PaperlayerView exact revision on arXiv

Source-linked AI summary

Understanding Game Coaching on Gig Platforms

Hwijoon Lee, Saiph Savage

arXiv:2609.12695v1cs.HCcs.CY

TL;DR

Everyday, non-institutional game coaching is largely unstudied compared with professional team coaching, leaving its practices insufficiently understood. Researchers conducted semi-structured interviews with 20 experienced coaches across 17 competitive games on Fiverr, examining session structure, diagnosis, student relationships, and platform constraints. Coaches converged on situated diagnosis and adaptive feedback while navigating platform and game volatility, continually establishing competitive legitimacy, and favoring AI support for preparation over live coaching.

  • Problem

    Everyday, non-institutional game coaching is largely unstudied compared with professional team coaching, leaving its practices insufficiently understood.

    Source: §1 · pp. 1–2

  • Method

    Researchers conducted semi-structured interviews with 20 experienced coaches across 17 competitive games on Fiverr, examining session structure, diagnosis, student relationships, and platform constraints.

    Source: §1 · p. 2

  • Results

    Coaches converged on situated diagnosis and adaptive feedback while navigating platform and game volatility, continually establishing competitive legitimacy, and favoring AI support for preparation over live coaching.

    Source: §1 · p. 2, §6 · p. 18

  • Takeaways & Limitations

    Computational coaching tools may augment coordination and preparation, while coaching’s relational and diagnostic core remains tied to situated human expertise and gameplay context.

    Source: §1 · p. 2

    The Fiverr-based, English-language sample privileged experienced, visible coaches, leaving non-English contexts, coaches outside established platforms, and those struggling to secure initial orders unexamined.

    Source: §5.5 · p. 17

Abstract

from arXiv · show

Freelance game coaches monetize their gaming expertise by offering personalized instruction to players seeking to improve, working through gig platforms, yet little is known about how they operate. To address this gap, we conducted semi-structured interviews with 20 experienced freelance coaches across 17 competitive games on Fiverr. Despite lacking shared formal training, these coaches converged on similar practices centered on rapport-building, individualized diagnosis, and adaptive feedback. We identify two structural conditions shaping this work: dual precarity, in which coaches navigate both gig platform instability and the lifecycle volatility of live-service games; and earned authority, in which coaches must continually establish legitimacy through visible competitive achievement within the same gaming spaces as their students. These coaches welcomed AI for administrative and analytic support but resisted its use in live interactions where trust, relational engagement, and situated judgment remained central. We discuss implications for Games HCI and the design of computational coaching systems.

1 Introduction

The paper examines freelance game coaching as an understudied form of platform-mediated instruction. Interviews show coaches converging on relational, adaptive practices while navigating platform and game volatility and establishing authority through competitive achievement.

1 Introduction

  • Everyday, non-institutional game coaching remains understudied because prior research has concentrated on professional team settings.Understanding these practices is presented as a prerequisite for designing tools that fit how competitive games are taught and learned.Source: §1 · p. 1
  • The study uses semi-structured interviews with 20 experienced Fiverr coaches across 17 competitive games to examine session structure, diagnosis, relationships, and platform constraints.Its analysis characterizes coaching as shaped by game mechanics, metagame cycles, and changing player needs.Source: §1 · p. 2
  • Without shared formal training, coaches converged on rapport-building, individualized diagnosis, and adaptive feedback grounded in players’ gameplay, habits, emotional states, and learning needs.They created value through situated interpretation rather than mainly repeating information available in guides or videos.Source: §1 · p. 2
  • Coaches face dual precarity from gig-platform instability and game lifecycles, while continually establishing earned authority through visible achievement in students’ competitive spaces.They welcomed AI for administrative and preparatory work but resisted replacing live coaching or gameplay diagnosis, which they viewed as requiring situated judgment and trust.Source: §1 · p. 2
  • The paper contributes concepts of dual precarity and earned authority, and argues that coaching tools should augment coordination and preparation while respecting situated human expertise and interaction.The relational and diagnostic core remains tied to the lived context of gameplay.Source: §1 · p. 2

2 Related Work

Related work describes platform labor as governed by algorithmic visibility and ongoing reputational work, while gigified education depends on relational collaboration. Research on game coaching and computational learning has not yet examined freelance coaching practice directly.

