Interpretive structural modelling

for factors leading to consumer privacy trade

off in online shopping
Sanyam Jain, Dipanjan Dasgupta, Ananya Kulkarni, Parth Pandya, Aarya Mandrekar
ASMSOC, NMIMS, Mumbai

I.ABSTRACT
Purpose – In the digital marketplace, the trade-off between consumer privacy and the benefits of data sharing has become a
critical issue. While consumers value personalized experiences, they also have apprehensions about how their personal data is
collected and used. This study aims to examine the key factors that influence consumers' willingness to share private
information with digital retailers and establish a hierarchical relationship among these factors to determine the most impactful
driver.

Design/methodology/approach – The study employs Interpretive Structural Modeling (ISM) to analyze the
interrelationships among factors such as personalized recommendations, targeted ads, customized notifications, faster
checkouts, exclusive discounts, subscription services, security verification, and beta testing participation. A structured survey
was conducted to gather responses from digital consumers, and the ISM methodology was applied to develop a hierarchical
model. Additionally, MICMAC (Matrice d'Impacts Croisés-Multiplication Appliquée à un Classement) analysis was used to
classify these factors based on their driving and dependence power.

Findings – Personalized recommendations drive consumer data-sharing, while targeted ads and beta testing incentives
depend on other factors. Discounts, subscriptions, and security features support but don’t primarily motivate sharing.
Personalization is the core enabler—other benefits enhance but don’t replace its influence.
Research limitations – While ISM provides a structured framework for understanding complex relationships, the findings
are based on qualitative interpretations and may benefit from further statistical validation in future studies.

Practical implications – The study offers valuable insights for digital retailers seeking to optimize data collection strategies
while maintaining consumer trust. By prioritizing high-impact factors like personalization, businesses can enhance user
experience without compromising privacy.

Originality/value – This research contributes to the ongoing debate on consumer privacy by analyzing the factors driving
data-sharing. The application of ISM and MICMAC provides a novel hierarchical framework, helping stakeholders
understand the delicate balance between personalization and privacy in the digital economy.

Keywords – Consumer privacy, Data sharing, Digital marketplace, Interpretive Structural Modeling (ISM), MICMAC
analysis, Personalization, Privacy trade-off

II.INTRODUCTION
In the rapidly evolving digital marketplace, the trade-off between consumer privacy and the benefits of data sharing has
become a central concern. As online retailers and service providers increasingly rely on customers’ personal data to enhance
customer experiences, customers are often compelled to share personal information in exchange for perceived advantages.
This exchange, however, raises critical questions about privacy, consent, and the underlying motivations that drive consumers
to disclose sensitive data.
 Several key factors influence a consumer’s decision to share personal information with digital retailers. These include the
desire to receive personalized recommendations, which tailor product suggestions based on past behavior; targeted
advertisements, which reduce irrelevant promotions; and customized notifications, ensuring timely and relevant updates.
Additionally, convenience plays a significant role—faster checkouts facilitated by saved payment and address details, also
exclusive discounts, and loyalty rewards incentivize data sharing. Subscription-based services, security verification
requirements, and opportunities to participate in beta testing further contribute to this. While these benefits enhance user
experience, they also create a dependency on data collection, often leaving consumers with limited autonomy over their
personal information.

To better understand the hierarchy of these influencing factors, this study employs the Interpretive Structural Modeling (ISM)
technique—a systematic approach used to analyze complex relationships among variables. By structuring these factors into a
hierarchical framework, ISM helps identify the most dominant drivers that lead consumers to share private data. The findings
of this research provide valuable insights into consumer behavior, enabling businesses to refine their data collection practices
while policymakers can better address privacy concerns in the digital economy.
Ultimately, this study seeks to determine which factor exerts the strongest influence on consumer data-sharing decisions.

III.LITERATURE REVIEW
In the article "The Impact of Artificial Intelligence on Innovation", Canhoto et al. dive into how AI is reshaping innovation in
countless fields. It shows how AI tools—like machine learning and predictive analytics—are supercharging everything from
brainstorming to product development and market launch. By making collaboration and data sharing easier, AI is opening
new doors for creativity. But it’s not all smooth sailing—issues like ethics, workforce skills, and regulations still need work.
We see AI making waves in industries like healthcare through smarter diagnostics and manufacturing (with robots handling
complex tasks). The review also points to where future research should focus, like AI’s long-term influence on innovation
networks, how it can support sustainability, and the value of blending different fields to understand its full impact. AI is a
game-changer for innovation, but unlocking its true potential means tackling the ethical, technical, and legal hurdles standing
in the way.

