LAWSEEK-O
AI-powered legal assistance: finding users pain points, validating market fit, creating target users and creating user interfaces.
I worked on this project with other 3 designers during the research phase and independently in the designs. It was a two-weeks sprint so we worked fast-paced, implementing and iterating quicky to create a viable product.
Team
4 Designers
Company’s stakeholder
Deliverables
User Research, Personas, Journey Mapping, Sketching, Wireframing, Prototyping, Usability Testing
Timeline
Discovery & Research
Design & Testing
Presentation deck
My Design Process
Problem statement
LawSeek-O needs to provide a product that is low cost, jargon-free, and brings efficiency in the way users access credible legal information because the legal world can be confusing and expensive.
Stakeholder meeting
We had various meetings with the stakeholders to learn more about the project. We talked about the company's background, current users, services, and existing products. Our objective was to ensure that the solution would align with the business objectives and user pain points. During these meetings, we found that:
A considerable number of users who access LawSeek-O have limited or no legal literacy.
LawSeek-O would like to make legal advice easy and friendly to receive and understand for their users.
LawSeek-O intends to develop an AI system able to tailor each response to their users' specific circumstances.
User Research
Competitor analysis
This research revealed that LawSeek-O had 3 top competitors:
Experience map
While analysing our findings from user interviews, we identified common scenarios when seeking legal information. Through an experience map, we identified user emotions at each stage. By empathising with these emotions, we aimed to understand the user experience as a whole and find design opportunities based on the common use cases.
Opportunities for design
Address the needs of both corporate entities and individuals.
Provide users with a straightforward, user-friendly, and instant means of accessing precise, reliable legal guidance customised to their requirements.
The AI will produce personalised contract drafts based on the user's unique circumstances.
Combine affordable pricing with smart AI technology to simplify legal jargon, making legal assistance effortless and comprehensible.
Persona
Our user research revealed three primary user personas who are likely to use this product, including:
Sketches
To begin the design process, I created initial sketches to explore layout options and core platform sections. The goal was to design a solution that satisfies user needs, based on research insights and findings from the team.
Low-fidelity wireframes
Then, I translated those sketches into digital, creating wireframes using Figma, which I used for testing with potential users. The results of those user tests informed multiple iterations of the design flow.
High fidelity
Building on low-fidelity prototypes, I developed high-fidelity prototypes with a new brand's look, feel, font, and visual components. I focused on improving the UI and customising the UX to better engage the target audience. The high-fidelity prototype aimed for an approachable and reliable aesthetic design.
Onboarding
I designed the onboarding process to be simple and intuitive, guiding new users through account creation and profile setup in just a few steps.
Chat
An AI legal chat system which offers document upload and analysis, information extraction, source verification, natural language understanding, video explanations, conversation export, and user feedback for continuous AI improvement and personalised responses.
Chat History
Resources
A tailored resource section for users to access, learn, and keep up to date with legal topics in their industry.
User testing and feedback
Throughout the design process, I conducted various rounds of testing with potential users to gather qualitative insights and refine the product experience. Some design modifications implemented based on user feedback include:
Content division & Searching tool
To improve the user experience when trying to find a specific resource in the resources hub, I divided the content into areas and added a search tool.
User recommendations
To better align with user needs and considering this product's potential as an early AI legal assistance platform, I added a recommendation feature for users to request additional resources to the platform.
Some key insights that came directly from users during the testing phases:
User Experience
Testers found the platform intuitive and easy to navigate thanks to its clean, simple interface design. Many commented on the ease of finding their way around the site.
Resources Section
Users greatly appreciated having a dedicated and customised resources section with curated, relevant legal information, as it allowed them to stay up-to-date on the latest developments in their field of interest.
Trust Features
The ability to verify the sources behind the AI's answers built important trust. Users valued this transparency for fact-checking the legal information from the chatbot, given the sensitivity of legal advice.
Next Steps & Recommendations
More testing
To conduct more user testing with potential users and make necessary iterative improvements wto the product.
Design a phone version
To do more research to understand if people need for legal information on the go.
Legal market place
To allow users to gain on-demand access to lawyers if they need more in-depth legal assistance.
Final thoughts
What have I learnt?
Through this project I learnt how important user research is to uncovering user pain points and opportunities in the market, informing product requirements and positioning the product from a business standpoint. By conducting user interviews and analysing competitors, I learned the key frustrations and insights that users have when seeking legal advice.
What would I have done differently?
If I had more time, I would have conducted more testing throughout the process. Quick paper and clickable prototypes would have provided valuable user feedback to tweak the designs. Additionally, I would have conducted a workshop with the stakeholder to show them early sketches or low-fidelity wireframes that could have given useful feedback.