Tuesday, 27 January 2026

Asking AI to offer an MBA research theme and a dissertation proposal: on AI strategy for the HK supermarket business

Asking AI to offer an MBA research theme and a dissertation proposal: on AI strategy for the HK supermarket business (information from perplexity.ai)


My question 1: "Please suggest an MBA research theme, via the pragmatism lens, based on the following key words: AI strategy, AI digital era in Hong Kong, corporate innovation performance of the Hong Kong supermarket sector"

The reply: "You could frame your project around how supermarkets in Hong Kong use AI strategically to drive innovation outcomes in practice, then study this through a mixed‑methods, pragmatist design.

Example overarching theme

“AI strategy and innovation performance in the Hong Kong supermarket sector: a pragmatist mixed‑methods inquiry in the AI‑driven digital era.”

This gives you:

  • Context: Hong Kong’s AI‑enabled retail digital transformation.

  • Sector: Supermarkets and grocery retail.

  • Focus: How AI strategy shapes corporate innovation performance (products, services, processes, customer experience, supply chain).

  • Lens: Pragmatism, emphasising actionable, context‑specific insights for managers.

Possible refined titles

You could choose one of these as your formal dissertation title:

  1. “From AI Strategy to Store‑Level Innovation: A Pragmatist Study of Hong Kong Supermarkets in the AI Digital Era.”

  2. “Designing Pragmatic AI Strategies for Innovation: Evidence from the Hong Kong Supermarket Sector.”

  3. “AI‑Enabled Digital Transformation and Innovation Performance in Hong Kong Supermarkets: A Pragmatist Mixed‑Methods Investigation.”

Each keeps your core keywords (AI strategy, AI digital era, Hong Kong, supermarket sector, corporate innovation performance) but signals a clear, applied business orientation.

Indicative research questions (pragmatist framing)

You can position the study around questions like:

  • How are Hong Kong supermarket chains formulating and implementing AI strategies (e.g., customer analytics, smart supply chains, automation, personalization)?

  • What kinds of innovation outcomes (new services, process improvements, omni‑channel offerings, data‑driven marketing) are associated with these AI initiatives?

  • Under what organisational conditions (leadership vision, culture, data capability, partnership ecosystems) do AI strategies most effectively translate into innovation performance gains?

  • What practical guidelines can supermarket executives in Hong Kong use to design “fit‑for‑purpose” AI strategies that measurably enhance innovation performance?

These questions focus on “what works, for whom, and under what conditions,” which aligns naturally with a pragmatist philosophy.

Why pragmatism fits this theme

  • It allows you to combine surveys (innovation performance measures), interviews (managerial sense‑making around AI), and possibly secondary data (digital initiatives, financials) in one coherent study.

  • It emphasises producing actionable knowledge for practitioners, which suits an AI‑strategy topic where supermarkets must make concrete investment and implementation decisions.

  • It is well aligned with studying organisational processes and change, such as AI‑enabled digital transformation in retail."


My question 2: "Based on the proposed research theme , please offer a part-time MBA dissertation proposal with the following information: 

1. Dissertation report title

2. 3 research issues (including research gaps) and 3 Hong Kong public media reported issues
3. 4 research objectives and 3 research questions
4. 2 primary research methods and 1 secondary research method
5. 2 Information systems strategy theories, 1 AI business application theory, and 1 innovation performance theory (with a reference list in Harvard reference format)
6. A brief justification of the dissertation proposal in terms of academic, practical and originality value."." 


The reply: "

Dissertation Report Title

AI Strategy and Corporate Innovation Performance in Hong Kong Supermarkets: A Pragmatist Mixed-Methods Inquiry in the AI Digital Era

Research Issues and Media-Reported Issues

3 Research Issues (with Gaps)

·        Limited empirical evidence on how AI strategies translate into measurable innovation outcomes (e.g., process efficiencies, new services) in mature retail sectors like Hong Kong supermarkets, despite global AI adoption trends.

·        Gap in context-specific studies linking AI implementation to supermarket performance amid Hong Kong's unique regulatory and competitive pressures, such as high rents and cross-border data flows.

·        Insufficient pragmatist-framed research exploring actionable AI strategy frameworks for mid-sized supermarkets, focusing on practical contingencies rather than universal models.

