, 2009). “The unit of analysis identifies what constitutes a “case,” that is, what is a case all about” (Patnaik and Pandey, 2019, p. 167) This idea helps define the focus of the study at the analysis stage, and the findings can then be related to research question and theoretical frames by situating the case within the theoretical contribution of the research. The unit of analysis is important for comparability and replication across cases, and for enabling the generation of more substantive cross-case insights (Rashid et al., 2019). A well-specified unit of analysis also increases the degree of congruence with theoretical boundaries and with the research scope, the context of the research, the philosophical underpinning, the type of case study, and the type of case study selected. The selection of a specific industry (Darke et al., 1998) is also key, and influenced the use of a specific theory. The unit of analysis is a key feature of case studies and refers to the ‘thing’ that is the case study. For the purposes of replication and cross-case comparison, this ensures some adherence to theoretical constraints, keeping the study consistent (Darke et al., 1998; Yin, 2009).
The selection of the unit of analysis is also based on the type of case that is chosen. For example, a cumulative case study is a research process whereby multiple individual cases are studied in order to identify patterns. It is valuable especially when researchers want to bring together complex and diverse data from different sites into an integrated interpretation of a phenomenon (Baxter & Jack, 2008). A cumulative case study allows for generalisation and increases the strength of, and confidence in the results across cases (Stake, 2013). Yet, to make this possible, the method needs to properly combine different types of data, which can be costly both in terms of resources and complexity. Such studies are of great importance in areas such as education and psychology, where different contextual factors can influence the results (Yin, 2013).
The level of analysis varies across different types of case studies depending on the research question. In terms of exploratory phenomena, the researcher must identify emerging patterns, or predict links between some factors. Explanatory analysis restricts the researcher to finding the cause-effect relationship between the units/variables (Hancock & Algozzine, 2016). Descriptive case studies involve starting with a descriptive theory, which they then test through a description of the phenomenon in relation to selected theory (Merriam, 1998). Intrinsic case studies consider the case to be the study’s object, selected from the point of view of its uniqueness , whereas instrumental studies (Yazan, 2015) use the case as a vehicle to examine a wider theoretical or academic problem which also relies on a theoretical underpinning (Mills et al., 2009; Swanborn, 2010). Cumulative case studies take multiple comparable cases as the unit of comparison which can lead to the development of a broader appreciation in the area of study. These types vary in terms of the degree to which the unit is circumscribed (Patnaik and Pandey, 2019; Swanborn, 2010).
Naeem’s (2025) research is based on a single exploratory case. Therefore, column 1 and column 2 synthesized the major units of analysis that need to be considered in the case of exploratory case study research. Various perspectives as to the selection of a unit of analysis for a case study have been consolidated to generate prompts for ChatGPT. The third column contains these prompts, which are based on authorial viewpoints which are summarised in column 2. The fourth column contains AI-propagated recommendations for the inclusion of particular units of analysis with consideration given to the research aim and objectives of Naeem (2025). The fifth column describes which units of analysis were considered by Naeem (2025) , without the use of AI intervention.
Table 8 shows that AI suggested very rich and wide ranging units of analysis, including user interactions, team dynamics, retail setting, and decision-making analysis. In contrast, Naeem (2025) study was limited to the negative impacts on consumer behaviour, and a thematic analysis of consumer experience. This establishes that, while AI gives us a complete structure to analyse a wide range of units of analysis, Naeem (2025) were more focused on the finer details, but included an additional unit of analysis suggested by AI to provide a deeper interpretation of subset domains.
