Where the Rubber Meets the Road
Sprint 3 is underway and our team is ready to start understanding what our hours of research are trying to tell us. While we are still conducting interviews, we are ready to start using (and creating) tools for us to organize that research in a way to help us make sense of the problem and hopefully shine a light on a design opportunity or two.
Yes, more research
Our second sprint was all about research interviews, but we still had more to learn from customers, CarMax, and experts in the field of conversational agents. Even though we are moving into a state of understanding, our team is committed to continually use research to explore the space, test our ideas, and, most importantly, tell us when we are wrong.
Affinity Diagram
As our team started accumulating data from interviews with car shoppers, CarMax insiders, domain experts, and a few lucky pretotype participants we began to synthesize our findings in parallel. Given that we are on a continuous path of research, this stack of information was only going to grow. We had notes, and we wanted insights.
Our team looked to the affinity diagram as a method of working with countless notes and findings to pull out insights or at least some understanding of the problems and opportunities in front of us.
The notes from the interviews were recorded to sticky notes in Miro, grouped into related clusters, the clusters are labeled with an ‘I’ statement, and then grouped again to derive an understanding or, hopefully, an insight.
- The identity of the conversational agent is important in framing its relationship with the user.
- Users have expectations of what a voice agent could do, and its utility is connected to its ability to meet those expectations.
- Customer satisfaction in the car buying process is influenced by their expectations for a salesperson, ability to access information, wait times, and finding the ‘right’ car.
- Sales associates’ success is directly related to their ability to guide the customer to the right car for them, rather than just any car, and their ability to accomplish this often lies in their ability to get useful information from the customer.
Ideally, the result of an affinity diagram is to gain a rich understanding of what the data is telling us at a high-level. However, the real value our team derived from creating this diagram was in the conversations had along the way, and we felt that there may be another tool for us to use. If there wasn’t, then we would create it.
Creating the tools we need
Opportunity Space Exploration
The team wanted a new way of organizing the problem space we were in. We decided that we should create a tool that could capture some of the opportunities that arose in conversation while tirelessly moving notes around the affinity diagram.
The beautiful mess above is the Opportunity Space Exploration Board (OSE). It is a way for our team to identify problem statements that arose in our research and explore their efficacy as viable design opportunities. At the top are problem statements from the point of view of a stakeholder. Below are things we know about that problem, things we don’t, potential solutions, and their benefits to various stakeholders.
Simplifying the Service Blueprint
Our team knew that we needed to develop a rich understanding of the service-side of CarMax. To improve any part of that service using conversation, we would need to see where CarMax’s service interacted with the customer journey and what may be happening behind the scenes.
The problem with many service blueprints lies in their complexity. They can be large visuals with lots of arrows, and we already have the daunting affinity diagram. So we decided to take a new approach that gave our team license to add interventions and focus on what was absolutely necessary for our ideation. For each problem statement in the OSE, we have started creating mini service blueprints that encapsulate that problem and potential conversational solutions.
We see more uses for this tool in the near future. Maybe we start storyboarding the solutions below or start to fill in the more complex behind-the-scene action of the service provider. We are also searching for a better name than ‘mini service blueprint’, so please comment some ideas below.
Next up, test drive (prototype)
In the weeks to come our team will be exploring these opportunity spaces further, from both the customer and service provider sides. We are all looking forward to designing, building, and testing a conceptual prototype to prove (or disprove) the insights we have so far.