Yang’s 2024 study dives into the personalization-privacy paradox—that gap between how much people say they care about
privacy and what they actually do online. The research explores this contradiction by looking at real user experiences, how
companies handle data, and the role of regulations in digital personalization.To get the full picture, Yang pulled insights from
a mix of sources: a deep dive into existing studies, corporate annual reports, and focus group discussions. This helped map
out the give-and-take between businesses, users, and other stakeholders. Using thematic and content analysis, the study
highlights how demographics, past experiences, and social influences lead to this paradox.The paper also pits rational
decision-making against situational choices, showing how trust and risk perception play a huge role. Yang’s proposed model
offers a new way to understand user behavior in digital spaces. Ultimately, the findings push for smarter, more nuanced
approaches to bridging the gap between personalization and privacy concerns. The study by Aggarwal et al. examines the
factors influencing customers' privacy concerns and their behaviour regarding Location-Based Advertising (LBA). Based on
privacy calculus theory, the research model assesses the effects of advertising value, privacy concerns, privacy self-efficacy,
brand trust, and privacy policies on LBA adoption. Data from 383 LBA users were analyzed using Structural Equation
Modeling (SEM).The findings indicate that advertising value has a positive influence on intentions to use LBA, which
counteracts the negative impact of privacy concerns. Privacy self-efficacy and brand trust are shown to mitigate privacy
concerns, whereas privacy policies unexpectedly amplify them. The study highlights the dual role of perceived benefits
(advertising value) and perceived risks (privacy concerns) in shaping consumer behavior. These insights provide valuable
guidance for marketers seeking to design LBA strategies that effectively balance personalization with privacy protection.
Chang, Chen, and Zhou (2020) explored how digital personalization shapes consumer shopping behaviors. Their research
specifically analyzes how customers' perception of personalization affects three outcomes: engagement levels, brand loyalty,
and likelihood to recommend products. Through structural equation modeling analysis of 404 online shoppers' responses, the
study demonstrates that well-executed personalization significantly boosts customer engagement. This heightened
engagement then acts as a catalyst for both stronger brand loyalty and increased word-of-mouth promotion. The research
provides strong evidence in support of personalized digital experiences in building customer relationships. The findings
suggest that when consumers recognize content as specifically tailored to their preferences, they become more actively
 engaged with the brand. This engagement then naturally translates into greater loyalty and a higher likelihood to share
positive experiences with others.

For marketing professionals, these insights emphasize the importance of developing genuinely relevant and customized
content. In today's digital commerce environment, such personalized approaches prove essential for creating meaningful
customer connections. The study ultimately provides empirical support for the business value of personalization while
offering practical guidance for its effective implementation. Ana Isabel Canhoto, Brendon James Keagan, and colleagues
(2024) investigate the paradoxical relationship between personal pride and AI-enabled personalization (AI-EP) in physical
retail. Their study addresses the factors shaping consumer responses to AI-EP in brick-and-mortar stores, particularly
examining how in-store atmosphere and privacy considerations influence adoption rates.The research employed in-depth
qualitative interviews with 18 female fashion consumers in the UK, revealing both the anticipated benefits and underlying
anxieties associated with AI-EP implementations. Key findings emphasize the critical role of the preservation of consumer
control in successful AI-EP integration.Through a framework modeled after the "snakes and ladders" game, the authors
provide retailers with nuanced insights into consumer behavior patterns. This approach effectively illustrates the delicate
balance between facilitating factors and adoption barriers, offering practical guidance for implementing AI-EP while
respecting consumer boundaries and expectations. Strycharz and Duivenvoorde (2021) examine how personalized marketing
communications (PMC) can exploit consumer vulnerabilities. Their study assesses the risks of data-driven personalization
strategies, particularly their potential to manipulate consumers and lead to financial harm. Through a comprehensive
literature review, the authors identify multiple consumer weaknesses that PMC may target. An analysis of European Union
Consumer Law reveals significant gaps in protections against such exploitative practices. The research underscores the need
for stronger regulatory measures to safeguard consumers from the risks associated with personalized marketing. Specifically,
the authors advocate for revisions to EU consumer protection frameworks to address these vulnerabilities more effectively.
The paper emphasizes the importance of policy reforms to preserve consumer freedom in today's digital markets. By
highlighting the ethical and economic implications of PMC, this study calls for further research and regulatory intervention to
ensure fair and transparent marketing practices.