Guidelines for Selecting Units of Analysis in Single Exploratory Case Study Research
Unit of analysis
Rationale ChatGPT prompt
ChatGPT prompt
Guidelines for S&G app case study research
Consideration in study
Individual
Focuses on the experiences, and perceptions, individual (Yin, 2018)
“Given the depth and specificity of my current research project, selecting the correct unit of analysis is crucial for ensuring the integrity and applicability of the findings. To help identify the most suitable unit of analysis, here is the necessary information that will guide your recommendations 1.Research Rationale: [Detail why this research is essential, what broader issues it addresses, and the unique aspects it explores.] 2. Research Gap: [Explain the specific knowledge gaps that this research aims to fill, highlighting areas that have been underexplored or misunderstood in previous studies.] 3. Research Objectives: [List the precise goals that the research intends to achieve, specifying what the study aims to uncover or determine.] Based on the above, develop a strategy for selecting the unit of analysis that includes • Individual: How focusing on single individuals might reveal in-depth insights into personal experiences or behaviours relevant to the research objectives • Group: The potential benefits of examining group dynamics or decision-making processes within the context defined by the research rationale • Decision Processes: Whether studying cognitive and affective processes leading to decisions could elucidate the complexities identified in the research gap • Organizational Routines: The relevance of analysing entire organizations as units to understand broader systemic functions or impacts • Social Interactions: How interactions within the case study setting could illuminate the social dynamics under investigation • Event: The significance of focusing on particular events and their relevance to the overarching research questions • Technological Artifacts: The role and impact of technology within the study’s scope and its alignment with the technological dimensions of the research gap • Policy and Regulatory Frameworks: Considering the influence of policy on the case study, especially if regulatory impacts are a central aspect of the research gap • Cultural Context: How cultural norms and values within the study context might affect behaviours and outcomes relevant to the research objectives • Economic Factors: The importance of economic conditions in shaping the phenomena under study • Process or Activity: The benefits of tracing specific processes or activities to understand the operational dynamics highlighted by the research • Program/Interventions: The potential insights from examining structured programs or interventions as they relate to the study’s aims • Community or Setting: The impact of community-specific factors and how they influence the phenomenon being studied • Embedded Units (Sub-units): The value of exploring sub-cases within a main case to provide a more detailed and comparative analysis Can you provide a detailed recommendation on which unit of analysis would best suit my research, taking into consideration the stated rationale, gaps, and objectives to maximize the relevance and impact of the findings?” (ChatGPT prompt)
“Explore user interactions: Detail individual user interactions with S&G apps to understand how features like speed, usability, and personalization affect their shopping experience and behaviour
Not explicitly mentioned
Group
Rationale of decision-making, or interaction patterns within a team (Creswell, 2013)
Analyse team dynamics in retail settings: Investigate how groups (e.g., families or friends shopping together) use S&G apps differently from individuals. Focus on collaborative decision-making and shared app interactions
Focus groups discussed; group dynamics not specified
Decision processes
Cognitive and affective processes leading to decisions (Stake, 1995)
Decision-making analysis: Study how decisions to use or avoid S&G technology are made, including factors like trust, perceived efficiency, and previous experiences with technology
Implicitly considered through thematic analysis of consumer experiences
Organizational routines
Treats the whole organization (Yin, 2009), e.g., a supermarket as a single, and bounded case in context of selected problem
Organizational impact assessment: Evaluate how S&G technology integration affects organizational routines and operations, particularly in inventory management and customer service
Not mentioned
Social interactions
Context of interaction of individuals or groups (Simons, 2009)
Social dynamics evaluation: Observe and analyse how social influences (peer pressure, social proof) affect the adoption and use of S&G apps within retail environments
Considered through focus groups and interviews discussing consumer interactions with technology
Event
Focuses on a significant or unique events links together (Stake, 1995)
Event-based study: Examine a rollout of new S&G technology or policy changes affecting its use, focusing on the immediate reactions and adaptations by consumers and staff
Technology effectiveness analysis: Critically assess the functionality and user interface of S&G apps, investigating issues like app crashes or errors and their impact on the shopping experience
Central to study; scan and go technologies analysed in-depth