2.1 Platform Labor and Reputation

  • Platform labor is precarious and institutionally unaffiliated workers must sustain credibility, while algorithmic ranking and reputation shape visibility and continued access to work.Freelancers cultivate profiles and reviews because platform-distributed opportunity makes reputation an ongoing resource.Source: §2.1 · p. 3

2.2 Gigified Knowledge Services

  • Gigified knowledge services require specialized skill and sustained collaboration, but ratings capture transactional satisfaction rather than relational and diagnostic quality.Research on gig-mediated tutoring highlights rapport and emotional labor; AI tutoring and augmentation studies have focused on traditional educational settings.Source: §2.2 · pp. 3–4
  • Freelance game coaching is increasingly visible on Fiverr, with over 2,500 active coaching-related listings reported as of 2026, despite limited academic study of the practice.The passage places coaching within broader platform-mediated game labor and competitive ecosystems.Source: §2.2 · p. 4, §2.3 · p. 4

2.3 Competitive Game Ecosystems and Coaching

  • Research has studied professional and academy coaching and computational player-learning tools, but has not empirically examined how independent marketplace coaches practice.Prior work identifies tactical change and coaching-toolkit challenges in organized settings, while AI-supported learning research focuses on individual players or elite teams and academies.Source: §2.3 · pp. 4–5

3 Methods

The study uses purposive recruitment and semi-structured interviews to investigate experienced Fiverr coaches across competitive games. Reflexive thematic analysis combines inductive coding with team discussion, while the lead author’s gaming familiarity informs interpretation.

3 Methods

  • The researchers conducted IRB-approved semi-structured interviews with 20 coaches to surface practices, challenges, and technology perspectives in an understudied context.The team included members familiar with competitive gaming and gig platforms, and no researcher had a prior relationship with participants.Source: §3 · p. 5

3.1 Participants

  • The study recruited 20 experienced coaches actively offering services on Fiverr across 17 games; purposive selection targeted accumulated experience, requiring at least 30 verified client reviews.Sample composition reflected available freelance coaches on Fiverr rather than purposive sampling decisions; review count was the platform’s only publicly exposed practice-volume signal.Source: §3.1 · pp. 5–6

3.2 Interview Procedure

  • Interviews took place by video over Zoom or Discord from November 2025 to February 2026, lasted about 35 minutes, and covered backgrounds, coaching practices, challenges, and views on AI.Sessions were conducted in English, recorded, transcribed, and checked against recordings by the first author.Source: §3.2 · pp. 6–7

3.3 Data Analysis

  • The first author analyzed transcripts through two rounds of inductive coding and developed themes with regular team discussion and refinement.The first author had played each represented game for at least 30 hours; analysis focused on recurring platform-mediated patterns across structurally different titles.Source: §3.3 · p. 7

4 Results

The section introduces the main themes that emerged from the study.

4 Results

  • The section presents the main themes that emerged from the study.Source: §4 · p. 7

4.1 Entering and Sustaining the Game Coaching Gig

Coaches entered gig work through competitive expertise but learned service delivery informally; sustaining it meant adapting to changing games and platform-mediated demand. Their work remained vulnerable to game popularity, platform visibility, and the limits of transferring expertise across titles.

4.1.1 How Coaches Enter Freelance Game Coaching.

  • Many coaches began after reaching high rank and seeing market demand, while others turned to coaching when outside constraints made continued competition difficult.Several learned teaching and service delivery on the job; some waited months for first clients, and one waited nearly two years for consistent orders.Source: §4.1.1 · pp. 7–8

4.1.2 How Game Lifecycles Shape Freelance Game Coaching.

  • Coaching demand depended on game popularity, and expertise transferred only partly across titles, even within a genre.Updates could also make existing resources unreliable, requiring coaches to keep their knowledge current.Source: §4.1.2 · p. 8

4.1.3 Adapting to the Metagame Cycle.

  • Metagame change varied by title: chess resources remained comparatively durable, while frequent patches could invalidate curricula and older coaching materials.Hearthstone Battlegrounds and TFT coaches described adapting to every patch or set; a Dota 2 coach found resources from one or two years earlier unreliable.Source: §4.1.3 · p. 8

4.1.4 Economic Realities and Platform Dependency on Game Coaching.

  • Platform visibility and game demand shaped income, while opaque Fiverr rankings could disrupt even established coaches’ work.Some diversified by moving recurring clients off-platform or developing other client-acquisition channels; others combined coaching with separate businesses.Source: §4.1.4 · pp. 8–9

4.2 Session Practices and Coaching Approaches

Without shared training, coaches developed similar one-on-one practices centered on personalization, rapport, and adaptive communication. They used lightweight tools and relational follow-up to sustain engagement beyond sessions.