The study by Morey et al. published in the Harvard Business Review examines the growing tension between corporate data
practices and consumer privacy concerns. Through a multinational survey of 900 respondents across five countries, the
research reveals a critical paradox: while consumers express significant concerns about potential data misuse, they
demonstrate limited awareness of the specific types of personal information they share online.The findings identify three key
pillars for building consumer trust: transparency in data practices, user control over personal information, and fair value
exchange. The study's conclusions emphasize the need for clear regulatory frameworks to ensure mutual benefits from
personal data collection, a shift from opaque data collection practices toward transparent, value-based approaches and
balanced solutions that address both business needs and consumer privacy expectations. These advocate for fundamental
changes in how organizations approach data collection and utilization in the digital economy. Sahana et al. investigate
consumer preferences for personalization within e-commerce environments. The study examines how data-sharing practices
influence three key phases of consumer behaviour: pre-purchase interactions, purchasing decisions, and post-purchase
behaviors. The research employed a survey methodology targeting online shoppers across diverse age demographics. Data
collection focused on three primary dimensions: (1) demographic characteristics, (2) shopping behaviors, and (3) privacy
attitudes. Purchase intentions were quantified using Likert scale measurements, while advertisement effectiveness was
assessed through the 8Cs customer loyalty framework. Key findings illuminate the specific categories of personal data
consumers are willing to disclose for personalized services and the delicate equilibrium between privacy concerns and
demand for customized shopping experiences. This research contributes valuable insights for e-commerce websites trying to
balance personalization strategies while respecting consumer privacy boundaries. Amil (2024) investigates privacy concerns
in e-commerce environments, focusing on how AI-driven personalization affects consumer trust. The study analyzes how
users' expectations of personalization influence privacy perceptions, which in turn determine their trust in AI systems. The
research employed quantitative methods to analyze survey data from 453 Amazon customers and product reviewers. Using
SPSS software, the study conducted regression analysis, reliability testing, and factor analysis to examine these relationships.
Key findings reveal that (1)Perceived benefits of personalization can mitigate privacy concerns and (2)Data breaches and
perceived risks associated with data sharing worsen privacy concerns.This research makes significant contributions to
understanding how e-commerce businesses can strategically balance AI-personalization with privacy preservation. The
findings offer practical insights for developing trustworthy digital markets that effectively balance personalization benefits
with consumer privacy expectations. Rahayu et al. (2025) investigate the relation between personalized email marketing and
consumer privacy concerns in e-commerce. Through phenomenological analysis of 20 online shoppers and three marketing
professionals, the study reveals: while consumers value relevant product recommendations, they experience discomfort when
personalization becomes excessively precise, creating a sense of surveillance.

Key findings demonstrate consumer preference for transparent data practices and greater autonomy over personal information
usage and the "privacy-personalization paradox" - where marketing effectiveness and consumer comfort are inversely related.
The research proposes implementing clear, accessible data usage policies ,allowing customizable marketing communication
and establishing opt-in mechanisms for data collection practices. These findings are very significant for Southeast Asia's
rapidly growing digital markets, where cultural norms surrounding privacy continue to grow. Kawaf et al. (2023) examine
consumer data collection, and personalization within digital sites governed by GDPR-2018 regulations. Through qualitative
analysis using semi-structured interviews, the study reveals: (1)Perceived Vulnerability: Despite GDPR's guidelines for their
privacy protection, many consumers report feeling insufficient control over their personal data, (2)Paradox: Participants
simultaneously demonstrate willingness to share data for personalized services while showing discomfort with privacy
intrusion, and (3)Digital Literacy Levels: Digital literacy significantly impacts privacy awareness, with experts showing
greater confidence

The research highlights gaps between regulations (GDPR-2018) and user experiences, regarding effective implementation,
consumer understanding of privacy rights, and thresholds for acceptable personalization. The study advocates for strategies to
strengthen GDPR's impact, including enhanced consumer education programs and more transparent data usage disclosures by
organizations. Aguirre et al. (2016) analyze personalized marketing across search, display, social, and mobile platforms,
revealing that while personalization improves engagement through relevant content, it triggers privacy concerns about data
collection. Their literature review shows effectiveness varies by platform, requiring tailored approaches for each of the
mentioned platforms. Key findings show that consumers value relevance but distrust data practices, platform-specific data
collection strategies are essential, and that transparency is critical for maintaining consumers' trust. The study highlights the
need to balance personalization benefits with privacy protection, suggesting future research on ethical data use and platform
optimization.