Cultural analysis: Examine how different cultural backgrounds affect perceptions and usage of S&G apps, particularly in terms of privacy concerns and technology trust
Intervention effectiveness analysis: Study the effectiveness of specific interventions, such as user training on app usage or promotional campaigns to boost S&G adoption
Not mentioned
Community or setting
Focus on community , beliefs, and cultural norms (Stake, 1995). How community-specific factors affect the phenomenon (Creswell, 2013)
Community impact analysis: Investigate how different communities (urban vs. rural, high-income vs. low-income) adopt and adapt to S&G technology. Analyse the cultural attitudes towards technology, privacy concerns, and the overall shopping habits that influence the use of S&G apps in varying community settings
Not mentioned but hints at community through focus on consumer experiences in retail settings
Embedded units (sub-units)
Consider subunits like sub departments, user types, or time phases in a single case (Yin, 2018) Enables richer analysis of internal comparison of similarity and difference of cases (Yin, 2018)
Comparative sub-unit evaluation: Within the larger case of a retail chain using S&G technology, compare different departments (e.g., groceries vs. electronics) or different types of user groups (e.g., tech-savvy millennials vs. older, less tech-oriented shoppers) to assess how S&G app usage varies. Evaluate the effectiveness of S&G technology during different times (e.g., holiday seasons vs. regular days) to identify patterns and effects on consumer behaviour and store operations” (This column is ChatGPT response)
Not explicitly mentioned but more focused on stages of the customer journey analysed
Conclusion
This paper outlines the case study development process guided using AI. This has resulted in seven distinct stages, each informed by a range of academic approaches. The paper describe a circular structure whereby GenAI is used as a tool to assist researchers in the development of case studies. The generated prompts for each phase are based on analyses of literature. The contribution of this work is to empower future researchers to develop prompts for other purposes and other case studies. The seven stages developed in this paper are key inputs that guide researchers as to the development of qualitative case studies. The framework is innovative in providing a systematic step-by-step procedure for the construction of a case study to develop rigorous research. Equipped with this robust case-study framework, researchers can go further with qualitative analysis with more accuracy by linking the analysis back to these stages. This comparison shows that, in every phase of a case study, AI can provide richer and more fundamental assistance with respect to traditional approaches. This research is limited to the development of a case study to carry out research. Future researchers can perhaps improve this methodology to specifically consider:
(1) How the seven steps can contribute to the rigour, truthfulness, trenchancy, and credibility of research?
(2) How might the framework might contribute to development of a conceptual framework, model, and theory?
(3) How might the seven steps might contribute to the rigour of thematic analysis, and what elements can be considered to embed case study development as a foundation to carry out systematic thematic analysis as described by Naeem et al. (2023)
(4) How might units of analysis before and after data collection help to develop themes and codes during AI assisted systematic thematic analysis, similar to the process introduced by Naeem et al. (2025)?
Acknowledgement of the Use of Gen AI
As this paper looks at the use of Gen AI development in case study development, Gen AI was used to complete several experiments to develop AI prompts and and structured tables in this paper. Therefore, all prompts were heavily edited and restructured by AI to make them more ‘Gen AI friendly’. Additionally, each table also contains ChatGPT responses which are replaced exactly as they were generated to effect comparisons with Naeem’s (2025) case study. Additionally, in order to compare ChatGPT’s output at each stage of the case study development process with the findings of Naeem (2025), that particular article was shared with the AI to identify any elements that might be overlooked in a manual analysis. This highlights how researchers might make use of ChatGPT to cross-check their own interpretations and to ensure that nothing critical is missed. Additionally, several articles were shared with ChatGPT to establish if any important element of the case study was missing from the seven stages introduced in the paper. As such, the paper was frequently copied and pasted to affect a comparison with existing literature and other qualitative case study papers. The language of the paper was enhanced using generative AI tools, which were also employed to restructure sentences for improved clarity and coherence. This was followed up with manual reading and editing to ensure accuracy and consistency. Finally, the paper was professionally proofread, and a proofreading certificate was submitted to the editor during the revision stage.
BaxterJ.ChuaW. F. (2003). Alternative management accounting research: Whence and whither. Accounting, Organizations and Society, 28(2–3), 97–126. https://doi.org/10.1016/s0361-3682(02)00022-3
4.