4.2.1 How Freelance Game Coaches Structure Coaching Sessions.

  • Coaches converged on a session flow of rapport building, assessing the student, observing gameplay, diagnosing weaknesses, giving targeted feedback, and closing with a summary.They used either live observation or recorded replays.Source: §4.2.1 · p. 9

4.2.2 Personalized and Relationship-Centered Pedagogy.

  • Coaches tailored instruction to students’ needs and adjusted tone, communication, and pacing; they saw this personalization and relationship-building as paid coaching’s value over free online content.These practices developed through experience and ongoing interaction rather than formal training.Source: §4.2.2 · pp. 9–10
  • Coaches relied on informal tools such as Discord, replay systems, and personal notes, while follow-up messages and communities helped maintain student engagement beyond sessions.Some coaches noted that follow-up approaches did not always scale effectively.Source: §4.2.3 · p. 10

4.3 Authority and Expectation Conflicts in Freelance Game Coaching

Coaching authority was fragile because students sometimes resisted feedback and game coaching lacked formal credentials. Coaches had to establish legitimacy while negotiating expectations shaped by the devaluation of games as work.

4.3.1 Student Resistance and Fragile Authority in Freelance Game Coaching.

  • Students sometimes equated long play experience with expertise and resisted feedback, so coaches continually established legitimacy through visible rank, competitive history, and platform ratings.Students often requested proof even when credentials were already displayed on coach profiles.Source: §4.3.1 · p. 10, §4.3 · p. 10
  • Coaches said games’ limited social and professional legitimacy devalued coaching, including when students compared its cost unfavorably with spending on game cosmetics.Source: §4.3.2 · p. 11

4.3.2 Devalued Labor and Misaligned Expectations.

  • Some students expected rapid improvement after one session, and coaches said negative reviews or refund demands could damage platform visibility and reputation.Source: §4.3.2 · p. 11

4.4 Perspectives on AI in Game Coaching

Coaches had encountered AI in game-learning tools but drew a boundary between automated support and human coaching. They were skeptical of AI in live instruction, while welcoming behind-the-scenes help that preserved the coach’s personal presence.

4.4.1 Encounters with AI Coaching Tools in Practice.

  • Coaches encountered AI in game-specific training tools, chess engines, and chatbots, but chatbot advice could not keep pace with rapidly changing game states.A Dota 2 content creator reportedly selected only one or two chatbot item recommendations before returning to in-game judgment.Source: §4.4.1 · p. 11

4.4.2 Skepticism Toward Automated Coaching.

  • Coaches doubted AI could keep strategies current or provide individualized diagnosis in many games, though some saw potential for chess or beginner support.They also said AI could not replicate human warmth or motivational support.Source: §4.4.2 · pp. 11–12
  • Coaches welcomed AI for scheduling, student intake, summaries, and recordkeeping, but resisted client-facing automation that could undermine trust or displace their presence.One coach disabled an AI response tool after it replied to clients without notifying him.Source: §4.4.3 · p. 12

5 Discussion

Freelance game coaching couples platform-mediated labor with evolving game ecosystems and relational instruction. Coaches manage instability across both platforms and game lifecycles while maintaining expertise and adapting their practices.

5 Discussion

  • Coaching combines platform labor, game lifecycles, and relational instruction in a configuration not fully anticipated by platform-labor or game-coaching research.Coaches face instability in platform visibility and game populations while establishing authority among students in the same competitive systems.Source: §5 · p. 13

5.1 Dual Precarity

  • Dual precarity spans gig-platform instability and live-service game volatility, including updates and declining player populations that can rapidly reshape coaching demand.Stability on one axis cannot compensate for instability on the other; coaches must manage both without formal esports protections.Source: §5.1 · p. 13
  • Expertise transfer across games is limited: coaches must relearn title-specific mechanics and norms, tying their livelihoods to particular game ecosystems.As a game's player base contracts, the pool of potential students also contracts; freelance coaches experience these shifts more directly than salaried esports coaches.Source: §5.1 · p. 13, §5.1.2 · p. 14

5.1.1 Adapting to the Metagame Cycle.

  • Game updates can invalidate coaching knowledge, requiring coaches to maintain, revise, and sometimes relearn title-specific strategies and materials.The speed of knowledge expiration varies by title, making stable curricula and expertise assumptions unsuitable for many games.Source: §5.1.1 · pp. 13–14

5.1.2 Coach Strategies for Managing Dual Precarity.

  • Coaches reduce reliance on platform algorithms through off-platform clients, external content, diversified income, follow-up support, and student communities.These practices extend client relationships and continuity but do not eliminate structural pressures.Source: §5.1.2 · p. 14
  • Coaching tools should strengthen coordination, community-building, and cross-title adaptation rather than displace coaches' existing practices.Responses to instability negotiate pressures across platform, relational, and gameplay-expertise layers.Source: §5.1.2 · p. 14

5.2 Relational Gameplay Diagnosis in Freelance Coaching

Freelance coaches diagnose gameplay individually and calibrate feedback to maintain trust with students who are also paying clients and evaluators. Their authority must be demonstrated within shared competitive spaces rather than granted by credentials.

