AI Tools for UX Research / Market Research / Competitive Analysis – Project 1
Collaborative Design with AI Assistance
Pharmaceutical Industry – Failure Mode and Effects Analysis
As a Principal UX Designer at a QMS software company, my role involves extensive discovery across diverse industries, a notably time-intensive aspect. In integrating AI into the design process, I initially focused on the life science sector, seeking AI-driven insights to develop solutions that assist organizations in complying with FDA regulations, reducing quality and compliance risks, and safeguarding their brand value.
Utilizing ChatGPT, I started with a general question. Then, I progressively refined my approach to gain deeper insights into user experiences.
- Prompt #2 – I’m looking to understand the most common or interesting problems that people face when it comes to FMEA in the pharmaceutical industry. What do you know about their struggles?
ChatGPT answered with 15 challenges that everyday production faces , ranging from regulatory hurdles to technological complexities.
ChatGPT: “Failure Mode and Effects Analysis (FMEA) in the pharmaceutical industry is a critical process, but it comes with its own set of challenges and problems. Understanding these issues is key to improving the FMEA process and ensuring product quality and safety. Here are some of the most common and interesting problems faced in this context:”
PW: AI4Design!!
With the next prompt, I wanted to delve a bit deeper:
- Prompt #3 – I’m particularly interested in learning more about what issues arise in regards to Cost of Poor Quality (COPQ) associated with providing poor quality products s in the pharmaceutical industry and what strategies companies in this industry use to document these issues. Can you expand on these two points?
AI-Generated Interviews and Personas – Project 1
Gaining Valuable Insights
Creating sharable User Personas and User Journey maps.
- Prompt #5 – Can you imagine an ideal persona of someone working in a role associated with COPQ in the pharmaceutical industry. What is their title, possible pain points, and their attributes and characteristics?
ChatGPT: Certainly! Creating a persona for someone working in a role associated with the Cost of Poor Quality (COPQ) in the pharmaceutical industry can help in understanding the challenges and needs specific to this position. Let’s imagine a persona named Dr. Sarah Bennett:
PW: AI4Design!!
UX Design Insights and Artifacts
AI Tools for Research
Generative AI platforms like ChatGPT have absorbed a vast amount of information from the internet. They offer intriguing insights as you envision potential solutions and the typical traits of your target users.
Employing AI tools to mimic a discussion with a character might be seen as contentious and tends to produce rather clichéd and broad information. However, it serves as a useful preliminary step in identifying key themes. It’s important to note that actual conversations with real users are likely to both test and confirm the initial findings from AI research.
- Prompt #6 – Can you give me an interview script for an interview unpacking product development costs in the pharmaceutical industry problems with Dr. Sarah Bennett?
ChatGPT: Certainly! Below is an interview script designed to unpack product development costs and associated problems in the pharmaceutical industry with Dr. Sarah Bennett, a Quality Assurance Manager. The script is structured to explore her insights, experiences, and perspectives on this topic.
- Prompt #7 – Can you turn this interview script into an actual conversation with Dr. Sarah Bennett?
ChatGPT: Absolutely! Let’s imagine a conversation based on the provided interview script with Dr. Sarah Bennett, a fictional Quality Assurance Manager in the pharmaceutical industry.
Defining the Problem Statement – Project 1
What is the Research Attempting to Address?
Bridge the gap between the existing issue and the target objective.
The next step is to translate all these insights into a problem statement that you are aiming to solve, and then moving into a set of features and user stories. I began this phase by asking for insights that identifies and prioritizes problems brought forward from my previous interactions.
- Prompt #8 – What are the main insights from this interview? What are the main problems Sarah is facing?
ChatGPT: From the interview with Dr. Sarah Bennett, a Quality Assurance Manager in the pharmaceutical industry, several key insights and main problems can be discerned:
- Prompt #9 – Can you prioritize these in the order of importance and how often they were mentioned?
ChatGPT: Certainly! Based on the frequency of mention and the implied importance in the interview with Dr. Sarah Bennett, the main problems can be prioritized as follows:
- Prompt #10 – Using the problems and priorities you’ve listed above, consolidate everything into a single Problem Statement that will serve as the design goal we set to solve as we continue with the UX process.
ChatGPT: Based on the identified problems and their prioritization, the consolidated Problem Statement for the design goal in the UX process can be articulated as follows:
PW: AI4Design!!
Problem Statement for UX Design
AI generated from the insights realized.
- My Prompt #11 – Can you shorten this?