3 Hong Kong Public Media-Reported Issues

·        Supermarket chains like PARKnSHOP and AEON are trialing AI tools (e.g., unmanned stores, self-checkouts, loss prevention systems) but face scaling challenges in dense urban environments.

·        Intense competition from e-commerce and experiential retail demands AI-driven personalization, yet high operational costs and talent shortages hinder adoption.

·        Regulatory pressures, including AI transparency laws and cybersecurity bills effective 2026, complicate dynamic pricing and customer data use in Hong Kong retail.

Research Objectives and Questions

4 Research Objectives

·        To critically review AI strategy and innovation performance literature through a pragmatist lens tailored to Hong Kong supermarkets.

·        To identify key AI applications and their links to innovation outcomes in the sector.

·        To empirically assess how organisational factors moderate AI strategy effectiveness on performance.

·        To develop practical guidelines for supermarket executives on AI-driven innovation strategies.

3 Research Questions

·        How do Hong Kong supermarkets formulate and implement AI strategies, and what innovation outcomes result?

·        What contextual factors (e.g., leadership, regulations) influence the AI-innovation performance relationship?

·        What pragmatist-derived recommendations optimise AI strategies for sustained innovation in this sector?

Research Methods

2 Primary Methods

·        Semi-structured interviews with 15-20 executives from major chains (e.g., PARKnSHOP, AEON, Wellcome) to explore AI strategy sense-making and perceived innovation impacts (qualitative, interpretivist phase).

·        Online survey of 100+ store managers using validated scales for AI adoption, innovation performance, and contingencies (quantitative phase, targeting statistical associations).

1 Secondary Method

·        Analysis of company reports, annual results, and media archives (e.g., AEON 2024 results, HK retail news) for triangulation and longitudinal trends in AI initiatives.

Theoretical Frameworks

2 Information Systems Strategy Theories

·        Information Systems Strategy Triangle (Pearlson et al.): Aligns business, organisational, and IS strategies, ensuring AI initiatives support supermarket goals like efficiency and customer experience.

·        Galliers' IS/IT Strategy Model: Holistic framework integrating information, IT, management, and change strategies for emergent AI planning in dynamic retail contexts.

1 AI Business Application Theory

·        AI Business Value Chain (Davenport & Ronanki): Stages from data acquisition to cognitive engagement, explaining AI deployment for retail analytics, automation, and decision-making.

1 Innovation Performance Theory

·        Complex Innovation Strategy Framework (Karlsson et al.): Links combinations of product/process/marketing/organisational innovations to productivity gains, applied to supermarket AI outcomes.

 Harvard Reference List

Karlsson, C., Warda, P. and Gråsjö, U. (2015) Innovation Strategies and Firm Performance. Stockholm: Royal Institute of Technology. Available at: 

https://static.sys.kth.se/itm/wp/cesis/cesiswp401.pdf

 (Accessed: 27 January 2026).

Pearlson, K.E., Saunders, C.S. and Galletta, D.F. (year) Managing and Using Information Systems: A Strategic Approach. Wiley.

Davenport, T.H. and Ronanki, R. (2018) 'Artificial Intelligence for the Real World', Harvard Business Review, January-February. (Derived from AI value chain applications).

Galliers, R.D. (1993) 'Towards a flexible information architecture: Integrating business strategies, information systems and technology', Journal of Information Systems, 3(3), pp. 199-213.

Proposal Justification

This proposal holds academic value by advancing pragmatist MBA research on AI in retail through mixed methods and sector-specific theories, filling gaps in Hong Kong-focused IS studies. Practically, it equips supermarket leaders with evidence-based AI guidelines amid digital pressures, supporting HK's smart city ambitions and regulatory compliance. Originality stems from its novel integration of supermarket media insights with strategy theories, yielding actionable frameworks absent in prior literature."

My question 3: "As I only have 3 friends being senior managers of 2 Hong Kong supermarkets, 3 friends being middle manager of 2 Hong Kong supermarkets, 2 IT managers each of whom working a supermarket company, plus 1 university lecturer specialized in IS strategy, could you suggest a research methodology that can make use of my social network. I could spare more time on secondary research to make up for my social capital limitations."

