Generative AI in developing User Experience Research Point of View: A NotebookLM case study

Huseyin Dogan; Mona Giff; Stephen Giff

arxiv: 2605.31125 · v1 · pith:YYFB2JFKnew · submitted 2026-05-29 · 💻 cs.HC

Generative AI in developing User Experience Research Point of View: A NotebookLM case study

Mona Giff , Stephen Giff , Huseyin Dogan This is my paper

Pith reviewed 2026-06-28 21:21 UTC · model grok-4.3

classification 💻 cs.HC

keywords Generative AIUser Experience ResearchNotebookLMUXR Point of ViewPrompt EngineeringCollaborative AIData-driven approachesStrategic product impact

0 comments

The pith

A five-prompt method lets NotebookLM build evidence-based UXR points of view that drive product decisions.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops and tests a formal methodology for using NotebookLM to support the User Experience Research Point of View framework. It breaks the process into four stages with five specific prompts that guide the AI from raw data to strategic narratives. When tested on eleven existing UXR papers, the AI successfully followed the framework at every step. This approach addresses the time lag in traditional UXR methods by making GenAI a reliable collaborator rather than an extra burden. A sympathetic reader would see this as a way to make research outputs more timely and impactful for product teams.

Core claim

The proposed methodology of five prompts across four stages—leveraging the framework, establishing roadmaps, applying best-practices, and crafting PoV narratives—enables NotebookLM to augment the UXR PoV process. On eleven test papers, it successfully leveraged the framework across all stages, demonstrating that NotebookLM can serve as an effective collaborative partner when provided with sufficient context and specific prompting.

What carries the argument

The UXR Point of View (PoV) framework, which transitions from raw data collection to an evidence-based PoV that drives strategic product impact, paired with a structured five-prompt methodology for NotebookLM.

If this is right

NotebookLM can process and structure UXR data into PoV narratives without additional user effort beyond the initial prompts.
The methodology reduces the work intensity typically associated with GenAI use in research by minimizing prompt engineering time.
Success across all stages on multiple papers indicates the approach can scale to various UXR contexts within the defined framework.
GenAI integration in this way supports the shift from traditional usability testing to data-driven UXR approaches.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Teams using this method might integrate NotebookLM directly into their research workflows for quicker iteration on product decisions.
Similar prompting structures could be adapted for other AI tools to support UXR in different organizational settings.
The framework's emphasis on evidence-based PoVs could encourage more consistent application of research insights across product development cycles.

Load-bearing premise

The UXR PoV framework itself offers a reliable and broadly applicable way to convert research data into actionable product strategy.

What would settle it

A new set of UXR papers processed with the same five prompts yields outputs that fail to form coherent evidence-based PoVs or do not align with actual product impact outcomes.

read the original abstract

User Experience Research (UXR) is currently undergoing a transition from traditional usability testing towards design-led and data-driven approaches, yet it faces an identity crisis due to a lack of methodological grounding in UXR and time-intensive methodologies which often lag behind product decision cycles. To address this, the UXR Point of View (PoV) framework formalises the UXR process by transitioning from raw data collection to forming an evidence-based PoV which drives strategic product impact. Furthermore, the use of GenAI in UXR has been investigated, but researchers often face increased work intensity when using GenAI, attributed to time spent on prompt engineering, data cleaning, and verification of AI outputs. This paper proposes and evaluates a formalised methodology for leveraging GenAI, specifically Google's NotebookLM, to augment the UXR PoV process. The methodology consists of five prompts across four stages: (1) leveraging the framework, (2) establishing roadmaps, (3) applying best-practices, and (4) crafting PoV narratives; and was tested on eleven UXR papers. Results showed that by using the proposed methodology, NotebookLM successfully leveraged the UXR PoV framework across all stages of PoV creation. These findings demonstrate that NotebookLM can serve as an effective collaborative partner in UXR, so long as it is provided with sufficient context and specific prompting.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper gives a specific five-prompt sequence for NotebookLM tied to their UXR PoV framework and tests it on eleven papers, but supplies no metrics, baselines, or controls to show the prompts actually drove the outcome.

read the letter

The core takeaway is that this work supplies a ready-made prompting recipe for NotebookLM to follow the authors' UXR Point of View framework across four stages, and reports that the tool handled eleven test papers without issue. That concrete sequence is the main new piece; prior work on AI-assisted synthesis exists, but the explicit tie to this particular framework and the staged prompts appear original here.