5.2 Relational Gameplay Diagnosis in Freelance Coaching

  • Coaches ground expertise in individualized diagnosis of gameplay, problems, and each learner's needs, personality, and goals.Because students can disengage, request refunds, or leave reviews, direct criticism risks losing clients and platform visibility.Source: §5.2 · p. 14
  • Coaches soften corrections, ask questions rather than command, and build rapport before critique, adapting tone to students' expectations and emotional responses.Feedback, diagnosis, trust, and legitimacy are intertwined when students are learners, customers, and evaluators.Source: §5.2 · pp. 14–15
  • The single woman coach's accounts suggest gender may shape relational conditions, but one participant cannot support systematic characterization.Prospective clients sometimes approached her based on gender, while parents sometimes preferred a woman coach for children.Source: §5.2 · p. 15

5.3 Earned Authority and the Negotiation of Coaching Expertise in Competitive Gaming

  • Students often ask about rank, achievements, or competitive experience, so coaches establish legitimacy through visible performance in the same games.Students may treat their own gameplay experience as grounds to evaluate or contest coaching advice.Source: §5.3 · p. 15
  • Coaches display ranks, badges, tournament histories, gameplay footage, or team affiliations, but students may demand further proof and contest expertise.Authority remains relational and ongoing, negotiated through interaction, communication, and continued competitive relevance.Source: §5.3 · p. 15
  • Professional-level accomplishments can reduce repeated legitimacy work for a few coaches, but such credentials are rare and formal coaching certification was absent.Most participants were highly ranked non-professional players; the study did not find coaching licenses on their profiles or referenced by them.Source: §5.3 · pp. 15–16

5.4 Design Implications

The design implications favor computational support for coaches' administrative and analytic work while preserving human diagnosis and relational engagement. Systems can also help coaches make authority legible and maintain knowledge as games change; the study's sample bounds these implications.

5.4 Design Implications

  • Coaches welcomed AI for administrative tasks and coaching analytics but resisted replacing human coaches in real-time gameplay and relational interactions.Suggested support includes scheduling, session summaries, progress tracking, replay analysis, and follow-up management; visible client-facing automation drew negative reactions from some students.Source: §5.4 · p. 16
  • Hybrid systems can support benchmarking, longitudinal pattern detection, adaptive lesson planning, and coordination while coaches retain relational and diagnostic work.Human coaches remain responsible for timing criticism, building rapport, responding to emotions, and interpreting gameplay in context.Source: §5.4 · p. 16

5.4.1 Designing Coaching Systems Around Earned Authority.

  • Verified coaching portfolios could aggregate ranks, badges, tournament histories, clips, and reviews to reduce the burden of repeatedly proving legitimacy.Systems could also display coaching outcomes, student-improvement trajectories, and adaptive lesson histories tied to changing metas.Source: §5.4.1 · p. 17
  • Coaching knowledge is dynamic, so tools could track updates, flag outdated strategies and materials, and preserve how advice changes across game versions.Expertise is shaped by patches, shifting metas, and player strategies rather than fully captured in a fixed guide or encoded once.Source: §5.4.2 · p. 17

5.5 Limitations

  • The study centers on experienced, visible Fiverr coaches, mostly serving US and European students in English; less-visible entrants and other language or platform contexts remain unexamined.The sample also cannot represent barriers faced by coaches struggling to secure initial orders or those who moved to alternative channels.Source: §5.5 · p. 17
  • A heavily gender-skewed sample and one woman's account cannot support systematic claims about how gender shapes coaching practice.Recruitment was not stratified by gender, and coach gender was often not visible on Fiverr listings.Source: §5.5 · p. 17

6 Conclusion

The study finds that experienced freelance game coaches converge on situated diagnosis and adaptive feedback, while navigating precarious work and continually establishing authority. Its conclusion favors AI support that preserves coaching's relational and diagnostic work.

6 Conclusion

  • Interviews with 20 experienced coaches across 17 competitive titles found a shared practice of diagnosing situated gameplay and giving adaptive feedback.Coaches worked independently and lacked shared formal training.Source: §6 · p. 18
  • Coaches continually earn authority within students' competitive spaces while navigating instability across platform algorithms and game lifecycles.These conditions shape where AI support can meaningfully enter coaching work.Source: §6 · p. 18
  • Coaches welcomed AI for behind-the-scenes administrative work but resisted its use in diagnostic interactions where trust and relational expertise are built.The conclusion calls for systems that support coaching ecosystems without displacing their relational and diagnostic work.Source: §6 · p. 18
Loading 2609.12695v1…