The reply: "A pragmatist sequential mixed-methods design leverages your 9 supermarket contacts (senior/middle managers, IT managers) for targeted primary insights, supplemented by extensive secondary data to achieve robust MBA-level validity.

Feasibility Assessment

Your network provides 9 high-value informants across 4 firms (2 seniors, 3 middles, 2 ITs per unnamed chains), plus 1 IS expert—ideal for purposeful sampling in qualitative-dominant studies.

 

This covers diverse roles (strategy, operations, tech), enabling triangulation without needing 100+ surveys.

Secondary sources compensate for scale, common in pragmatist retail studies where access is constrained.

Proposed Methodology: Pragmatist Embedded Case Study

Adopt an embedded multiple-case study with qualitative primacy (your contacts) and quantitative secondary data embedded for validation—fits part-time MBA timelines (6-9 months).

Phase 1: Qualitative Primary (Months 1-3)

·        Semi-structured interviews (45-60 mins each) with all 10 contacts (9 supermarket + 1 lecturer).

·        Supermarket managers: Probe AI strategy formulation, implementation barriers/outcomes (e.g., "How has AI changed inventory or customer service?").

·        Lecturer: Validate frameworks (e.g., IS Triangle).

·        Sample size: Exactly your network (N=10)—saturation likely at 8-12 for case studies.

·        Analysis: Thematic coding via NVivo/Excel, yielding propositions (e.g., "AI succeeds when IT-business alignment is strong").

Phase 2: Quantitative Secondary (Months 2-4, parallel)

·        Analyse public data from 6-8 major chains (e.g., PARKnSHOP, AEON, Wellcome, U Select):

·        Annual/interim reports (HKEX filings): AI mentions, capex, revenue growth, innovation KPIs.

·        Media/press releases: 50+ items on AI pilots (unmanned stores, apps).

·        Industry reports (JLL, KPMG): Retail performance metrics (sales/sqft, digital adoption).

·        Metrics: Code for AI strategy intensity (scale 1-5), innovation performance (e.g., % digital sales).

·        Analysis: Simple regressions/correlations in Excel/SPSS (e.g., AI investment vs. growth).

·        Time-efficient: 20-30 hours sourcing/analysing.

Phase 3: Integration and Validation (Months 4-6)

·        Use primary themes to guide secondary coding; test propositions quantitatively (e.g., "High-alignment firms show 15% better outcomes").

·        Triangulation: Cross-validate via matrix (interviews vs. reports).

·        Snowballing: Ask contacts for 2-3 referrals each (aim +6 more, optional).

Role/ Network

Number

Contribution

Access Plan

Senior Managers

3 (2 chains)

Strategy vision, performance links

LinkedIn/ email: "Exploring AI impact -min chat?"

Middle Managers

3 (2 chains)

Implementation realities

Same chains: Coffee meetup in Fanling/ North District

IT Managers

4 (2 chains)

Tech specifics (e.g., data stacks)

Internal referrals

IS Lecturer

1

Theoretical framing

Academic network - virtual interview

Secondary Firms

4-6 more

Broader benchmarks

Public HKEX/ media - no access needed

 

Total cases: 4 firms primary + 4 secondary = credible scope.

Data Analysis Workflow

  1. Transcribe/record interviews → Thematic analysis (pragmatist codes: "what works?").

  2. Compile secondary dataset (Excel: Firm | Year | AI_score | Innovation_KPIs).

  3. Visualise: Innovation matrix plotting primary/secondary firms.

  4. Pragmatist synthesis: Guidelines matrix (e.g., "For cost-focused chains, prioritise supply-chain AI").

Advantages for Your Constraints

  • Network-maximising: Uses 100% of contacts deeply, no cold outreach.

  • Time-flexible: Interviews (20-30 hours total), secondary scalable to your schedule.

  • MBA-rigorous: Meets validity via multi-source triangulation, common in IS/retail theses.

  • Ethical: Low-risk (anonymous, informed consent).

This yields ~15,000 words: Intro/lit review (30%), Methods (15%), Findings (30%), Discussion (25%). If needed, I can draft your ethics form or interview guide.


A collection of blog notes on using chatgpt for research purpose.

No comments:

Post a Comment