What the paper does reasonably is lay out the prompts in enough detail that a practitioner could copy them and try the workflow on their own data. It also flags real pain points in current UXR practice, such as time spent on prompt engineering and output verification, which the structured approach aims to reduce.

The soft spot is the evaluation. The claim that NotebookLM "successfully leveraged" the framework rests on an assertion with no rubric, no inter-rater checks, no comparison to generic prompting or unaided work, and no description of how the eleven papers were selected. The test corpus is existing papers rather than raw user-study transcripts, so it is hard to know whether the results reflect the proposed method, NotebookLM's built-in behavior, or author selection. That gap makes it difficult to judge whether the methodology generalizes or simply worked on the chosen examples.

This is the sort of practical case study that UXR teams experimenting with NotebookLM might find useful as a starting point. Readers looking for rigorous evidence on whether structured prompting improves research synthesis will come away wanting more. The paper is coherent on its own terms and shows clear thinking about the workflow, so it deserves a serious referee who can push for proper controls and metrics rather than a desk reject.

Referee Report

2 major / 2 minor

Summary. The paper introduces the UXR Point of View (PoV) framework to structure the process from raw data collection to evidence-based strategic product insights. It proposes a five-prompt methodology spanning four stages for using Google's NotebookLM to apply this framework and reports that, when tested on eleven UXR papers, NotebookLM successfully leveraged the framework across all stages provided sufficient context and specific prompting, positioning the tool as an effective collaborative partner in UXR.

Significance. If the methodology can be shown to produce reliable outputs through controlled evaluation, the work would offer a concrete, replicable template for reducing prompt-engineering overhead and time lags in UXR, directly addressing the methodological and temporal challenges described in the introduction. The explicit staging of prompts also supplies a starting point for comparative studies with other GenAI systems.

major comments (2)

[Results] Results section (description of the eleven-paper test): the central claim that NotebookLM 'successfully leveraged the UXR PoV framework across all stages' is presented without any definition of success criteria, scoring rubric, quantitative metrics, inter-rater reliability checks, or baseline condition (e.g., generic prompting or unaided human application). This absence makes it impossible to attribute observed performance to the five-prompt methodology rather than to paper selection, NotebookLM's training data, or author interpretation.
[Methodology] Methodology and Evaluation sections: the test corpus consists exclusively of eleven existing published UXR papers rather than raw user-study transcripts or primary data. No selection criteria, sampling frame, or justification for this choice are supplied, raising the possibility that the reported success reflects properties of the chosen papers rather than general applicability of the prompts to typical UXR workflows.

minor comments (2)

[Abstract] The abstract and introduction would benefit from a brief enumeration of the five prompts or at least their high-level structure so readers can assess replicability without reading the full methods.
[Introduction] Notation for the four stages is introduced only in the abstract; a numbered list or table in the main text would improve clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. We address each major comment below and describe the revisions we will make to strengthen the manuscript.

read point-by-point responses

Referee: [Results] Results section (description of the eleven-paper test): the central claim that NotebookLM 'successfully leveraged the UXR PoV framework across all stages' is presented without any definition of success criteria, scoring rubric, quantitative metrics, inter-rater reliability checks, or baseline condition (e.g., generic prompting or unaided human application). This absence makes it impossible to attribute observed performance to the five-prompt methodology rather than to paper selection, NotebookLM's training data, or author interpretation.

