: 10.56472/25835238/IRJEMS-V5I6P130
Bhadrika Evandito Atmomintarso, Sahat Hutajulu. "From Regulation Search to Synthesis: A Design Thinking Approach for Model Context Protocol (MCP) Adoption in a Tax Firm" International Research Journal of Economics and Management Studies, Vol. 5, No. 6, pp. 263-274, 2026. Crossref. http://doi.org/10.56472/25835238/IRJEMS-V5I6P130
This study examines how the Model Context Protocol (MCP) can bridge a tax firm's internal knowledge base to the generative AI ecosystem, closing the retrieval friction that separates validated regulatory content from the AI tools practitioners use, and evaluates the integration through a qualitative Task-Technology Fit (TTF) lens. The study adopts a qualitative single-case design at a mid-sized Indonesian tax consultancy, guided by the five-stage Design Thinking framework. Data came from three focus group discussions with 11 tax professionals, a key informant interview, and think-aloud usability testing with 5 participants. Reflexive Thematic Analysis structured the exploratory data, and Framework Analysis using the eight TTF dimensions structured the evaluative data. Two prototypes were developed in parallel: a firm-owned front end with a citation guardrail and a thin client on a mature chat surface. Operational burden consistently converged on a burden of trust, elevating verifiable source citation to a non-negotiable design requirement. The evaluation produced a consistent dimensional split: the firm-owned front end was stronger on governance dimensions, and the thin client was stronger on surface dimensions, while willingness to integrate split by work context. The element bearing the integration load is the firm's MCP layer, not either front end. The study offers the first application of MCP as a bridging technology evaluated through a qualitative TTF lens in the Indonesian civil-law tax context, and advances three transferable propositions: adoption fit is gated by output verifiability rather than model capability; the integration load concentrates in the shared backend; and willingness to integrate is work-context contingent. It reframes AI investment toward mitigating malpractice risk.
[1] A. F. Aji, G. I. Winata, F. Koto, S. Cahyawijaya, A. Romadhony, R. Mahendra, K. Kurniawan, D. Moeljadi, R. E. Prasojo, T. Baldwin, J. H. Lau, and S. Ruder, One country, 700+ languages: NLP challenges for underrepresented languages and dialects in Indonesia, Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (2022) 7226-7249.
[2] B. Alarie, A. Niblett, and A. H. Yoon, Using machine learning to predict outcomes in tax law, Canadian Business Law Journal 58(3) (2018) 231-254.
[3] E. Ammenwerth, C. Iller, and C. Mahler, IT adoption and the interaction of task, technology and individuals: a fit framework and a case study, BMC Medical Informatics and Decision Making 6 (2006) 3.
[4] Anthropic, Introducing the Model Context Protocol, available at: https://www.anthropic.com/news/model-context-protocol (accessed 16 June 2026) (2024).
[5] Anthropic, Model Context Protocol specification, available at: https://modelcontextprotocol.io/ (accessed 16 June 2026) (2024).
[6] R. Bommasani, D. A. Hudson, E. Adeli, R. Altman, S. Arora, S. von Arx, and P. Liang, On the opportunities and risks of foundation models, arXiv (2021), available at: https://arxiv.org/abs/2108.07258 (accessed 16 June 2026).
[7] V. Braun, and V. Clarke, Reflecting on reflexive thematic analysis, Qualitative Research in Sport, Exercise and Health 11(4) (2019) 589-597.
[8] T. Brown, Design thinking, Harvard Business Review 86(6) (2008) 84-92.
[9] T. B. Brown, B. Mann, N. Ryder, M. Subbiah, J. Kaplan, P. Dhariwal, and D. Amodei, Language models are few-shot learners, Advances in Neural Information Processing Systems 33 (2020) 1877-1901.
[10] A. Cahyadini, J. I. G. Hutagalung, and Z. Muttaqin, The urgency of reforming Indonesia's tax law in the face of economic digitalisation, Cogent Social Sciences 9(2) (2023) article 2285242.
[11] C. M. Christensen, T. Hall, K. Dillon, and D. S. Duncan, Know your customers' jobs to be done, Harvard Business Review 94(9) (2016) 54-62.
[12] M. Coccia, Fishbone diagram for technological analysis and foresight, International Journal of Foresight and Innovation Policy 14(2/3/4) (2020) 225-247.
[13] M. Dahl, V. Magesh, M. Suzgun, and D. E. Ho, Large legal fictions: profiling legal hallucinations in large language models, Journal of Legal Analysis 16(1) (2024) 64-93.
[14] T. H. Davenport, Thinking for a Living: How to Get Better Performance and Results from Knowledge Workers, Harvard Business School Press, Boston, MA (2005).
[15] F. D. Davis, Perceived usefulness, perceived ease of use, and user acceptance of information technology, MIS Quarterly 13(3) (1989) 319-340.
[16] N. K. Denzin, Sociological Methods: A Sourcebook, 5th ed., Aldine Transaction, New Brunswick, NJ (2006).
[17] S. P. Dow, A. Glassco, J. Kass, M. Schwarz, D. L. Schwartz, and S. R. Klemmer, Parallel prototyping leads to better design results, more divergence, and increased self-efficacy, ACM Transactions on Computer-Human Interaction 17(4) (2010) 1-24.
