Conversational Experience | Startup Client
Architecting a Chatbot Conversation Using Iterative Ideation & Prototyping
How might we structure a conversational experience using a research-driven, content first approach?
Leveraging natural human interaction patterns to set context and mitigate communication gaps between the users within a product ecosystem.
Scenario & Background
Reserve was a restaurant technology startup with a connected ecosystem of B2B and B2C technology products that facilitated a two-way communication network - allowing for real-time demand for a reservations to be disseminated from consumers directly to restaurants.
For restaurants, this system allowed for a better understanding of customer demand for tables at given time and better understand, optimize, and manage their business. For diners, this system democratized the supply chain for tables at trendy restaurants, allowing them to ‘request’ a reservation at a restaurant directly from a restaurant, and receive a yes/no answer, or potentially an ‘offer’ with alternate accommodations - sometimes within minutes, but often longer.
The existing, static UI pattern implementation being used to communicate information to a user regarding the request process and the status of their “request” proved to be confusing and frustrating for consumers, resulting in: (1.) them being confused by the delay in receiving a response - often immediately cancelling the request, (2.) assuming they had a confirmed reservation when they did not - resulting in them arriving at a restaurant without a reservation, or (3.) confused or angry when confirmed for a reservation hours after beginning the process - often immediately cancelling their confirmed reservation.
The ask for this phase of work was to bring a conversational experience to life though that would clarify and provide transparency into the reservation request process, while setting and managing a user’s expectations throughout. A previous phase of work that included multiple rounds of iterative research and concept testing validated that a text-based, conversational UI could clarify the request process for our users.
Reserve was a small startup with big goals, a lean team and limited resources. I was the design lead on this project and responsible for leading planning, research and design from the ideation phase through to development and launch. Additionally, I consulted and collaborated directly with an additional designer during the initial ideation and design sprint as well as the CPO, VP of Design, CEO, COO, and CTO throughout the lifecycle of the project to foster inclusion, generate ideas, and facilitate progress and decision making.
My Role & Contribution
As the design lead on this project, I was responsible for leading all aspects of design definition, design execution, and product management for this feature.
Design Definition Activities
Included collaborating with business and technology stakeholders to define feature requirements, KPI's, & goals, defining and conducting user research and discovery, and organizing collaborative ideation activities with users and stakeholders.
Design Execution Activities
Included the definition of system architecture, key flows and interactions, the design of iterative wireframe design, the planning and facilitating of both in-person and online user testing sessions, user interface & visual design, and the creation of developer specifications & documentation.
Product Management Activities
Included the definition of project timelines & milestones, defining requirements & mapping corresponding user stories, setting and maintaining the cadence of design sprints, facilitating stakeholder discussion & decision making, and collaborating with development team through development, QA, design reviews & launch.
Process & Deliverables
The final output of this project was a conversational experience (chatbot) and corresponding technology infrastructure that connected diners and restaurants, providing them with the ability to communicate within the system via an ecosystem of applications.
The architecture and flow of the final conversational architecture was heavily influenced by research into the interactions and conversations that were taking place between diners and restaurants in the physical world. Observations and insights from this research informed our designs, allowing us to better empathize with our users needs and structure moments and language that catered to these needs while creating a familiar and trustworthy environment.
Phase 1: Discovery & Planning
As a first step, it was critical to gather internal and external data that would help frame the problem, contextualize the landscape, and garner a thorough understanding of both the players and factors influencing the challenge ahead.
Phase 2: Generative Ideation
Design thinking techniques were used in various capacities to generate and synthesize ideas, extract and leverage domain knowledge and key insights from the core team, and establish trust and good will toward the project.
Phase 3: Experience Design
The ideation and prototyping process yielded a foundational framework of our conversation ecosystem. In-depth experience and interaction design was now required to take the experience to the next level, bringing it to life.
Phase 4: Visual Design
Iterative wireframe design, user testing, and visual design were the final steps in the design process – taking place after the structure of the experience and core interactions had been defined and battle-tested.
Outcomes, Learnings, and Takeaways
After launch, the first release of the chatbot and corresponding conversational experience was met with a high-level of positive customer feedback. Analytics also reflected how the new experience delivered on two key metrics - increasing the % of accepted ‘offers’ by double digits and also substantially decreasing the number of immediate reservation cancellations.
Learnings & Takeaways:
Designing content-first is a powerful way to define an experience
Initial discovery and research is still a requirement, but utilizing a collaborative, content-first, collaborative approach to designing an experience will provide a completely different context and perspective that will help you uncover information and generate ideas.
Understanding a users active intent should be your top priority
Conversational interactions offer an opportunity to exchange information with a user. If structured properly, this exchange allows us to understand not only what they are asking for at a given time, but also the context and their active intent - creating the opportunity to deliver personalized content that surprises and delights.
User expectations rise sharply when shown a chat interface
Chat interfaces are familiar to users, and they are used to using them in a specific way - to communicate with other humans in a two-way, open-ended, multi-layered conversation. This mental model results in heightened expectations that even the best chatbots have a hard time meeting.
Human or machine? The line can be hard for a user to discern
It is important to consider how you want your chatbot to be perceived by a user - as a machine or another human. But even then, it may be hard for users to discern what they are interacting with, don’t be surprised by how many people will respond with “thank you!”. That’s a good thing though! ;0
Conversational interaction patterns can be utilized in many ways
The power and flexibility of conversational interaction patterns can be leveraged in a multitude of ways, with substantially less development effort, by thinking differently about how they can be used — then designing creative solutions that push conversational experiences in novel directions.
Want to know more?
An in-depth case study of the design process & deliverables for this project is available on Behance.
This case study is also published in the Chatbot.com Publication on Medium.com
Additionally, I discussed the process of designing a chatbot experience with Jared Spool & Steph Hay in Episode 14 of The UIE Podcast: "More Human than Human?: Designing a Conversational UI"