BaxterP.JackS. (2008). Qualitative case study methodology: Study design and implementation for novice researchers. Qualitative Report, 13(4), 544–559. https://doi.org/10.46743/2160-3715/2008.1573
5.
BenfordR. D.SnowD. A. (2000). Framing processes and social movements: An overview and assessment. Annual Review of Sociology, 26(1), 611–639. https://doi.org/10.1146/annurev.soc.26.1.611
6.
BergerP. L.LuckmannT. (1966). The social construction of reality: A treatise in the sociology of knowledge. Anchor Books.
7.
BogdanR.BiklenS. (1982). Qualitative research for education: An introduction to theory and methods. Allyn & Bacon.
CollinsC. S.StocktonC. M. (2018). The central role of theory in qualitative research. International Journal of Qualitative Methods, 17(1), 1–10. https://doi.org/10.1177/1609406918797475
CrottyM. J. (1998). The foundations of social research: Meaning and perspective in the research process. Sage.
19.
CroweS.CresswellK.RobertsonA.AveryA.SheikhA. (2011). The case study approach. BMC Medical Research Methodology, 11(1), 100. https://doi.org/10.1186/1471-2288-11-100
20.
DarkeP.ShanksG.BroadbentM. (1998). Successfully completing case study research: Combining rigor, relevance and pragmatism. Information Systems Journal, 8(4), 273–289. https://doi.org/10.1046/j.1365-2575.1998.00040.x
21.
DavisN. (2017). Digital technologies and change in education: The arena framework. Routledge.
22.
DenzinN. K.LincolnY. S. (1998). Collecting and interpreting qualitative materials. Sage.
23.
DubéL.ParéG. (2003). Rigor in information systems positivist case research: Current practices, trends, and recommendations. MIS Quarterly, 27(4), 597–636. https://doi.org/10.2307/30036550
24.
EisenhardtK. M. (1989). Building theories from case study research. Academy of Management Review, 14(4), 532–550. https://doi.org/10.2307/258557
25.
EisenhardtK. M.GraebnerM. E. (2007). Theory building from cases: Opportunities and challenges. Academy of Management Journal, 50(1), 25–32. https://doi.org/10.5465/amj.2007.24160888
26.
EssénA.VärlanderS. W. (2018). How materiality enables and constrains framing practices: Affordances of a rheumatology e-service. Journal of Management Inquiry, 28(4), 459–472. https://doi.org/10.1177/1056492618760722
27.
FayardA. L.MetiuA. (2014). The role of writing in distributed collaboration. Organization Science, 25(5), 1391–1413. https://doi.org/10.1287/orsc.2013.0893
28.
FayardA. L.WeeksJ. (2007). Photocopiers and water-coolers: The affordances of informal interaction. Organization Studies, 28(5), 605–634. https://doi.org/10.1177/0170840606068310
29.
FlindersD. J. (1993). From theory and concepts to educational connoisseurship. In FlindersD. J.MillsG. E. (Eds.), Theory and concepts in qualitative research: Perspectives from the field (pp. 117–140). University of Chicago Press.
GaffikinM. (2007). Accounting research and theory: The age of neo-empiricism. Australasian Accounting, Business and Finance Journal, 1(1), 1–19. https://doi.org/10.14453/aabfj.v1i1.1
32.
GarciaL.A.NúñezA.-M.SansoneV. A. (2019). Toward a Multidimensional Conceptual Framework for Understanding “Servingness” in Hispanic-Serving Institutions: A Synthesis of the Research. Review of Educational Research, 28(4), 745–784. https://doi.org/10.1177/10564926187607
GubaE. G.LincolnY. S. (1985). Naturalistic inquiry. Sage Publications.
37.
GubaE. G.LincolnY. S. (1994). Competing paradigms in qualitative research. In DenzinN. K.LincolnY. S. (Eds.), Handbook of qualitative research (pp. 105–117). Sage.
38.
GubaE. G.LincolnY. S. (2005). Paradigmatic controversies, contradictions, and emerging confluences. In DenzinN. K.LincolnY. S. (Eds.), The Sage handbook of qualitative research (pp. 195–196). Sage.