AI for HMW (How Might We’s) and User Stories – Project 1
How Might We’s?
Transforming the problem statement into actionable design queries.
“How might We’s” are a useful way to formulate actionable design questions based on the problem statement. They are a tool used in design thinking and problem-solving processes to open up the space for creative thinking and innovative solutions. They are framed in a way that invites expansive thinking and encourages a positive and exploratory approach to challenges.
- Prompt #12 – Based on this problem statement, can you generate 4 How Might We’s?
ChatGPT: Certainly! Based on the problem statement, here are four “How Might We” (HMW) questions to inspire solutions in the UX design process:
User Stories
Translating How Might We’s into user stories.
- Prompt #13 – Transform these into user stories.
ChatGPT: Transforming the “How Might We” questions into user stories, we get the following:
These user stories focus on the needs and goals of Quality Assurance Managers in the pharmaceutical industry, providing a user-centered perspective for designing solutions that address their specific challenges.
PW: AI4Design!!
AI for Wireframes, Mock-ups, Prototyping, and UX Writing – Project 1
Generative AI – Exploring UX Wireframes
Wireframes are foundational in interface design, often being the first deliverables in a project. Designers initially face the challenge of a blank canvas, which is where generative AI tools become useful. These tools help grasp project needs and suggest early wireframe ideas, aiding in the design process. For example, using a tool like ChatGPT, you can visualize a feature’s interface and then use specialized tools to begin creating and refining wireframes.
Generative AI Tools I explored:
- WireGen: This plugin for Figma creates modifiable wireframes from textual descriptions. With a single click, you can produce various wireframe sets. If the initial outcome doesn’t meet your expectations, you can easily instruct the AI to make modifications or generate new versions with minor alterations in the prompt.
- Uizard: This adaptable tool is ideal for crafting wireframes, mockups, and prototypes. It offers the capability to generate interfaces from text prompts, transform hand-drawn sketches into wireframes, and reverse-engineer images into wireframes. Uizard features a collection of ready-to-use templates and UI components for designing interfaces, and it supports both drag-and-drop editing and integration with Figma.
Other resources:
- Galileo AI: This tool creates high-quality, customizable UI designs from text descriptions. It blends UI elements with AI-generated images and content to convert language prompts into attractive designs. Galileo AI’s integration with Figma allows for seamless incorporation of AI-designed elements into your projects.
- Visily: An AI-powered tool that turns screenshots or text into editable wireframes and prototypes. It includes a component library and supports both low- and high-fidelity designs. Visily facilitates collaborative editing and offers brainstorming tools like flowcharts and sticky notes, all integrated with Figma.
- Typper: Acting as a virtual design assistant, Typper provides design recommendations to enhance the layout and accessibility of interfaces. It generates text, icons, and images from text prompts, aiding in the design process.
PW: AI4Design!!
AI Tools for User Testing: Evaluate Your Designs
Once you’ve developed high-fidelity designs complete with appealing UX copy, the next step is testing. While AI will never – and hopefully should never – substitute real users, it can assist in assessing and refining your designs before presenting them to real people. This could potentially save you a couple of rounds of revisions.
Attention Insight: This tool predicts the areas users are likely to focus on within your design. Beyond generating predictive heatmaps, it aids in design comparison. You can use different versions of your own design or even analyze a competitor’s design to assess the effectiveness of your interface.
How does Attention Insight interpret Clarity Score?
The Clarity score evaluates the user-friendliness of a design for newcomers, determining whether it appears overly busy with elements and conflicting colors or if it’s neat and straightforward. This score is calculated by an algorithm that considers various design aspects, such as the quantity of text, text size and contrast, color vibrancy, and the number and size of images. 
A higher score signifies a cleaner, more streamlined design. Conversely, a lower score suggests that the design is cluttered, making it difficult to spot key elements.
For instance, a Clarity score of 79% indicates optimal clarity, which is an excellent outcome. You can also benchmark your design’s clarity against industry standards using our Clarity benchmark tool.
Clarity scores range from 1 to 100 and are categorized into four distinct groups.
- Severe Difficulty (0-29)
- Moderate Difficulty (30-59)
- Optimal Clarity (60-94)
- Too Simple (95-100)
AI Tools for UX Research / Market Research / Competitive Analysis – Project 2
Collaborative Design with AI Assistance
Locating Content in a Multiple Module QMS application.