Authors: We agree that explicit success criteria are needed. The evaluation is a qualitative case study. In revision we will add a dedicated 'Success Criteria' subsection defining success as observable alignment of NotebookLM outputs with each of the four UXR PoV framework stages, illustrated by representative excerpts from all eleven papers. We will also explicitly state the limitations of the current design, including the lack of quantitative metrics, inter-rater checks, and baselines, and position the work as a feasibility demonstration rather than a controlled comparison. revision: yes
Referee: [Methodology] Methodology and Evaluation sections: the test corpus consists exclusively of eleven existing published UXR papers rather than raw user-study transcripts or primary data. No selection criteria, sampling frame, or justification for this choice are supplied, raising the possibility that the reported success reflects properties of the chosen papers rather than general applicability of the prompts to typical UXR workflows.

Authors: We accept that justification for the corpus must be supplied. The eleven papers were selected for public availability to ensure full reproducibility. We will revise the Methodology section to state the selection criteria (peer-reviewed UXR papers from the last five years, spanning qualitative and quantitative approaches) and sampling rationale. We will also add a Limitations paragraph noting that future studies should evaluate the prompts on raw transcripts and primary data. revision: yes

Circularity Check

0 steps flagged

No circularity: case study evaluation does not reduce to self-definition or fitted inputs

full rationale

The paper proposes the UXR PoV framework and a five-prompt methodology for NotebookLM, then evaluates the combination on eleven external papers by reporting that the tool 'successfully leveraged the UXR PoV framework across all stages.' No mathematical derivations, equations, parameter fitting, or load-bearing self-citations appear in the provided text. The central claim is an empirical observation about prompting outcomes rather than a result that is definitionally equivalent to the inputs or forced by prior author work. The evaluation is therefore self-contained against the described test corpus and does not match any enumerated circularity pattern.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The central claim rests on the untested validity of the newly introduced UXR PoV framework as a general solution to UXR's identity crisis and on the assumption that NotebookLM outputs remain reliable when the stated prompting structure is followed.

axioms (1)

domain assumption The UXR PoV framework provides a valid and complete formalization of the process from raw data to evidence-based strategic impact.
The methodology is built directly on this framework; if the framework is incomplete or biased, the prompting stages inherit those limitations.

invented entities (1)

UXR PoV framework no independent evidence
purpose: To formalize the transition from data collection to an evidence-based point of view that drives product decisions.
Introduced in the paper to address the stated identity crisis and time lag in UXR; no independent evidence of its effectiveness is supplied beyond the NotebookLM case study.

pith-pipeline@v0.9.1-grok · 5780 in / 1414 out tokens · 25613 ms · 2026-06-28T21:21:11.101943+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

15 extracted references

[1]

UX Research is Dead. Long Live UX Research,

S. Giff and H. Dogan, “UX Research is Dead. Long Live UX Research,” 2016

2016
[2]

The Past, present, and future of UX empirical research,

J. Robinson, C. Lanius and R. Weber, “The Past, present, and future of UX empirical research,” Communication Design Quarterly Review, vol. 5, no. 3, pp. 10-23, 2018

2018
[3]

User Experience Research: Point of View Playbook,

H. Dogan, S. Giff and R. Barsoum, “User Experience Research: Point of View Playbook,” Chi EA '24: Extended Abstracts of the CHI Conference on Human Factors in Computing Systes, pp. 1-7, 2024

2024
[4]

Toward a Theory of Situation Awareness in Dynamic Systems,

M. Endsley, “Toward a Theory of Situation Awareness in Dynamic Systems,” Human Factors: The Journal of the Human Factors and Ergonomics Society, vol. 37, pp. 32- 64, 1995

1995
[5]

Defining a UX Research Point of View,

H. Dogan and R. G. S. D. A. C. E. Barsoum, “Defining a UX Research Point of View,” Conference on Human Factors in Computing Systems Proceedings, 2025

2025
[6]