39.
HancockD. R.AlgozzineB. (2016). Doing case study research: A practical guide for beginning researchers. Teachers College Press.
40.
HarrisonH.BirksM.FranklinR.MillsJ. (2017). Case study research: Foundations and methodological orientations. Forum for Qualitative Social Research, 18(1), 19. https://doi.org/10.17169/fqs-18.1.2655
41.
HathawayD.NortonP. (2018). Understanding problems of practice: A case study in design research. Springer International Publishing.
42.
HumphreyC.LeeB. (Eds.), (2004). The real life guide to accounting research: A behind-the-scenes view of using qualitative research methods. Elsevier.
43.
HumphreyC.ScapensR. W. (1996). Methodological themes: Theories and case studies of organizational accounting practices: Limitation or liberation?Accounting, Auditing & Accountability Journal, 9(4), 86–106. https://doi.org/10.1108/09513579610129435
JarvensivuT.TornroosJ. A. (2010). Case study research with moderate constructionism: Conceptualization and practical illustration. Industrial Marketing Management, 39(1), 100–108. https://doi.org/10.1016/j.indmarman.2008.05.005
46.
JonesA.TorresV.ArminioJ. (2013). Negotiating the complexities of qualitative research in higher education: Fundamental elements and issues. Routledge.
47.
JorgensenD. L. (1989). Participant observation: A methodology for human studies. Sage Publications.
48.
JungY.LyytinenK. (2013). Towards an ecological account of media choice: A case study on pluralistic reasoning while choosing email. Information Systems Journal, 24(3), 271–293. https://doi.org/10.1111/isj.12024
49.
KaplanB.MaxwellJ. A. (1994). Qualitative research methods for evaluating computer information systems. In AndersonJ. G.AydinC. E. (Eds.), Evaluating health care information systems: Methods and applications (pp. 45–68). Sage.
LeonardiP. M. (2011). When flexible routines meet flexible technologies: Affordance, constraint, and the imbrication of human and material agencies. MIS Quarterly, 35(1), 147–168. https://doi.org/10.2307/23043493
52.
LlewelynS. (2003). Methodological issues: What counts as “theory” in qualitative management and accounting research? Introducing five levels of theorizing. Accounting, Auditing & Accountability Journal, 16(4), 662–708. https://doi.org/10.1108/09513570310492344
53.
MaxwellJ. A. (2008). Designing a qualitative study. In BickmanL.RogD. J. (Eds.), The handbook of applied social research methods (2nd ed., pp. 214–253). Sage Publications.
54.
MaxwellJ. A. (2009). Evidence: A critical realist perspective for qualitative research. In DenzinN. K.GiardinaM. D. (Eds.), Qualitative inquiry and social justice (pp. 108–122). Left Coast Press.
55.
MerriamS. B. (1998). Qualitative research and case study applications in education. Jossey-Bass.
56.
MeyersM. (2004). Qualitative research and the generalizability question: Standing firm with Proteus. Qualitative Report, 9(3), 494–500. https://doi.org/10.46743/2160-3715/2000.2925
57.
MillsA. J.DureposG.WiebeE. (Eds.). (2009). Encyclopedia of case study research. Sage Publications.
MorehouseR. E.MaykutP. (2002). Beginning qualitative research: A philosophical and practical guide. Routledge.
60.
NaeemM. (2025). Emerging trends in global e-retailing: Exploring the dark side of scan and go in-store technologies in consumer shopping journeys. International Marketing Review, 42(2-3), 249–282. https://doi.org/10.1108/imr-06-2023-0110
61.
NaeemM.OzuemW.HowellK.RanfagniS. (2023). A step-by-step process of thematic analysis to develop a conceptual model in qualitative research. International Journal of Qualitative Methods, 22, 16094069231205789. (Original work published 2023). https://doi.org/10.1177/16094069231205789
62.