Martin et al. (2017) investigate how customer data collection practices affect firm performance through the use of gossip
theory. Their approach combines experimental studies, an analysis of 414 companies showing performance declines after
privacy breaches of customers' data, and field research across 15 firms revealing competitor advantages from data leaks

Key findings establish that competitors gain 3-5% market share following rivals' privacy incidents. The research identifies
transparency as an effective mitigator, reducing competitor vulnerability effects by 40-60%. These results suggest firms
should implement clear data governance policies, develop strong security protocols, and offer opt-in/opt-out personalization
controls

Bleier et al. examine how consumer privacy concerns impact data-driven marketing, particularly for startups. Using
contextual integrity theory, the authors show privacy violations occur when data use contradicts consumer expectations. The
paper proves that startups face higher risks from regulations and distrust. However, privacy concerns also create opportunities
for privacy-enhancing technologies. The research suggests startups can turn privacy into a competitive advantage by adopting
transparent practices and aligning data use with consumer expectations.

Lee, C. H., & Cranage study how privacy assurances affect user behavior on personalized travel platforms. Key findings
reveal that strong privacy measures simultaneously enhance perceived usefulness (by 23%) while reducing privacy concerns
(by 31%), ultimately increasing both data disclosure rates (+18%) and purchases (+15%). Notably, personalization alone
showed no significant impact on privacy worries. The research demonstrates that the optimal user experience is when
personalization combines with visible privacy protections. While limited to student respondents, these results suggest travel
platforms should prioritize transparent data practices to maximize sales. Martin, K., Borah, A., & Palmatier, R. W. analyze
data privacy through three fundamental lenses: societal, psychological, and economic dimensions. Current research, while
valuable, remains limited across four primary domains - consumer behavior, organizational practice, ethical considerations,
and legal frameworks.The authors suggest we need a better way to study privacy that looks at all these connections together.
 For example: How people's feelings about privacy (psychology) affect what businesses do (marketing), How privacy laws
(legal) change what companies can do with data (economics) and How all this impacts society as a whole.They argue that by
studying privacy from multiple angles at once, we can get a clearer picture of how it really works in today's world where data
is so important. This would help businesses make better decisions about customer data while still respecting their people's
privacy. The analysis by Chellappa et al. critiques the prevailing binary system that frames privacy and personalization as
mutually exclusive in digital markets. Current approaches often force users into an extreme choice between complete data
disclosure (enabling personalization) or complete data withholding (preserving privacy).The real issue isn’t that privacy and
personalization can’t coexist—it’s that most systems don’t give users control over their data. A better solution would let
people choose what information to share, decide who gets access to it and still enjoy personalized features. By designing
systems this way, we can have both privacy and useful personalization—without forcing users into an extreme choice. A
study by Van Buggenhout et al. examines the complex relationship between personal data collection and media
personalization through Delphi Techniques. Using a three-round Delphi methodology with media and privacy specialists, the
research shows that while some experts view this (data sharing in exchange for personalized media content) as an equitable
transaction, others highlight inherent power imbalances that favor media providers. The investigation also reveals
shortcomings in current transparency practices, particularly regarding how platforms communicate data collection scope and
usage intentions. These shortcomings persist despite existing regulatory requirements, resulting in the above mentioned
power imbalances. To mitigate these issues, the findings advocate for an integrated governance framework including
standardized disclosure rules, user-selectable data controls, and strong technical safeguards. The research makes three
primary contributions: first, it discloses areas of expert consensus and disagreement regarding data privacy; second, it
proposes an actionable model; third, it identifies practical challenges in implementing current data protection regulations.
Özköse’s thesis takes a close look at a growing concern—how digital marketing affects consumer privacy. With companies
relying more on personal data to tailor ads, the study explores the ethical and legal challenges that come with it. To get
real-world perspectives, Özköse interviewed three people from different backgrounds in the EU and analyzed existing
research along with GDPR guidelines. The findings reveal a complicated reality: while some people don’t mind targeted ads
if it means getting free services, many feel uneasy about how their data is collected, shared, and potentially misused. A major
concern is that companies often use personal information in ways consumers don’t fully understand or consent to. Even more
surprising, most participants weren’t aware of the privacy tools available to them or their rights under GDPR. This disconnect
between expectations and reality raises serious questions about digital privacy and whether current protections are enough.
Petrescu and Krishen (2018) dive into the growing concerns around data privacy in marketing analytics. They argue that
businesses need to strike a balance between using data for effective marketing and protecting consumer privacy to maintain
