Generative Artificial Intelligence (GenAI) has entered knowledge-work environments at an unprecedented pace, yet little academic research has examined how it is being adopted within the specific practice of product discovery. This thesis investigates how Product Managers (PMs) are integrating GenAI into their discovery workflows, and explores the perceived value of these tools, the barriers limiting their effective use, and the implications of adoption for the evolving role of the PM. Adopting a pragmatist research philosophy and a convergent parallel mixed-methods design, the study combines a quantitative survey of 82 practicing PMs and Product Owners with a qualitative case study of four leading product companies — Figma, Notion, Miro, and Atlassian — drawn from publicly available documentation. The analysis is framed by three theoretical lenses: the Technology Acceptance Model (TAM) and the Unified Theory of Acceptance and Use of Technology (UTAUT) as adoption frameworks; Daugherty and Wilson's augmentation theory; and the AI Fluency Framework developed by Dakan and Feller, applied here for the first time in the academic literature to interpret professional GenAI adoption behaviour. The findings indicate that GenAI has become a routine instrument of product discovery for most practitioners, valued chiefly for accelerating repetitive cognitive work such as research synthesis, drafting, and coordination. This value is, however, bounded: practitioners trust GenAI markedly less than they find it useful, with hallucination risk, missing organisational context, and governance constraints emerging as the principal barriers to deeper adoption. Trust is therefore shown to be conditional and situational, and human discernment remains indispensable to the discovery process. The study concludes that the trajectory of GenAI in product discovery is one of augmentation rather than replacement, shifting the centre of gravity of the PM role away from manual production and toward judgement, interpretation, validation, and orchestration. As its principal original contribution, the thesis proposes a GenAI Adoption Maturity Model for product teams, positing that organisations advance not by using AI more, but by using it with greater discernment, calibrated trust, and governance maturity. Keywords: Generative AI (GenAI); Product Discovery; Product Management; AI Adoption; Technology Acceptance Model; AI Fluency; Human-AI Collaboration; Augmentation; Maturity Model.
L'Intelligenza Artificiale Generativa (GenAI) ha fatto il suo ingresso negli ambienti di lavoro della conoscenza a un ritmo senza precedenti, eppure la ricerca accademica ha dedicato scarsa attenzione a come essa venga adottata nell'ambito specifico del product discovery. La presente tesi indaga le modalità con cui i Product Manager (PM) stanno integrando la GenAI nei propri flussi di lavoro di discovery, esplorandone il valore percepito, le barriere che ne limitano un uso efficace e le implicazioni dell'adozione sull'evoluzione del ruolo del PM. Adottando una filosofia di ricerca pragmatista e un disegno misto convergente parallelo, lo studio combina un'indagine quantitativa condotta su 82 PM e Product Owner con un'analisi qualitativa di quattro aziende leader nel settore dei prodotti digitali — Figma, Notion, Miro e Atlassian — basata su documentazione pubblica. L'analisi si avvale di tre lenti teoriche: il Technology Acceptance Model (TAM) e la Unified Theory of Acceptance and Use of Technology (UTAUT) come framework di adozione; la teoria dell'augmentation di Daugherty e Wilson; e il Framework sull'AI Fluency sviluppato da Dakan e Feller, applicato qui per la prima volta nella letteratura accademica per interpretare i comportamenti di adozione professionale della GenAI. I risultati indicano che la GenAI è diventata uno strumento ordinario del product discovery per la maggior parte dei professionisti, apprezzata principalmente per la capacità di accelerare attività cognitive ripetitive quali la sintesi della ricerca, la redazione di documenti e il coordinamento. Tale valore risulta tuttavia limitato: i professionisti si fidano della GenAI sensibilmente meno di quanto la ritengano utile, con il rischio di allucinazioni, la mancanza di contesto organizzativo e i vincoli di governance che emergono come principali ostacoli a un'adozione più profonda. La fiducia si dimostra pertanto condizionale e situazionale, e il discernimento umano rimane indispensabile nel processo di discovery. Lo studio conclude che la traiettoria della GenAI nel product discovery è di augmentation piuttosto che di sostituzione, spostando il centro di gravità del ruolo del PM dalla produzione manuale verso il giudizio, l'interpretazione, la validazione e l'orchestrazione. Come contributo originale principale, la tesi propone un Modello di Maturità dell'Adozione della GenAI per i team di prodotto, postulando che le organizzazioni avanzino non utilizzando l'AI di più, bensì con maggiore discernimento, fiducia calibrata e maturità di governance. Parole chiave: Intelligenza Artificiale Generativa (GenAI); Product Discovery; Product Management; Adozione dell'AI; Technology Acceptance Model; AI Fluency; Collaborazione Uomo-AI; Augmentation; Modello di Maturità.
Generative AI in Product Discovery: Adoption Patterns, Perceived Value, Barriers, and Role Evolution Among Product Managers
TODKARI, YASH SUDHIR
2025/2026
Abstract
Generative Artificial Intelligence (GenAI) has entered knowledge-work environments at an unprecedented pace, yet little academic research has examined how it is being adopted within the specific practice of product discovery. This thesis investigates how Product Managers (PMs) are integrating GenAI into their discovery workflows, and explores the perceived value of these tools, the barriers limiting their effective use, and the implications of adoption for the evolving role of the PM. Adopting a pragmatist research philosophy and a convergent parallel mixed-methods design, the study combines a quantitative survey of 82 practicing PMs and Product Owners with a qualitative case study of four leading product companies — Figma, Notion, Miro, and Atlassian — drawn from publicly available documentation. The analysis is framed by three theoretical lenses: the Technology Acceptance Model (TAM) and the Unified Theory of Acceptance and Use of Technology (UTAUT) as adoption frameworks; Daugherty and Wilson's augmentation theory; and the AI Fluency Framework developed by Dakan and Feller, applied here for the first time in the academic literature to interpret professional GenAI adoption behaviour. The findings indicate that GenAI has become a routine instrument of product discovery for most practitioners, valued chiefly for accelerating repetitive cognitive work such as research synthesis, drafting, and coordination. This value is, however, bounded: practitioners trust GenAI markedly less than they find it useful, with hallucination risk, missing organisational context, and governance constraints emerging as the principal barriers to deeper adoption. Trust is therefore shown to be conditional and situational, and human discernment remains indispensable to the discovery process. The study concludes that the trajectory of GenAI in product discovery is one of augmentation rather than replacement, shifting the centre of gravity of the PM role away from manual production and toward judgement, interpretation, validation, and orchestration. As its principal original contribution, the thesis proposes a GenAI Adoption Maturity Model for product teams, positing that organisations advance not by using AI more, but by using it with greater discernment, calibrated trust, and governance maturity. Keywords: Generative AI (GenAI); Product Discovery; Product Management; AI Adoption; Technology Acceptance Model; AI Fluency; Human-AI Collaboration; Augmentation; Maturity Model.| File | Dimensione | Formato | |
|---|---|---|---|
|
YashSTodkari_MastersThesis_GenAI_in_ProductDiscovery_UNIVPM.pdf
accesso aperto
Dimensione
1.74 MB
Formato
Adobe PDF
|
1.74 MB | Adobe PDF | Visualizza/Apri |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.12075/26990