Enhancing B2B User Research with AI: Introducing AnAÏs, Our Virtual Design Persona. (1/4)
Chapter 1: Context and Utilizing AI in User Research
Welcome to the first article* of our series exploring how AI is transforming the landscape of user research at Akeneo! As a leading provider of Product Information Management (PIM) software, we are constantly seeking innovative ways to better understand our clients’ needs and pain points. To this end, we are excited to introduce a groundbreaking project that aims to leverage the power of artificial intelligence to enhance our user research process.
The Evolution of User Research in a B2B context
User research has always been at the heart of our design and development processes. It allows us to gain insights into our clients’ perspectives, challenges, and aspirations, helping us build software that truly meets their requirements. Traditionally, this process involves conducting interviews, surveys, and usability testing, which are invaluable but can be time-consuming and resource-intensive.
Conducting user research in a business-to-business (B2B) context presents unique challenges compared to the more common business-to-consumer (B2C) scenarios. B2B companies deal with a different set of users, limited pool of participants, limited availability of participants, limited access to users, fragmented feedback channels, and of course, confidentiality and privacy concerns.
Diverse User Roles and Needs: B2B products often serve multiple stakeholders within a single organization. This means that user research needs to encompass a wide range of user roles, each with distinct goals, pain points, and requirements. Designing research methodologies that capture insights from various user personas can be challenging and resource-intensive.
Limited Pool of Participants: B2B products typically have a smaller user base compared to B2C products. This limited pool of potential participants can make it difficult to recruit enough diverse participants for research studies. It requires strategic participant recruitment methods to ensure a representative sample.
Limited Availability of Participants: B2B professionals are often busy and may have limited availability for research sessions. Scheduling interviews or usability tests with these professionals can be challenging, leading to longer research timelines.
Limited Direct Access to Users: B2B companies might not have direct access to end users, especially when their product is used internally by another organization. This can make it difficult to directly observe user behaviors and gather qualitative insights. In addition, if we don’t talk to the right people, we can’t do user-centric design and improve our products while taking the user experience into account.
Fragmented Feedback Channels: B2B users might provide feedback through various channels, including customer support, sales teams, and online forums. Collating and analyzing this feedback to extract actionable insights can be challenging. Furthermore, going through intermediaries to collect user feedback carries a risk of not formulating the problem correctly.
Confidentiality and Privacy Concerns: B2B products often handle sensitive business information. This presents challenges in terms of data security and the willingness of participants to share information during research studies. Companies must navigate privacy concerns while still gathering valuable insights.
“…exploring AI-driven approaches to complement traditional research efforts without distorting the fundamental principle of the discipline: human first.“
In light of these challenges, B2B companies must adopt tailored user research strategies that account for the nuances of their user base and business environment. Strategies might include building relationships with key clients, conducting in-depth interviews, meeting with clients in face to face, utilizing remote research methods, integrating in-app tools, and exploring AI-driven approaches to complement traditional research efforts without distorting the fundamental principle of the discipline: human first. In this series of articles we will focus on the latter point, exploring one of the many possibilities that AI offers when it comes to user research.
Unleashing the Power of AI
In our quest for more efficient and comprehensive user research, facing those challenges, we’ve embarked on a journey to integrate AI into the process. We’ve named our AI creation “AnAÏs,” and she’s more than just lines of code. AnAÏs is designed to be a generative AI that will serve as our main design persona. But what exactly does that mean?
Imagine AnAÏs as a virtual team member who embodies the characteristics, preferences, and behaviors of a typical user. We’re not talking about a mere chatbot or automated response system. AnAÏs is being trained to think, react, and respond like a human user, enabling us to simulate user interactions and gather insights on-demand.
We are convinced that this approach could help us overcome the participant recruitment difficulties we encounter in certain specific situations. Indeed, it is not a question of using this innovation to replace our real users who will always be essential due to their unique, unpredictable, sensitive, and quite simply human nature. Indeed, AnAÏs could help guide certain decisions very early in projects, or when we must act quickly to correct a bug, deliver a critical improvement, or even deliver a new feature for which we already know the strong potential.
The Path Ahead
In this article, we’ve set the stage for our exciting AI-driven user research project. Our intention is to harness the capabilities of AI to transform the way we gather insights, validate assumptions, and enhance our understanding of client needs. The journey ahead involves training AnAÏs to become a virtual design persona and integrating her into our research processes seamlessly.
We are aware that this AI tool will never replace real users. This is the reason why we want, in this series of articles, to reflect and initiate discussions on the use that will be made of this virtual persona. When and how to use it? And what value should be given to the answers she will provide us?
Stay tuned for our next article, where we will dive deeper into the technical aspects of setting up our generative AI, AnAÏs, and how we’ve worked to imbue her with the characteristics of a relatable and valuable design persona. We’re excited to share our progress and insights with you as we embark on this innovative endeavor!
In the meantime, feel free to share your thoughts and questions about the role of AI in user research. We believe that by embracing these advancements, we can better serve our clients and create a software that truly addresses their pain points and aspirations.
*An article written with the help of generative AI (ChatGPT 3.5).
