Enhancing Customer Onboarding Workflows In B2B SaaS Through AI Interventions: A Qualitative Case Study
Ghaderiazad, Parnian (2025)
2025
Master's Programme in Computing Sciences and Electrical Engineering
Informaatioteknologian ja viestinnän tiedekunta - Faculty of Information Technology and Communication Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2025-12-29
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-2025122612132
https://urn.fi/URN:NBN:fi:tuni-2025122612132
Tiivistelmä
The prevalence of Software-as-a-Service (SaaS) has been on the rise both globally and in Finland, making more digital products accessible to end users via the internet rather than local installations. From a business perspective, SaaS products are offered via a subscription-based business model. Perceiving the value of SaaS offerings is thus crucial for product adoption and avoiding churn. In Business-to-Business (B2B) contexts, Customer Success teams, and more specifically Customer Success Managers (CSMs), are usually responsible for facilitating the process of onboarding customers and helping them realize the product’s value to reduce churn, the rate at which customers stop using a product.
This thesis was conducted in collaboration with a Nordic B2B SaaS company, referred to as the case company in this document. The case company provides expert organizations with features for managing projects, resources, finances, time, and customer relationships.
The use of artificial intelligence (AI) in the workplace and human-AI collaboration has long been studied. However, it is understudied what challenges CSMs face and how AI can be used to streamline their workflow. Therefore, I aim to answer two research questions: (1) What are the dimensions and challenges of customer onboarding in B2B SaaS products? and (2) How could AI support the Customer Success teams in streamlining onboarding tasks?
To answer the research questions, first, two briefing sessions in the form of semi-structured interviews were conducted with the UX Researcher and Product Manager of the case company to gain an understanding of the broad customer onboarding landscape. Next, semi-structured interviews with all four CSMs were conducted to develop a more detailed understanding of their customer onboarding workflow, the challenges they face, and how they create value. After analyzing the interview transcripts using thematic analysis, two mid-fidelity AI-assisted concepts were designed, aiming to address the most recurring challenges CSMs face and to explore how AI can streamline their workflows. Lastly, a survey was run with the CSMs consisting of video prototypes of the designed concepts, each followed by five open-ended questions, allowing them to express their opinions freely.
Findings regarding the first research question, informed by CSMs interviews, are (1) a comprehensive mapping of the nine-step customer onboarding process; (2) conceptualizing how CSMs create value in this process, and how their tacit knowledge and idiosyncrasies can result in the use of distinct practices; (3) explaining why customer onboarding is never a one-size-fits-all process and the role of customer characteristics; and lastly, (4) identifying the current challenges CSMs face and the AI intervention opportunities.
Regarding the second research question, it is evident from the survey results that both manual software environment preparation and manual error resolution and data field formatting of CSV files can be augmented with AI. However, the perceived usefulness can differ among CSMs due to their idiosyncrasies. Additionally, actual concept implementation and longitudinal exposure to the system are needed to evaluate the amount of trust and perceived control.
The main contribution of this thesis is the synthesis of four dimensions that may help other designers/researchers when exploring how AI could streamline CSM workflows in B2B SaaS contexts: (1) The focus should expand from user onboarding to customer onboarding; (2) Human-to-human interactions should be preserved; (3) The use of AI should still be flexible; and (4) New AI interventions require proper onboarding themselves.
This thesis was conducted in collaboration with a Nordic B2B SaaS company, referred to as the case company in this document. The case company provides expert organizations with features for managing projects, resources, finances, time, and customer relationships.
The use of artificial intelligence (AI) in the workplace and human-AI collaboration has long been studied. However, it is understudied what challenges CSMs face and how AI can be used to streamline their workflow. Therefore, I aim to answer two research questions: (1) What are the dimensions and challenges of customer onboarding in B2B SaaS products? and (2) How could AI support the Customer Success teams in streamlining onboarding tasks?
To answer the research questions, first, two briefing sessions in the form of semi-structured interviews were conducted with the UX Researcher and Product Manager of the case company to gain an understanding of the broad customer onboarding landscape. Next, semi-structured interviews with all four CSMs were conducted to develop a more detailed understanding of their customer onboarding workflow, the challenges they face, and how they create value. After analyzing the interview transcripts using thematic analysis, two mid-fidelity AI-assisted concepts were designed, aiming to address the most recurring challenges CSMs face and to explore how AI can streamline their workflows. Lastly, a survey was run with the CSMs consisting of video prototypes of the designed concepts, each followed by five open-ended questions, allowing them to express their opinions freely.
Findings regarding the first research question, informed by CSMs interviews, are (1) a comprehensive mapping of the nine-step customer onboarding process; (2) conceptualizing how CSMs create value in this process, and how their tacit knowledge and idiosyncrasies can result in the use of distinct practices; (3) explaining why customer onboarding is never a one-size-fits-all process and the role of customer characteristics; and lastly, (4) identifying the current challenges CSMs face and the AI intervention opportunities.
Regarding the second research question, it is evident from the survey results that both manual software environment preparation and manual error resolution and data field formatting of CSV files can be augmented with AI. However, the perceived usefulness can differ among CSMs due to their idiosyncrasies. Additionally, actual concept implementation and longitudinal exposure to the system are needed to evaluate the amount of trust and perceived control.
The main contribution of this thesis is the synthesis of four dimensions that may help other designers/researchers when exploring how AI could streamline CSM workflows in B2B SaaS contexts: (1) The focus should expand from user onboarding to customer onboarding; (2) Human-to-human interactions should be preserved; (3) The use of AI should still be flexible; and (4) New AI interventions require proper onboarding themselves.