In my second phase of exploration with ChatGPT for research generation, I aimed to uncover insights into the what, why, and how of users navigating the search for specific documents within a sophisticated QMS application.
Leveraging ChatGPT for this study, I initiated my inquiry with the same broad question I used in project 1 above. Subsequently, I narrowed down my focus to gain more precise insights.
- Prompt #1 – “I’m looking to understand the most common or interesting problems that people face while locating a particular item in a complex document heavy multiple module QMS app. What do you know about their struggles?”
ChatGPT answered with 12 challenges that everyday users face, ranging from navigation complexity to poor metadata or tagging.
ChatGPT: “When users are trying to locate a specific item in a complex, document-heavy, multi-module Quality Management System (QMS) application, they often face several common and interesting challenges. Here’s an overview of these struggles:”
Today manufacturers must build products on time, first time, and every time to achieve and maintain competitive standing in global markets.
In every conversation with ChatGPT that I have, I ask a similar question; I want to learn where I can find more resources for my research and perhaps validate the findings being presented by ChatGPT.
- Prompt #3 –“You’ve presented and shown some interesting insights; can you point me towards some research that provides data for: What users struggle with while locating particular items in a complex content management system?”
Turning Findings Into Sharable Artifacts – Project 2
UX Design Insights and Artifacts
Generative AI streamlines prioritizing user experience redesigns and new feature exploration by analyzing user data to identify key trends and gaps. It predicts user responses and highlights high-impact areas, enabling efficient focus on enhancements that significantly improve user satisfaction and engagement. This data-driven approach ensures resources are effectively allocated to the most beneficial initiatives, leading to more successful, user-centric products.
- Prompt #17 – What are 5 phases in the user journey that Jacob Martinez would experience working in a QMS app that does not have the features described above? What makes him feel emotionally happy along this journey, what makes him feel emotionally anxious along the journey?
ChatGPT: In a QMS app lacking the advanced features described earlier, Jacob Martinez’s user journey can be broken down into five distinct phases. Each phase has elements that could contribute to his emotional satisfaction or anxiety.
PW: AI4Design!!
User Needs, Wants, and Desires – Utilizing Generative AI
Generative AI can play a significant role in determining user needs, wants, and desires in several impactful ways:
- Data Analysis and Pattern Recognition: Generative AI can analyze large datasets of user interactions, feedback, and behavior patterns. By identifying trends and commonalities in this data, AI can uncover underlying user needs and preferences that might not be immediately apparent.
- Personalization and Customization: AI algorithms can generate personalized content, recommendations, or product suggestions based on individual user data. This helps in understanding and catering to specific user desires and preferences.
- Predictive Modeling: Generative AI can predict future user behavior and preferences based on historical data. This predictive capability allows businesses to anticipate user needs and develop solutions proactively.
- Sentiment Analysis: By analyzing user feedback, reviews, and social media conversations using natural language processing (NLP), AI can gauge user sentiments, helping to understand their emotional responses and satisfaction levels.
- User Simulation and Scenario Analysis: AI can simulate various user scenarios to test how different user groups might interact with a product or service. This can reveal unmet needs or potential areas for improvement.
- Rapid Prototyping: AI can quickly generate prototypes for products, services, or interfaces based on user data, allowing for rapid testing and iteration. This helps in refining offerings to better match user expectations.
- Market Trend Analysis: Generative AI can analyze market trends and consumer behavior, providing insights into what users might want or need in the future.
- Enhancing User Research: AI can assist in user research by automating the collection and analysis of user data, enabling researchers to focus on interpreting the results and making informed decisions.
AI Tools for Design Iteration and Problem Solving – Project 2
Generative AI for Wireframes, Mock-ups, and Prototypes
Generative AI streamlines prioritizing user experience redesigns and new feature exploration by analyzing user data to identify key trends and gaps. It predicts user responses and highlights high-impact areas, enabling efficient focus on enhancements that significantly improve user satisfaction and engagement. This data-driven approach ensures resources are effectively allocated to the most beneficial initiatives, leading to more successful, user-centric products.
- Prompt #19 – From Jacob’s 3 interview, what are 3 outcomes learned for each of the 5 phases in the user journey and what are 3 associated opportunities learned for each of the 5 phases in the user journey for enhancing her user experience in the app?
ChatGPT: Based on Jacob’s interview and the outlined user journey phases in a QMS app, we can derive specific outcomes and associated opportunities for each phase to enhance his user experience. Here’s a breakdown:
PW: AI4Design!!
How Might We’s?