The "Augmented

T. A. Reyes Ponce de León, “The "Augmented" Researcher: UX Researchers' experiences with incorporating genAI into their work,” [Preprint], 2025

2025
[7]

Using generative ai in developing user experience points of view,

A. Sanandaji and R. Stegbauer, “Using generative ai in developing user experience points of view,” 2025

2025
[8]

NotebookLM Review 2025: AI Tool for Researchers,

I. Shabanov, “NotebookLM Review 2025: AI Tool for Researchers,” The Effortless Academic, 15 December 2025. [Online]. Available: https://effortlessacademic.com/notebook-lm-googles-newest-academic-ai-tool/. [Accessed January 2026]

2025
[9]

Supporting Human-AI Teams: Transparency, explainability, and situation awareness,

M. Endsley, “Supporting Human-AI Teams: Transparency, explainability, and situation awareness,” Computers in Human Behaviour, vol. 140, 2023

2023
[10]

Thinking Smarter, not Harder? Google NotebookLM's Misalignment Problem in Education,

C. Albrecht-Crane, “Thinking Smarter, not Harder? Google NotebookLM's Misalignment Problem in Education,” SIGDOC '25: Proceedings of the 43rd ACM International Conference on Design of Communication, pp. 121-127, 2025

2025
[11]

From 600 Tools to 1 Console: A UX- Driven Transformation,

M. K. Smith, J. Meijer-Irons and A. Millar, “From 600 Tools to 1 Console: A UX- Driven Transformation,” ACM, 2025

2025
[12]

Generative AI in User Experience Design and Research: How do UX Practitioners, Teams, and Companies Use GenAI in Industry?,

M. Tkaffoli, S. Li and V. Mäkelä, “Generative AI in User Experience Design and Research: How do UX Practitioners, Teams, and Companies Use GenAI in Industry?,” DIS '24 Proceedings of the 2024 ACM Designing Interactive Systems Conference, pp. 1579-1593, 2024

2024
[13]

Non-Deterministic AI and the Emergence of Gen AI: A New Frontier,

G. H. &. D. I. Foundation, “Non-Deterministic AI and the Emergence of Gen AI: A New Frontier,” Global Health & Digital Innovation Foundation, 18 August 2025. [Online]. Available: https://ghdif.org/news-%26-blog/f/non-deterministic-ai-and-the- emergence-of-gen-ai-a-new-frontier. [Accessed January 2026]

2025
[14]

Developing a UXR Point of View for Neuroinclusive Emotion Regulation with Generative AI,

M. Acka, M. Giff, D. Cetinkaya, H. Dogan and S. Giff, “Developing a UXR Point of View for Neuroinclusive Emotion Regulation with Generative AI,” unpublished
[15]

Exploring the Impact of Generative Artificial Intelligence on the Design Process: Opportunities, Challenges, and Insights,

Y.-r. Lai, H.-J. Chen and C.-H. Yang, “Exploring the Impact of Generative Artificial Intelligence on the Design Process: Opportunities, Challenges, and Insights,” in Artificial Intelligence, Social Computing and Wearable Technologies, Waldemar Karwowski and Tareq Ahram , 2023. Appendix A “From 600 Tools to 1 Console: A UX-Driven Transformation” Outputs Ca...

2023

[1] [1]

UX Research is Dead. Long Live UX Research,

S. Giff and H. Dogan, “UX Research is Dead. Long Live UX Research,” 2016

2016

[2] [2]

The Past, present, and future of UX empirical research,

J. Robinson, C. Lanius and R. Weber, “The Past, present, and future of UX empirical research,” Communication Design Quarterly Review, vol. 5, no. 3, pp. 10-23, 2018

2018

[3] [3]

User Experience Research: Point of View Playbook,

H. Dogan, S. Giff and R. Barsoum, “User Experience Research: Point of View Playbook,” Chi EA '24: Extended Abstracts of the CHI Conference on Human Factors in Computing Systes, pp. 1-7, 2024

