pith. sign in

arxiv: 2604.09583 · v1 · submitted 2026-02-26 · 💻 cs.HC

Trace-Aware Workflows for Co-Creating Branded Content with Generative AI

Taehyun Yang , Eunhye Kim , Zhongzheng Xu , Fumeng Yang This is my paper

Pith reviewed 2026-05-15 19:32 UTC · model grok-4.3

classification 💻 cs.HC
keywords small business ownersgenerative AIbranded contentiterative workflowstraceboardhuman-AI co-creationsocial media content creation
0
0 comments X

The pith

A traceboard that records branching AI image iterations helps small business owners articulate brand feel and steer refinements more effectively.

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

The paper reports a preliminary study with 12 small business owners who create their own social media content using generative AI. It identifies three recurring challenges: turning an abstract brand feel into usable prompts, revisiting and comparing earlier image generations, and interpreting differences between iterations to guide the next step. From these observations the authors built a prototype that scaffolds brand articulation, enables feedback-informed exploration, and maintains a traceboard of how image versions branch and change. The central idea is that preserving traces of the iterative process itself can reduce the cognitive load of keeping track, making sense of changes, and refining content without professional designers.

Core claim

Small business owners face three specific challenges when using generative AI for branded social media content: translating brand feel into effective prompts, revisiting and comparing prior generations, and making sense of changes between iterations to steer refinement. A prototype that scaffolds brand articulation, supports feedback-informed exploration, and maintains a traceboard of branching image iterations can help users keep track of explorations, interpret differences, and produce more consistent on-brand results.

What carries the argument

The traceboard, a maintained visual record of branching image iterations that lets users revisit prior outputs, compare versions, and understand incremental changes to guide further refinements.

If this is right

  • Users can more readily maintain visual consistency across multiple social media posts.
  • Comparison of alternative prompt variations becomes less cognitively demanding.
  • Feedback from one generation can be applied more precisely to steer the next.
  • The overall time and frustration of iterative content creation decreases.

Where Pith is reading between the lines

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

  • Trace mechanisms of this kind could be adapted to other generative tasks such as caption writing or short video editing.
  • Embedding explicit brand guidelines into the traceboard interface might further reduce the prompt-translation barrier.
  • Longer-term deployment studies could test whether the approach reduces reliance on external marketing help.

Load-bearing premise

The three challenges reported by the 12 small business owners in the questionnaire are representative enough to guide the design of a prototype that will meaningfully improve workflows for the broader population of small business owners.

What would settle it

A controlled comparison study in which small business owners using the prototype show no measurable improvement in on-brand content quality, iteration speed, or reported workflow satisfaction relative to standard generative AI tools.

Figures

Figures reproduced from arXiv: 2604.09583 by Eunhye Kim, Fumeng Yang, Taehyun Yang, Zhongzheng Xu.

Figure 1
Figure 1. Figure 1: Overview of our prototype for branded content creation: [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
read the original abstract

Generative AI tools have lowered barriers to producing branded social media images and captions, yet small-business owners (SBOs) still struggle to create on-brand posts without access to professional designers or marketing consultants. Although these tools enable fast image generation from text prompts, aligning outputs with a brand's intended look and feel remains a demanding, iterative task. In this position paper, we explore how SBOs navigate iterative content creation and how AI-assisted systems can support SBOs' content creation workflow. We conducted a preliminary study with 12 SBOs who independently manage their businesses and social media presence, using a questionnaire to collect their branding practices, content workflows, and use of generative AI alongside conventional design tools. We identified three recurring challenges: (1) translating brand "feel" into effective prompts, (2) difficulty revisiting and comparing prior image generations, and (3) difficulty making sense of changes between iterations to steer refinement. Based on these findings, we present a prototype that scaffolds brand articulation, supports feedback-informed exploration, and maintains a traceboard of branching image iterations. Our work illustrates how traces of the iterative process can serve as workflow support that helps SBOs keep track of explorations, make sense of changes, and refine content.

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.