NaeemM.OzuemW.HowellK.RanfagniS. (2024). Demystification and actualisation of data saturation in qualitative research through thematic analysis. International Journal of Qualitative Methods, 23, 16094069241229777. (Original work published 2024). https://doi.org/10.1177/16094069241229777
63.
NaeemM.SmithT.ThomasL. (2025). Thematic analysis and artificial intelligence: A step-by-step process for using ChatGPT in thematic analysis. International Journal of Qualitative Methods, 24, 16094069251333886. (Original work published 2025). https://doi.org/10.1177/16094069251333886
64.
O’NeilP. (2000). Handbook of community psychology. Springer.
65.
OrlikowskiW. J.BaroudiJ. J. (1991). Studying information technology in organizations: Research approaches and assumptions. Information Systems Research, 2(1), 1–28. https://doi.org/10.1287/isre.2.1.1
66.
PatnaikS.PandeyS. C. (2019). Case study research. In SubudhiR. N.MishraS. (Eds.), Case study research. In Methodological issues in management research: Advances, challenges, and the way ahead (pp. 163–179). Emerald Publishing Limited.
67.
PattonM. Q. (2014). Qualitative research & evaluation methods: Integrating theory and practice. Sage Publications.
68.
PetersonM. (2012). EFL learner collaborative interaction in Second Life. ReCALL.
69.
RashidY.RashidA.WarraichM. A.SabirS. S.WaseemA. (2019). Case study method: A step-by-step guide for business researchers. International Journal of Qualitative Methods, 18, 1–13. https://doi.org/10.1177/1609406919862424
70.
Sanders. (2010). Behavioural analysis. Theory and methods in political scienc. paigrave.
71.
SchochK. (2020). Case study research. In LeavyP. (Ed.), Case study research. Research design and methods: An applied guide for the scholar-practitioner (pp. 245–264). Sage Publications.
72.
SimonsH. (2009). Case study research in practice. Sage Publications.
73.
SmithM. (2022). Research Methods in Accounting. Sage Publications.
74.
StakeR. E. (1995). The art of case study research. Sage Publications.
75.
StakeR. E. (2000). Case studies. In DenzinN. K.LincolnY. S. (Eds.), Handbook of qualitative research (pp. 435–454). Sage.
76.
StakeR. E. (2013). Multiple case study analysis.Guilford press.
77.
StebbinsR. A. (2001). Exploratory research in the social sciences. Sage Publications.
78.
StrongD. M.JohnsonS. A.TuluB.TrudelJ.VolkoffO.PelletierL. R.Bar-OnI.GarberL. (2014). A theory of organization-EHR affordance actualization. Journal of the Association for Information Systems, 15(2), 53–85. https://doi.org/10.17705/1jais.00353
79.
SwanbornP. (2010). Case study research:Case study research: What, why and how?.Sage.
ThomasD. R. (2006). A general inductive approach for analyzing qualitative evaluation data. American Journal of Evaluation, 27(2), 237–246. https://doi.org/10.1177/1098214005283748
82.
Thompson, C. J., Locander, W. B., and Pollio, H. R. (1989). Putting consumer experience back into consumer research: The philosophy and method of existential-phenomenology. Journal of Consumer Research, 16(2), 133–146. https://doi.org/10.1086/209203
83.
VilliersC. J.DumayJ.MarounW. (2019). Qualitative accounting research: dispelling myths and developing a new research agenda.Accounting & Finance, 23(3), 1459–1487. https://doi.org/10.1111/acfi.12487
84.
VolkoffO.StrongD. M. (2013). Critical realism and affordances: Theorizing IT–associated organizational change processes. MIS Quarterly, 37(3), 819–834. https://doi.org/10.25300/misq/2013/37.3.07
85.
WilsonJ. (2010). Essentials of business research: A guide to doing your research project. Sage.
ZammutoR. F.GriffithT. L.MajchrzakA.DoughertyD. J.FarajS. (2007). Information technology and the changing fabric of organization. Organization Science, 18(5), 749–762. https://doi.org/10.1287/orsc.1070.0307