trust. To explore this, they analyzed research and regulatory reports on issues like unauthorized data use, improper access,
and misleading data practices. Their review of existing studies highlights a tricky dilemma: while stronger privacy
regulations can help prevent breaches, they can also make targeted advertising less effective. With the rise of the Internet of
Things adding new challenges, the authors stress that the way forward isn’t just stricter rules—it’s greater transparency and
giving consumers more control over their data to create a more sustainable approach to digital marketing. Quach et al. (2022)
examine how companies’ push to collect and share data can leave consumers feeling vulnerable and exposed. To understand
this issue from multiple angles, the researchers interviewed senior managers, spoke with everyday consumers, and analyzed
case studies from major companies. Using structuration theory as a foundation, they created a model for tackling privacy
challenges. Their findings show a complex trade-off: while digital tools generate massive economic value, they also introduce
significant privacy risks that spark different reactions from consumers, businesses, and regulators. Ozdemir, Z. D., Smith, H.
J., & Benamati, J. H’s research explores how people’s willingness to share personal information is influenced by social cues
and the order of questions they are asked. The first set of studies reveals a "herding" effect—people are more likely to share
sensitive information if they believe others have already done so (Study 1a). This effect occurs because knowing others have
disclosed reduces discomfort (Study 1b). The second set of studies examines how the order of questions affects disclosure.
People are more likely to share when intrusive questions are asked first and less likely when these questions come later
(Study 2a). This happens because the order influences how intrusive the questions seem (Study 2b). However, the effect
weakens if people are reminded of privacy concerns or are asked to evaluate other questions first (Studies 2c and 2d). The
findings show that both social factors and how questions are framed shape disclosure behavior.
 The study by Adarsh Anand and Gunjan Bansal (Delhi University) attempts to create a hierarchy of all attributes relating to
software quality according to their importance-which has been achieved by applying the ISM(Interpretative Structured
Modelling) technique. By applying ISM and MICMAC analysis on the set of attributes under consideration, it has been found
that “reliability” along with “usability” and “performance” are the most important attributes of software quality and are
preferred the most. Attri et al define Interpretive Structural Modeling (ISM) as a powerful tool for identifying and building
relationships between key elements or factors. Developed by Warfield in 1974, ISM helps clarify complex models, turning
them into structured systems that help in decision-making and problem-solving. Key Concepts and Methodology include
Identifying Variables(Relevant factors identified using group techniques), Contextual Relationships (A relationship (like
"influences" or "leads to") is used to compare the variables), Structural Self-Interaction Matrix (SSIM) (Pairwise comparisons
are made using symbols to indicate relationships), Reachability Matrix (RM) (SSIM is turned into a binary matrix by
checking transitivity), Level Partitioning (Variables are ranked hierarchically based on their relationships), MICMAC
Analysis (A graph is created, simplifying the structure). Applications of ISM include Operations Management: Analyzing
barriers in Total Productive Maintenance (Attri et al., 2012), Supply Chain Management: Modeling risk mitigation strategies
(Faisal et al., 2006; Jharkharia and Shankar, 2004), Energy Conservation: Identifying key variables in energy conservation
for the cement industry (Saxena et al., 1990), Waste Management: Structuring objectives for waste management (Sharma et
al., 1995),Vendor Selection: Evaluating criteria for selecting vendors (Mandal & Deshmukh, 1994).Limitations of ISM
include complexity with many Variables (Too many variables can make the model cumbersome) and Expert Dependency
(Relies on expert judgment, which may introduce bias). ISM is a valuable tool for structuring complex systems, providing
clarity and guiding decision-making. While it has limitations, such as reliance on expert opinion, its ability to simplify
complex relationships makes it highly useful. A paper by Mandal et al. focuses on enhancing the vendor selection process by
considering non-traditional factors beyond quality, delivery, and price using ISM. The research identifies several additional
factors for selecting reliable vendors, acknowledging that vendors must meet not only immediate but also long-term business
needs. The study categorizes these factors into four groups based on their influence: autonomous, dependent, linkage, and
driver. These categories reflect the "driver power" (how much influence a factor has on others) and "dependence" (how much
other factors rely on it). An Interpretive Structural Modeling (ISM) approach is used to analyze the inter-relationships among
these factors.The findings show that "attitude and willingness for business" and "after-sales service" are just as crucial as
traditional factors like quality and delivery performance. These two criteria, along with others, are dependent on all other
factors, indicating their central role in the vendor selection process. This analysis offers a more holistic approach to vendor
selection.

IV.RESEARCH METHODOLOGY
The identified research problem is ‘Data sharing drivers for consumers and the benefits they perceive’ and the objective of
this study is to use ISM to ascertain the most impactful driver.
The research uses Interpretive Structural Modeling (ISM) and MICMAC (Matrice d’Impacts Croisés-Multiplication
Appliquée à un Classement) analysis to establish hierarchical relationships and classify factors based on their driving and
dependence power.