Referee Report

2 major / 1 minor

Summary. The manuscript reports a preliminary questionnaire study with 12 small business owners (SBOs) that identifies three recurring challenges when using generative AI for branded social-media content: (1) translating brand 'feel' into effective prompts, (2) revisiting and comparing prior image generations, and (3) making sense of changes between iterations. On the basis of these findings the authors present a conceptual prototype that scaffolds brand articulation, supports feedback-informed exploration, and maintains a traceboard of branching image iterations, arguing that explicit traces of the iterative process can help SBOs keep track of explorations and steer refinement.

Significance. If the proposed trace-aware features were shown to mitigate the reported challenges, the work would contribute to HCI by illustrating how process traces can scaffold iterative creative workflows for non-expert users. The emphasis on small-business contexts and the explicit mapping from user-reported pain points to interface mechanisms could inform future tool design. At present, however, the absence of any evaluation data means the significance remains prospective rather than demonstrated.

major comments (2)
  1. [preliminary study] Preliminary study section: the central design requirements rest on questionnaire responses from only 12 participants with no reported recruitment details, response analysis, or comparison to existing literature on SBO branding practices; this small, unvalidated sample is load-bearing for the claim that the three challenges are general and sufficient to motivate the prototype features.
  2. [prototype] Prototype description: the manuscript asserts that the traceboard, brand-articulation scaffolding, and feedback-informed exploration address the three identified challenges, yet no usability testing, think-aloud study, or even informal walkthrough with participants is reported; the mapping from challenges to features therefore remains an untested design hypothesis rather than an empirically supported outcome.
minor comments (1)
  1. [abstract] Abstract and introduction: explicitly label the work as a position paper and add a sentence noting the preliminary status of the study and the lack of prototype evaluation to manage reader expectations.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive comments. We address the major concerns point by point below, clarifying the scope of this position paper while incorporating revisions where appropriate.

read point-by-point responses

  1. Referee: Preliminary study section: the central design requirements rest on questionnaire responses from only 12 participants with no reported recruitment details, response analysis, or comparison to existing literature on SBO branding practices; this small, unvalidated sample is load-bearing for the claim that the three challenges are general and sufficient to motivate the prototype features.

    Authors: We agree the sample is small and the study is preliminary, as stated in the manuscript. We will revise to add recruitment details (via online small-business forums and local networks), a brief summary of thematic analysis of responses, and a short comparison to existing literature on SBO digital branding practices. As a position paper, however, the goal is to surface recurring challenges to motivate conceptual design directions rather than to establish generalizability; the three challenges appeared consistently in the data and provide a reasonable basis for the proposed prototype. revision: partial

  2. Referee: Prototype description: the manuscript asserts that the traceboard, brand-articulation scaffolding, and feedback-informed exploration address the three identified challenges, yet no usability testing, think-aloud study, or even informal walkthrough with participants is reported; the mapping from challenges to features therefore remains an untested design hypothesis rather than an empirically supported outcome.

    Authors: We acknowledge that no usability evaluation of the prototype is reported. This is a position paper that presents a conceptual prototype as an illustration of how trace-aware mechanisms could address the identified challenges; the mapping is offered as a design rationale, not as an empirically validated result. We will revise the text to state this scope explicitly and note that empirical validation remains future work, consistent with the prospective contribution described in the referee summary. revision: yes

Circularity Check

0 steps flagged

No circularity; claims rest on direct questionnaire findings and design proposal

full rationale

The paper conducts a questionnaire study with 12 SBOs, identifies three challenges from responses, and proposes a prototype as a design response. No equations, fitted parameters, predictions, or self-citations are used in a load-bearing way. The central claim (prototype addresses challenges) is a hypothesis derived from user-reported data rather than reducing to its inputs by construction. This is a standard qualitative design paper with no self-referential loops.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The central claim rests on standard HCI assumptions about the value of user-centered design and qualitative identification of pain points; no free parameters or invented physical entities are introduced.

axioms (1)
  • domain assumption A small sample of 12 small business owners can surface recurring, generalizable workflow challenges for the broader population.
    Invoked when moving from questionnaire responses to prototype requirements.
invented entities (1)
  • traceboard no independent evidence
    purpose: Maintains a visual record of branching image iterations to support comparison and refinement.
    Newly proposed interface component whose utility is asserted but not yet measured.

pith-pipeline@v0.9.0 · 5530 in / 1355 out tokens · 67960 ms · 2026-05-15T19:32:49.592093+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

14 extracted references · 14 canonical work pages

  1. [1]