Generative AI aids in formulating “How Might We” (HMW) questions by analyzing user data, feedback, and behavior patterns to uncover challenges and opportunities in the user experience. These AI-generated insights can lead to targeted HMW questions that focus on specific user needs and pain points. In the design process, HMW questions are instrumental in framing problems as opportunities for creative thinking and innovation. They encourage a user-centric approach, fostering a mindset of possibility and open-ended exploration among design teams. This leads to the generation of a wide range of ideas and solutions, directly addressing user needs and enhancing the overall design strategy.
- Prompt #11 – Based on this problem statement, can you generate 5 How Might We’s?
ChatGPT: Certainly! “How Might We” (HMW) questions are a great way to spark innovative thinking and explore potential solutions. Based on the problem statement, here are five HMW questions:
AI Tools for UX Research / Market Research / Competitive Analysis – Project 3
Collaborative Design with AI Assistance
Form/Workflow builder in a QMS application.
For my third research effort, I set out to learn more about the pain-points for 2 different end users in a QMS application.
- Emily Johnson: Is the author/user of regulatory process documents. A part of her role is to create forms and workflows for different departments to utilize.
- Michael Smith: Needs to locate the correct forms to initiate the workflows that are typical in a QMS. Looking at his struggles to fill out, and submit the forms that are used in his role as a Production Line Supervisor in the manufacturing industry.
Using ChatGPT, I began with a general question about working with a form builder integrated into a QMS app to get the conversation rolling.
- Prompt #1 – “I’m looking to understand the most common or interesting problems that people face while working in a form builder that is integrated into a QMS desktop application. What do you know about their struggles?”
ChatGPT: “When working with a form builder integrated into a Quality Management System (QMS) desktop application, users often encounter a variety of challenges. These challenges can range from technical issues to usability concerns. Here are some of the most common or interesting problems they might face: ”
PW: AI4Design!!
With the next prompt, I wanted to delve a bit deeper:
- Prompt #2 –I’m particularly interested in learning more about what pain points arise in regards to user experience and what strategies user implement. Can you expand on these two points?
ChatGPT: Certainly! Let’s delve deeper into the user experience pain points associated with using a form builder in a Quality Management System (QMS) desktop application and explore the strategies users often implement to address these challenges.
Creating UX Artifacts with Collected Insights – Project 3
UX Design Insights and Artifacts
Creating sharable User Personas and User Journey maps.
- Prompt #4 – Can you imagine an ideal persona of someone that creates forms and regulatory workflows in the manufacturing industry? What is their attributes and characteristics?
ChatGPT: Certainly! Creating an ideal persona for someone who designs forms and regulatory workflows in the manufacturing industry involves understanding the unique blend of skills, experience, and attributes required for this role. Here’s a detailed persona:
- Prompt #5 – How about a persona for users within the company that use the forms and workflows that have been created by individuals like Emily Johnson in the manufacturing industry? What are their attributes and characteristics?
ChatGPT: Absolutely, creating a persona for users within a manufacturing company who interact with forms and workflows designed by individuals like Emily Johnson can help in understanding their needs and challenges. Here’s a detailed persona for such a user:
UX Design Insights and Artifacts
AI Tools for Research
User Personas and User Journey Maps are crucial in UX Design, providing deep insights into user needs and experiences. Personas represent key user groups, guiding designers to create more targeted and relevant products.
Journey Maps illustrate the steps a user takes to achieve a goal, revealing their emotions and challenges. These tools ensure a user-centered design approach, enhancing the product’s functionality and user engagement.
- Prompt #24 – What are 5 user journey phases Emily experiences when she is designing forms and workflows for other departments to utilize? For each phase, what are 5 outcomes that can be determined by her struggles and 5 possible enhancements that would help Emily with the form design process?
ChatGPT: In a QMS app without the enhanced features previously described, Emily Johnson’s user journey can be broken down into five distinct phases. Each phase presents its own set of emotional responses, ranging from happiness to anxiety, based on the challenges and successes she encounters.
- Prompt #27 – Michael is responsible for submitting CAPA forms, Accident and Safety Reporting, Nonconformance Reports, for each phase, what are 5 outcomes that can be determined by his struggles and 5 possible enhancements that would help Michael with the form design process?
ChatGPT: Based on Michael’s interview and the identified pain points in the existing QMS app, we can extract key outcomes and associated opportunities for each phase of his user journey. These insights can guide enhancements to improve his experience with the app.
PW: AI4Design!!