2024

[4] [4]

Toward a Theory of Situation Awareness in Dynamic Systems,

M. Endsley, “Toward a Theory of Situation Awareness in Dynamic Systems,” Human Factors: The Journal of the Human Factors and Ergonomics Society, vol. 37, pp. 32- 64, 1995

1995

[5] [5]

Defining a UX Research Point of View,

H. Dogan and R. G. S. D. A. C. E. Barsoum, “Defining a UX Research Point of View,” Conference on Human Factors in Computing Systems Proceedings, 2025

2025

[6] [6]

The "Augmented

T. A. Reyes Ponce de León, “The "Augmented" Researcher: UX Researchers' experiences with incorporating genAI into their work,” [Preprint], 2025

2025

[7] [7]

Using generative ai in developing user experience points of view,

A. Sanandaji and R. Stegbauer, “Using generative ai in developing user experience points of view,” 2025

2025

[8] [8]

NotebookLM Review 2025: AI Tool for Researchers,

I. Shabanov, “NotebookLM Review 2025: AI Tool for Researchers,” The Effortless Academic, 15 December 2025. [Online]. Available: https://effortlessacademic.com/notebook-lm-googles-newest-academic-ai-tool/. [Accessed January 2026]

2025

[9] [9]

Supporting Human-AI Teams: Transparency, explainability, and situation awareness,

M. Endsley, “Supporting Human-AI Teams: Transparency, explainability, and situation awareness,” Computers in Human Behaviour, vol. 140, 2023

2023

[10] [10]

Thinking Smarter, not Harder? Google NotebookLM's Misalignment Problem in Education,

C. Albrecht-Crane, “Thinking Smarter, not Harder? Google NotebookLM's Misalignment Problem in Education,” SIGDOC '25: Proceedings of the 43rd ACM International Conference on Design of Communication, pp. 121-127, 2025

2025

[11] [11]

From 600 Tools to 1 Console: A UX- Driven Transformation,

M. K. Smith, J. Meijer-Irons and A. Millar, “From 600 Tools to 1 Console: A UX- Driven Transformation,” ACM, 2025

2025

[12] [12]

Generative AI in User Experience Design and Research: How do UX Practitioners, Teams, and Companies Use GenAI in Industry?,

M. Tkaffoli, S. Li and V. Mäkelä, “Generative AI in User Experience Design and Research: How do UX Practitioners, Teams, and Companies Use GenAI in Industry?,” DIS '24 Proceedings of the 2024 ACM Designing Interactive Systems Conference, pp. 1579-1593, 2024

2024

[13] [13]

Non-Deterministic AI and the Emergence of Gen AI: A New Frontier,

G. H. &. D. I. Foundation, “Non-Deterministic AI and the Emergence of Gen AI: A New Frontier,” Global Health & Digital Innovation Foundation, 18 August 2025. [Online]. Available: https://ghdif.org/news-%26-blog/f/non-deterministic-ai-and-the- emergence-of-gen-ai-a-new-frontier. [Accessed January 2026]

2025

[14] [14]

Developing a UXR Point of View for Neuroinclusive Emotion Regulation with Generative AI,

M. Acka, M. Giff, D. Cetinkaya, H. Dogan and S. Giff, “Developing a UXR Point of View for Neuroinclusive Emotion Regulation with Generative AI,” unpublished

[15] [15]

Exploring the Impact of Generative Artificial Intelligence on the Design Process: Opportunities, Challenges, and Insights,

Y.-r. Lai, H.-J. Chen and C.-H. Yang, “Exploring the Impact of Generative Artificial Intelligence on the Design Process: Opportunities, Challenges, and Insights,” in Artificial Intelligence, Social Computing and Wearable Technologies, Waldemar Karwowski and Tareq Ahram , 2023. Appendix A “From 600 Tools to 1 Console: A UX-Driven Transformation” Outputs Ca...

2023