    Philip Alford and Stephen John Page. 2018. Marketing technology for adoption by small business. InSocial Media and Interactive Communications. Routledge, 99–113. doi:10.1080/02642069.2015.1062884

    work page doi:10.1080/02642069.2015.1062884 2018

  2. [2]

    2008.The New Strategic Brand Management: Creating and Sustaining Brand Equity Long Term

    Jean-Noël Kapferer. 2008.The New Strategic Brand Management: Creating and Sustaining Brand Equity Long Term. Kogan Page, London, UK

    work page 2008

  3. [3]

    Nimisha Karnatak, Adrien Baranes, Rob Marchant, Triona Butler, and Kristen Olson. 2025. ACAI for SBOs: AI Co-creation for Advertising and Inspiration for Small Business Owners. arXiv. arXiv:2503.06729 doi:10.48550/arXiv.2503.06729

    work page doi:10.48550/arxiv.2503.06729 2025

  4. [4]

    Sascha Kraus, Johanna Gast, Moritz Schleich, Paul Jones, and Michael Ritter. 2019. Content is king: How SMEs create content for social media marketing under limited resources.Journal of Macromarketing39, 4 (2019), 415–430. doi:10.1177/0276146719882746

    work page doi:10.1177/0276146719882746 2019

  5. [5]

    Jie Li, Hancheng Cao, Laura Lin, Youyang Hou, Ruihao Zhu, and Abdallah El Ali. 2024. User experience design professionals’ perceptions of generative artificial intelligence. InProceedings of the 2024 CHI conference on human factors in computing systems. 1–18. doi:10.1145/3613904.3642114

    work page doi:10.1145/3613904.3642114 2024

  6. [6]

    Xuye Liu, Annie Sun, Pengcheng An, Tengfei Ma, and Jian Zhao. 2025. Influencer: Empowering Everyday Users in Creating Promotional Posts via AI-infused Exploration and Customization. InProceedings of the 2025 CHI Conference on Human Factors in Computing Systems. Association for Computing Machinery, New York, NY, USA, 1–19. doi:10.1145/3706598.3713309

    work page doi:10.1145/3706598.3713309 2025

  7. [7]

    Midjourney. [n. d.]. Midjourney. https://www.midjourney.com/

    work page

  8. [8]

    OpenAI. [n. d.]. DALL·E. https://openai.com/dall-e

    work page

  9. [9]

    Heini Maarit Taiminen and Heikki Karjaluoto. 2015. The usage of digital marketing channels in SMEs.Journal of Small Business and Enterprise Development22, 4 (2015), 633–651. doi:10.1108/JSBED-05-2013-0073

    work page doi:10.1108/jsbed-05-2013-0073 2015

  10. [10]

    United Nations. 2025. Micro-, Small and Medium-sized Enterprises Day. https://www.un.org/en/observances/micro-small-medium-businesses-day Accessed: 2026-02-07

    work page 2025

  11. [11]

    Verizon Business. 2025. 2025 State of Small Business Survey: Surge in AI, Cybersecurity and Social Media Demand. https://www.verizon.com/about/ news/2025-state-of-small-business-survey. Accessed: 2026-02-06

    work page 2025

  12. [12]

    Justin D Weisz, Jessica He, Michael Muller, Gabriela Hoefer, Rachel Miles, and Werner Geyer. 2024. Design principles for generative AI applications. InProceedings of the 2024 CHI Conference on Human Factors in Computing Systems. 1–22. doi:10.1145/3613904.3642466

    work page doi:10.1145/3613904.3642466 2024

  13. [13]

    World Bank. 2025. Small and Medium Enterprises (SMEs) Finance. https://www.worldbank.org/en/topic/smefinance Accessed: 2026-02-07

    work page 2025

  14. [14]

    J Diego Zamfirescu-Pereira, Richmond Y Wong, Bjoern Hartmann, and Qian Yang. 2023. Why Johnny can’t prompt: how non-AI experts try (and fail) to design LLM prompts. InProceedings of the 2023 CHI conference on human factors in computing systems. 1–21. doi:10.1145/3544548.3581388 6 Appendices A Participant Demographics Table A.1. Participant demographics in...

    work page doi:10.1145/3544548.3581388 2023