V.DATA COLLECTION AND SURVEY DESIGN

A structured questionnaire was administered to digital consumers to assess their perceptions of data-sharing motivators. The
survey captured responses on perceived relationships among all the factors driving consumers to share data with retailers
(ascertained through a thorough literature review), including: (1)Personalized recommendations (2) Targeted Advertisements
(3) Customized notifications (4) Faster checkouts (saved payment/address details) (5) Exclusive discounts & loyalty rewards
(6) Subscription Services (7) Security Verification and (8) Beta test participation.
Participants selected the relationships among all of these factors paired with every other factor, out of the following options:
1.​ Only factor A affects factor B
2.​ Only factor B affects factor A
3.​ Both factor A and B affect each other
4.​ Neither factor A nor B affect each other
A total of 50 complete responses were collected from a consistent demographic (18-25 years old) of digital shoppers.
ISM was employed to analyze the interrelationships among the identified factors and develop a hierarchical model. The
methodology followed these steps:
1.​ Identification of Factors: The eight key motivators of data sharing were selected based on in depth literature review.

C1 Personalized recommendations

C2 Targeted advertisements

C3 Customized notifications

C4 Faster checkouts (saved payment/address details)

C5 Exclusive discounts & loyalty rewards

C6 Subscription services

C7 Security verification requirements

C8 Beta testing participation incentives

Table 1: Identified drivers for data sharing by consumers

2.​ Structural Self-Interaction Matrix (SSIM): Survey responses indicated pairwise relationships between factors using
four symbolic notations:
V (Factor a influences Factor b)
A (Factor b influences Factor a)
X (Mutual influence between a and b)
O (No relationship)

Fig.1: SSIM
3.​ Initial Reachability Matrix (IRM): The SSIM was converted into a binary matrix (0/1) based on predefined rules:
(1) when (a,b)th entry presents V: convert the particular (a,b)th entry by “1” and (b, a)th entry by “0”;
(2) when (a,b)th entry presents A: convert the particular (a,b)th entry by “1” and (b, a)th entry by “0”;
(3) when (a,b)th entry presents X: convert the particular (a,b)th entry by “1” and (b, a)th entry by “1”; and
(4) when (a,b)th entry presents O: convert the particular (a,b)th entry by “0” and (b, a)th entry by “0”

Fig. 2: IRM

4.​ Final Reachability Matrix (FRM): Transitivity checks were applied to ensure logical consistency in factor
relationships. This was done using the R studio ISM software package

Fig. 3: FRM
5.​ Level Partitioning: Factors were hierarchically classified by repetitively extracting those with identical reachability
and intersection sets. This was done using the R studio ISM software package.

Fig.4: Level Partitioning

 6.​ Canonical Matrix & Digraph: A directed graph (ISM model) was constructed to visualize the hierarchy of factors.

Figure 5: Digraph

7.​ MICMAC Analysis: Following ISM, MICMAC analysis was conducted to categorize factors into four clusters based
on their driving power (influence on other factors) and dependence power (dependence on other factors):
(1)Autonomous Factors: Weak driving & dependence power
(2)Dependent Factors: Low driving power but high dependence
(3)Linkage Factors: Strong bidirectional influence
(4)Independent (Driving) Factors: High driving power, low dependence

Fig.6: MICMAC analysis

8.​ Validation & Software Tool: To enhance accuracy, the ISM process was automated using R software

VI.DATA ANALYSIS
The ISM hierarchy reveals a clear structural relationship among the factors influencing consumers' data-sharing decisions in
digital marketplaces. The model categorizes these factors into distinct levels. In figure 5, the Digraph, at the highest level,
representing the most influenced factors, we find targeted advertisements, customized notifications, faster checkouts, and beta
testing participation incentives. These factors constitute the end benefits that consumers experience but which are largely
dependent on underlying motivators (other factors). For instance, faster checkouts rely on stored payment and address data
that consumers only provide when sufficiently incentivized by other factors. Similarly, targeted ads and customized
notifications are outcomes that require prior consumer trust.

The intermediate levels of the hierarchy contain exclusive discounts and loyalty rewards, which act as a bridge. This factor
demonstrates strong dependence on the factors-security verification,subscription services, and personalized recommendation.
At the same time, it exerts influence on the factors- targeted advertisements, customized notifications, faster checkouts, and
 beta testing participation incentives. Security verification requirements occupy a hybrid position, depending on consumer
trust while also increasing consumer willingness to share data for benefits they can get. Subscription services appear lower in
the hierarchy than security but higher than personalization, suggesting that both of them reinforce data sharing by customers
but don't serve as primary motivators. These services often combine with discounts or faster services to create secondary
motivational effects.