[18] P. F. Drucker, Knowledge-worker productivity: the biggest challenge, California Management Review 41(2) (1999) 79-94.
[19] J. Faulconbridge, A. Sarwar, and M. Spring, How professionals adapt to artificial intelligence: the role of intertwined boundary work, Journal of Management Studies (2025), doe: 10.1111/jams. 12936.
[20] L. Faulkner, beyond the five-user assumption: benefits of increased sample sizes in usability testing, Behaviour Research Methods, Instruments, & Computers 35(3) (2003) 379-383.
[21] N. K. Gale, G. Heath, E. Cameron, S. Rashid, and S. Redwood, Using the framework method for the analysis of qualitative data in multi-disciplinary health research, BMC Medical Research Methodology 13 (2013) 117.
[22] D. L. Goodhue, Understanding user evaluations of information systems, Management Science 41(12) (1995) 1827-1844.
[23] D. L. Goodhue and R. L. Thompson, Task-technology fit and individual performance, MIS Quarterly 19(2) (1995) 213-236.
[24] R. M. Grant, Toward a knowledge-based theory of the firm, Strategic Management Journal 17(S2) (1996) 109-122.
[25] G. Guest, E. Namey, and K. McKenna, How many focus groups are enough? Building an evidence base for nonprobability sample sizes, Field Methods 29(1) (2017) 3-22.
[26] M. M. Hennink, B. N. Kaiser, and V. C. Marconi, Code saturation versus meaning saturation: how many interviews are enough?, Qualitative Health Research 27(4) (2017) 591-608.
[27] X. Hou, Y. Zhao, S. Wang, and H. Wang, Model Context Protocol (MCP): landscape, security threats, and future research directions, arXiv (2025), available at: https://arxiv.org/abs/2503.23278 (accessed 16 June 2026).
[28] Ikatan Konsultan Pajak Indonesia, Kode Etik Profesi Ikatan Konsultan Pajak Indonesia, IKPI, Jakarta (2019).
[29] K. Ishikawa, Guide to Quality Control, 2nd ed., Asian Productivity Organisation, Tokyo (1986).
[30] M. Klesel, and H. F. Wittmann, Retrieval-augmented generation (RAG), Business & Information Systems Engineering 67 (2025) 551-561.
[31] R. A. Krueger and M. A. Casey, Focus Groups: A Practical Guide for Applied Research, 5th ed., Sage, Thousand Oaks, CA (2015).
[32] P. Lewis, E. Perez, A. Piktus, F. Petroni, V. Karpukhin, N. Goyal, and D. Kiela, Retrieval-augmented generation for knowledge-intensive NLP tasks, Advances in Neural Information Processing Systems 33 (2020) 9459-9474.
[33] J. Liedtka, Why design thinking works, Harvard Business Review 96(5) (2018) 72-79.
[34] Y. S. Lincoln and E. G. Guba, Naturalistic Inquiry, Sage, Beverly Hills, CA (1985).
[35] V. Magesh, F. Surani, M. Dahl, M. Suzgun, C. D. Manning, and D. E. Ho, Hallucination-free? Assessing the reliability of leading AI legal research tools, Journal of Empirical Legal Studies 22(2) (2024) 216-242.
[36] T. J. McGill, and J. E. Klobas, A task-technology fit view of learning management system impact, Computers & Education 52(2) (2009) 496-508.
[37] D. L. Morgan, Pragmatism as a paradigm for social research, Qualitative Inquiry 20(8) (2014) 1045-1053.
[38] OECD, Tax Administration 2022: Comparative Information on OECD and Other Advanced and Emerging Economies, OECD Publishing, Paris (2022).
[39] M. Q. Patton, Qualitative Research and Evaluation Methods, 4th ed., Sage, Thousand Oaks, CA (2015).
[40] H. Plattner, C. Meinel, and L. Leifer, Design Thinking: Understand, Improve, Apply, Springer, Berlin (2011).
[41] A. Schipp, F. Siahaan, and C. Sureth-Sloane, Determinants of tax complexity in tax regulations and tax procedures: evidence from a developing country, Intertax 54(2) (2024) 102-122.
[42] R. Susskind and D. Susskind, The Future of the Professions: How Technology Will Transform the Work of Human Experts, Oxford University Press, Oxford (2015).
[43] M. R. Tambunan and H. Rosdiana, Indonesia tax authority measures to address challenges in taxing the digital economy, The International Technology Management Review 9(1) (2020) 1-10.
[44] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, and I. Polosukhin, Attention is all you need, Advances in Neural Information Processing Systems 30 (2017) 5998-6008.
[45] D. J. Wood, R. K. Mitchell, B. R. Agle, and L. M. Bryan, Stakeholder identification and salience after 20 years: progress, problems, and prospects, Business & Society 60(1) (2021) 196-245.
[46] R. K. Yin, Case Study Research and Applications: Design and Methods, 6th ed., Sage, Thousand Oaks, CA (2018).
[47] T. Zhou, Y. Lu, and B. Wang, Integrating TTF and UTAUT to explain mobile banking user adoption, Computers in Human Behaviour 26(4) (2010) 760-767.
Artificial Intelligence Adoption, Design Thinking, Knowledge Management, Model Context Protocol, Professional Service Firms, Task-Technology Fit.