AI Tools for Design Iterating: Wireframes, Mock-ups, Prototyping, and UX Writing – Project 3
Prototypes and Usability Testing
Generative AI can significantly enhance the process of creating prototypes and conducting usability testing in several ways:
- Rapid Prototyping: Generative AI can quickly produce a range of design prototypes based on specified parameters or learned preferences. This accelerates the design process, allowing for more iterations and refinements in less time.
- Personalized User Experiences: By analyzing data on user behavior and preferences, AI can generate personalized prototypes that cater to the needs of different user personas, leading to more effective usability testing.
- Automated Feedback Analysis: AI can analyze feedback from usability tests, identifying patterns and insights that might be missed by human analysts. This can lead to more focused and effective iterations on the design.
- Predictive Modeling: Generative AI can predict how changes in the design might impact user experience, helping designers make informed decisions about which modifications will most improve usability.
- Accessibility Testing: AI can automatically test prototypes for accessibility issues, ensuring that the design is usable by people with a wide range of abilities.
- Real-time Adaptation: In advanced applications, AI can modify prototypes in real-time during usability testing based on user interactions, providing immediate insights into how changes affect user experience.
By integrating generative AI into the prototyping and usability testing process, designers can create more user-centric products efficiently and effectively, ultimately leading to better user experiences and more successful products.
PW: AI4Design!!
AI Generated Heatmaps and the Design Process
AI-generated heatmaps are a powerful tool for designers in the design process, offering several key benefits:
- User Engagement Insights: Heatmaps visually represent where users are most frequently clicking, touching, or looking on a screen. This data helps designers understand which areas of their design are attracting the most attention and engagement.
- Usability Analysis: By highlighting the most and least interacted areas, heatmaps can reveal usability issues. For instance, important elements that receive less attention might need to be redesigned or repositioned for better visibility.
- Optimizing Layouts: Designers can use heatmap data to optimize the layout of a webpage or application. Understanding where users focus their attention allows for strategic placement of key content or call-to-action buttons.
- A/B Testing: In comparing heatmaps from different design variations, designers can quantitatively assess which design performs better in terms of user interaction and engagement.
- Content Effectiveness: Heatmaps can show how users interact with content, like how far they scroll on a page. This helps in determining the effectiveness of content placement and length.
- Navigation Improvement: Analyzing how users navigate through a site or app, indicated by their interaction hotspots, can lead to improvements in the navigation structure, making it more intuitive.
- Reducing Clutter: Identifying areas of a design that are consistently ignored can help in decluttering the interface, leading to a cleaner, more focused user experience.
Overall, AI-generated heatmaps provide a data-driven approach to understanding user behavior, enabling designers to make informed decisions to enhance usability, engagement, and overall user experience.
Utilizing AI in the Design Process
Top 5 Key Take Aways
- Enhanced User Understanding: AI can analyze vast amounts of user data, providing deeper insights into user behavior, preferences, and needs. This allows UX designers to create more personalized and user-centric designs.
- Efficiency in Design Iterations: AI tools can automate repetitive tasks in the design process, such as wireframing or prototyping. This speeds up the iteration process, allowing designers to focus on more creative aspects of UX design.
- Predictive Design Solutions: AI’s predictive capabilities enable designers to foresee user needs and behavior patterns. This can lead to proactive design solutions that enhance user experience even before users articulate their needs.
- Accessibility and Inclusivity: AI can assist in making designs more accessible to a diverse range of users. By analyzing data from various user groups, AI can help in creating designs that are inclusive and cater to people with different abilities and preferences.
- Ethical and Responsible Design: AI in UX design brings a responsibility to use data ethically and design responsibly. Designers must ensure that AI-driven designs respect user privacy, promote fairness, and do not perpetuate biases, thereby fostering trust and a positive user experience.
Final Summary
To remain at the forefront of design, particularly with emerging technologies like AI, it’s essential to continuously adapt and learn. Embracing these advancements early on is key to staying ahead.
Consider integrating AI into your design practices and explore how it can enhance your workflow. Remember, AI cannot replace human attributes such as critical thinking, empathy, and creativity. Our unique capabilities in understanding intricate information and emotions, conjuring new concepts, and collaborative efforts distinguish us. As a designer, you represent your users’ interests. In this evolving AI landscape, ensure that you employ AI in an ethical and beneficial manner.
See full conversations with ChatGPT and links to design iterations in Uizard.io, and Figma.
PDF Password: AI4Design!!