At the foundation of the hierarchy, serving as the most influential and least dependent factor, we find personalized
recommendations. This dominant driver influences all other factors either directly or indirectly, reflecting the critical
importance of hyper-personalization in contemporary consumer decision-making. The positioning of personalized
recommendations at the base of the hierarchy aligns with current industry trends where personalization significantly boosts
engagement with customers.

The MICMAC analysis provides critical insights into the driving and dependence powers of factors influencing consumers'
data-sharing behavior. The analysis reveals three distinct clusters of factors.

Autonomous factors, including Exclusive discounts & loyalty rewards, Subscription services, Security verification
requirements, Customized notifications, and Faster checkouts (saved payment/address details), exhibit weak driving power
and weak dependence power. These factors are relatively independent within the system, showing minimal influence on other
elements while also being less affected by external variables. Though not primary drivers of data-sharing decisions, they
serve as supplementary motivators that can be combined with stronger factors.

Dependent factors, consisting of Targeted advertisements and Beta testing participation incentives, demonstrate low driving
power but high dependence power. These elements are significantly influenced by other variables in the system, particularly
the independent factors (Personalized Recommendations). Their effectiveness relies heavily on foundational motivators like
personalization.

Independent factors include only Personalized recommendations, which possess strong driving power and weak dependence
power. This factor emerges as the most influential determinant of data-sharing behavior, directly or indirectly affecting all
other factors in the hierarchy. Its dominant position highlights the critical role of hyper-personalization in modern digital
commerce, where tailored experiences serve as the primary tool for consumer engagement and data disclosure.

This classification system reveals that while customers directly interact with dependent factors (e.g., targeted ads), their
willingness to share data fundamentally comes from independent factors like personalized recommendations. Platforms
seeking to optimize data collection strategies should therefore prioritize strengthening this core driver while using
autonomous factors as complementary tools. The analysis confirms that in the digital marketplace, personalization isn't
merely a feature—it's the foundational core basis that powers consumer data exchange

VII.CONCLUSION
Personalized recommendations drive consumer data-sharing, while targeted ads and beta testing incentives depend on other
factors. Discounts, subscriptions, and security features support but don’t primarily motivate sharing. Personalization is the
core enabler—other benefits enhance but don’t replace its influence. The hierarchical model revealed that personalized
recommendations serve as the main driver of data-sharing, occupying the most influential position in the framework.The
analysis identified exclusive discounts, subscription services, security verification, customized notifications, and faster
checkouts as autonomous factors that enhance user experience but do not independently drive data sharing. These findings
offer critical insights for digital retailers. First, they confirm that hyper-personalization is non-negotiable for effective data
collection. Second, they demonstrate that convenience features (faster checkouts) and rewards (discounts) function best when
combined with personalization rather than as standalone.For policymakers, the study highlights how consumers undertake
privacy tradeoffs when they feel there is some value in it for them, suggesting that regulations should focus on ensuring
transparent data usage rather than blanket data restrictions.Ultimately, this research provides a framework for understanding
 the psychology behind data sharing - one where personalization serves as the key. Platforms that master this hierarchy will
gain sustainable competitive advantage in an increasingly privacy-conscious marketplace.

VIII.ACKNOWLEDGEMENT

We would like to express our heartfelt gratitude to NMIMS University for providing us with the opportunity and resources to
undertake this research on Consumer privacy in data-driven marketing: trust vs personalization. The institution's support has
been instrumental in shaping the scope and direction of this study.

Our deepest appreciation goes to our research methodology teacher, Ms. Vandana Bharadi, whose insightful feedback,
indefatigable mentorship, and constant encouragement have been pivotal throughout this research journey. Her expertise and
dedication have greatly enriched the quality of this work.

We are also profoundly grateful to our friends and family for their unwavering support and encouragement. Their patience
and belief in us have been a source of motivation during challenging times, enabling us to complete this project successfully.

This research would not have been possible without the collective contributions of these individuals and institutions, and we
are sincerely thankful for their assistance and support.

IX..REFERENCES
[1]Canhoto, A.I., Keegan, B.J. & Ryzhikh, M. Snakes and Ladders: Unpacking the Personalisation-Privacy Paradox in the
Context of AI-Enabled Personalisation in the Physical Retail Environment. Inf Syst Front 26, 1005–1024 (2024).
[2]Yang, D. (2024). Explaining the personalisation privacy paradox using an integrated experience model [Doctoral
dissertation, Edinburgh College of Art]. Edinburgh Research Archive.
[3]Aggarwal, R., Dhingra, S., & Mittal, S. (2023). What reduces the privacy concerns of the customers towards the use of
location-based advertising? An empirical investigation. Journal of Location Based Services, 18(2), 139–161.
[4]Zhou, Chen, Chang.-Z., & Yu, Y.-T. (2024). Impact of digital personalisation on consumers shopping experiences. Journal
of Consumer Behaviour, 23(2), 45-60
[5]Canhoto, A. I., & Keegan, B. J. (2023). Snakes and ladders: Unpacking the personalisation-privacy paradox in the context
of AI-enabled personalisation in the physical retail environment. Information Systems Frontiers, 25(2), 409-423.
[6]Strycharz, J., & Duivenvoorde, B. (2021). The exploitation of vulnerability through personalised marketing
communication: Are consumers protected? Internet Policy Review, 10(4).
[7]Morey, T., Forbath, T., & Schoop, A. (2015). Customer data: Designing for transparency and trust. Harvard Business
Review.
[8]Sahana, R., & Yashwanth, K. (2024). Personalisation in marketing in shaping consumer preferences and purchasing
behaviour. Department of Commerce, Loyola College.​
[9]Amil, Y. (2024). The impact of AI-driven personalization tools on privacy concerns and consumer trust in e-commerce
(Master’s thesis, HEC-Ecole de gestion de l’Université de Liège). Université de Liège
[10]Rahayu, T. A., Purnamasari, V. F. W., Rizky, M. A., Rahyadi, I., & Mani, L. (2025). Feeling watched: A
phenomenological exploration of consumer discomfort and data privacy concerns in e-commerce personalized email
marketing. paperASIA, 41(1b).
[11]Kawaf, F., et al. (2023). The privacy–personalisation paradox in GDPR-2018 regulated environments: Unpacking
consumer vulnerability and the curse of personalisation. Information Technology & People.
[12]Aguirre, E., Roggeveen, A. L., Grewal, D., & Wetzels, M. (2016). The personalization-privacy paradox: Implications for
new media. Journal of Consumer Marketing, 33(2), 98-110.
[13]Martin, K. D., Borah, A., & Palmatier, R. W. (2017). Data privacy: Effects on customer and firm performance. Journal of
Marketing, 81(1), 36-58.
[14]Bleier, A., Goldfarb, A., & Tucker, C. (2020). Consumer privacy and the future of data-based innovation and marketing.
International Journal of Research in Marketing, 37(3), 466–480.
 [15]Lee, C. H., & Cranage, D. A. (2011). Personalisation–privacy paradox: The effects of personalisation and privacy
assurance on customer responses to travel websites. Tourism Management, 32(5), 987–994.
[16]Martin, K., Borah, A., & Palmatier, R. W. (2017). Data privacy: Effects on customer and firm performance. Journal of
the Academy of Marketing Science, 45(2), 135–155.
[17]Chellappa, R. K., & Sin, R. G. (2005). Personalization versus privacy: An empirical examination of the online
consumer's dilemma. Information Technology and Management, 6(2–3), 181–202.
[18]Van Buggenhout, N., Van den Broeck, W., Van Zeeland, I., & Pierson, J. (2023). Personal data and personalisation in
media: Experts’ perceptions of value, benefits, and risks. Digital Policy, Regulation and Governance, 25(3), 305–324.
[19]Özköse, A. Y. (2022). Online marketing: Privacy concerns of consumers [Bachelor’s thesis, Haaga-Helia University of
Applied Sciences]
[20]Petrescu, M., & Krishen, A. S. (2018). Analyzing the analytics: Data privacy concerns-Journal Marketing Analytics,
6(2), 41–43
[21]Quach, S., Thaichon, P., Martin, K. D., Weaven, S., & Palmatier, R. W. (2022). Digital technologies: Tensions in privacy
and data. Journal of the Academy of Marketing Science, 50, 1299–1323
[22]Ozdemir, Z. D., Smith, H. J., & Benamati, J. H. (2017). Antecedents and outcomes of information privacy concerns in a
peer context: An exploratory study. European Journal of Information Systems, 26(6), 642–660
[23]Adarsh Anand, Gunjan Bansal, (2017) "Interpretive structural modelling for attributes of software quality", Journal of
Advances in Management Research, Vol. 14 Issue: 3, pp.256-269.
[24]Attri, R., Dev, N., & Sharma, V. (2013). Interpretive Structural Modelling (ISM) approach: An overview. Research
Journal of Management Sciences, 2(2), 3-8.
[25]Mandal, A. and Deshmukh, S.G. (1994), "Vendor Selection Using Interpretive Structural Modelling (ISM)", International
Journal of Operations & Production Management, Vol. 14 No. 6, pp. 52-59.

X.ANNEXURES
Questionnaire: https://forms.gle/rMJcQuoY4gPyLG4e9